Barney G. Glaser and Anselm L. Strauss-Awareness of Dying-Aldine (1965) PDF

IStCVAFICNS
A series edited by Howard S. Becker

Northw estern University
AWARENESS
OF
DYING
BY
Barney G. Glaser
AND
Anselm L. Strauss
ih
s ih
A L D IN E PU B LISH IN G C O M PA N Y / Chicago
To
W A LTER A. G LA SER
and
M IN N IE R O TH SC H IL D STRAUSS
Copyright 1965 by Barney G. Glaser and
Anselm L. Strauss
All rights reserved. No part of this publication may be
reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopy, recording,
or any information storage or retrieval system, without
permission in writing from the publisher.
First published 1965 by
Aldine Publishing Company
529 South Wabash Avenue
Chicago, Illinois 60605
Seventh Printing 1974
Library of Congress Catalog Card Number 65-12454
ISBN 0-202-30001-3
Designed by David Miller
Printed in the United States of America
Preface
Once upon a time a patient died and went to heaven,
but was not certain where he was. Puzzled, he asked
a nurse who was standing nearby: Nurse, am I dead?
The answer she gave him was: Have you asked your
doctor?
A n o n y m o u s,
circa 1964
Recently The N ew Y ork Tim es reported: v e r y i l l c h il d r e n

o f d i s e a s e ; Leukem ia Patients at N .I.H . Not Shielded
From Truth. . . . A child should always be told the truth, even
when he has an incurable disease such as leukemia, according
to two researchers who interviewed 51 children hospitalized at
the National Cancer Institute, Bethesda, M aryland, for treat
ment of leukemia. This kind of news item reflects the growing
concern among researchers and public about m atters which
touch on morality as m uch as on the technical aspects of medi
cine. The rapidly increasing proportion of elderly people in the
American population presents a range of personal and social
questions; not the least is how they view their newly won lon
gevity (often including anticipated years of chronic disease) as
well as their attitudes toward death. In consequence, many geri
atric specialists are beginning to study Am erican attitudes
toward death, while others, spurred on by what seems a sense
less prolonging of life within hospital walls by medical tech
nology run wild, are raising questions about death and dying in
American life.
O ur book is no exception to this trend; indeed, we would
further it. We wish to contribute toward making the manage
ment of dying by patients, families and health professionals
to ld
viii
AWARENESS OF DYING
m ore rational and compassionate (and the two are far from
incom patible). The chief differences between our approach and
others can be quickly summarized. Recognizing th at most
Americans are now dying inside hospitals, we have focused
upon what happens when people die there. We have focused
on the interaction between hospital staffs and patients, rather
than on the patients themselves. We have reported on contexts
of action rather than merely on attitudes toward death. And
we have been less concerned with death itself than the process
of dying a process often of considerable duration.
This approach reflects our sociological perspective, for we
have attem pted to channel our reform ing impulses into an in
quiry not at all medical in character. If increasingly Americans
are dying within medical establishments, surrounded m ore by
nurses and physicians than by kinsmen, then how do these
representatives of the wider society manage themselves and their
patients while the latter are dying? How is the hospitals organi
zation capitalized upon in this process? W hat forms of social
action, transitory or m ore permanent, arise while handling the
dying of people? W hat are the social consequences for the hos
pital and its staff, as well as for the patients and their families?
T o answer these kinds of questions, we did intensive field
work (involving a combination of observation and interviewing)
at six hospitals located in the Bay area of San Francisco. We
chose a num ber of medical services at each hospital, selected,
as we shall explain later, to give us m aximum exposure to dif
ferent aspects of dying locales where death was sometimes
speedy, sometimes slow; sometimes expected, sometimes unex
pected! sometimes unanticipated by the patient, sometimes antic
ipated; and so on. The reader who is unacquainted with this
style of field research need only imagine the sociologist moving
rather freely within each medical service, having announced his
intention of studying term inal patients and w hat happens
around them to the personnel. The sociologist trails personnel
around the service, watching them at work, sometimes question
ing them about its details. H e sits at the nursing station. He
Preface
ix
listens to conversations himself. Occasionally he queries the staff

members, either about events he has seen or events someone
has told him about. Sometimes he interviews personnel at con
siderable length, announcing an interview, perhaps even using
a tape recorder. H e sits in on staff meetings. H e follows, day by
day, the progress of certain patients, observing staff interaction
with those patients and conversation about the patients among
the personnel. H e talks with patients, telling them only that he
is studying the hospital. His fieldwork takes place during day,
evening and night, and m ay last from ten minutes to m any
hours.
In presenting what we observed by such methods, we might
have organized our analysis in this book to highlight differences
and similarities among the various medical services. Instead, we
chose to offer our readers a m ore abstract and so m ore power
ful explanatory theoretical scheme. This scheme arose from
scrutiny of the data and should illuminate the data far more
than a comparative analysis of the medical services. O ur an
alysis is based upon w hat we term awareness context, which
is discussed in C hapter I; here we need only note th at this term
refers to who, in the dying situation, knows what about the
probabilities of death for the dying patient. It m akes a great
deal of difference who knows what, and the use of this scheme
allows the organization of many events that otherwise might
seem disconnected o r paradoxical.
The efficiency of the scheme allows us to claim we believe
with some persuasiveness that discernible patterns of inter
action occur predictably, o r at least non-fortuitously, during the
process of hospitalized dying, and th at explicit knowledge of
these patterns would h elp the medical staff in its care o f dying
patients. Physicians and nurses tend to regard such events either
in mythological term s (som e mythologies are touched upon
later) o r to discount patterned events in favor of the uniqueness
of events ( everyone is a differ e n tj i e rsonality, so dies dif
ferently and must be handled differently) . A group o f eminent
physicians hearing of our analysis before its actual publication,
AWARENESS OF DYING
we were told, remarked flatly that sociologists have nothing

useful to offer physicians. Theirs was a natural reaction to the
invasion by outsiders of a delicate and somewhat mysterious
realm. We have not m eant to scientize this realm, nor to offer
commentary that would freeze and prerrtaturely professionalize
care for the dying by hospital staffs. O ur intent was, above all,
to ask whether people can die socially before they die bio
logically, and what this means for hum an relationships. If our
report makes m atters easier for people who m ust live around
the dying (and vice versa), it will only be because critical in
telligence is brought to bear on our findings and on common
practices in American hospitals. Perhaps, then, hospital person
nel will not laugh quite so wryly at the anonymous lines, quoted
above, about the patients puzzled query of a nurse.
Awareness o f Dying was planned as the first of a series of
four monographs resulting from a six-year research financed by
the National Institutes of H ealth (grant num ber N U 0 0 047).
The second m onograph will discuss the course, or trajectory, of
dying; and the third, by Jeanne Quint, will be titled The Nurse
Student and the Dying Patient. A fourth volume will deal with
staff-family interaction in dying situations.
The authors of Awareness of Dying are indebted to a great
many people. They wish especially to thank the third mem ber
of the project team, Miss Jeanne Quint, for her almost daily
invaluable support; also M rs. Elaine M acDonald and Miss R uth
Fleshman, who assisted in data collection during an early phase
of the project. Howard Becker, editor of the Observations
series in which this book appears, read our original manuscript
with an appreciative but unusually critical eye, and we wish to
thank him here. From a num ber of colleagues, we received
general support and specific commentary: among them, H erbert
Blumer, F red Davis, and Louis Schaw. Strauss appreciates his
conversations with Leonard Schatzman. We had a very useful
early exchange with Dr. Melvin Sabshin and equally useful later
conversations with several nurse educators, especially Miss
Helen Nahm, Miss Jeanne Hallburg and Mrs. M ildred M cIntyre.
Preface
Like all field researchers, we are especially indebted to many

persons who worked at the field-work locales. They are far too
many to cite by name, but we wish at least to express our
gratitude to them and their institutions: especially Moffitt Hos
pital (University of California M edical Center, San Francisco);
Providence Hospital, Oakland; the Veterans Administration
Hospital at O akland; the N apa State M ental H ospital; the San
Francisco General Hospital; and Highland Hospital in Oakland.
Miss Karen M any edited and M rs. Kathleen Williams helped
to type first draft manuscript. Miss Bess Sonoda, our project
secretary, is the hitherto without whom unsung heroine of our
manuscript; we thank her, too.
Barney G. Glaser and A nselm L . Strauss
San Francisco
Contents
PART I: INTRODUCTION
Chapter
Chapter
1. The Problem of Awareness of Dying

2. Expectations of D eath: A Problem in
Social Definition
3
16
PART I I : TYPES OF AWARENESS CONTEXTS
Chapter
Chapter
Chapter
Chapter
Chapter
3. Closed Awareness
4. Suspicion Awareness: The Contest for
Control
5. The R itual D ram a of M utual Pretense
6. The Ambiguities of O pen Awareness
7. Discounting Awareness
29
47
64
79
107
PART i n : PROBLEMS OF AWARENESS
Chapter S. Direct Disclosure of Terminality

Chapter 9. The Unaware Family
Chapter 10. The Aware Family
Chapter 11. Nothing M ore to D o : The Problem of
N o Recovery
Chapter 12. Nothing M ore to D o : The Problem of
Comfort
Chapter 13. Awareness and the N urses Composure
119
136
155
177
204
226
PART IV : CONCLUSIONS
Chapter 14.
Chapter 15.
a p p e n d ix :
The Practical Use of Awareness Theory

Awareness and the Study of Social
Interaction
M ethods of Collection and Analysis
of D ata
Index
259
274
286
295
Part One
Introduction
1
The Problem
of Awareness of Dying
American perspectives on death seem strangely paradoxical.
O ur newspapers confront the brutal fact of death directly, from
the front page headlines to the back page funeral announce
ments. Americans seem capable of accepting death as an every
day affair someone is always dying somewhere, frequently
under most unhappy circumstances. T o account for this absorb
ing interest in death (even death by violence), one need not
attribute to the reading public an especially vigorous appetite
for gruesome details. D eath is, after all, one of the character
istic features of hum an existence, and the people of any society
must find the means, to deal with this recurring crisis. Presum
ably one way to deal w ith jt is to talk and read about it.
Curiously, however, Americans generally seem to prefer to
talk about particular deaths rather than about death in the
abstract. D eath as such has been described as a taboo topic
for us, and we engage in very little abstract or philosophical
discussion of death.1 Public discussion is generally limited to
the social consequences of capital punishm ent o r euthanasia.
Americans are characteristically unwiljing to talk openly
bout the process of dying itself; and they are prone to avoid
telling a dying person that he is dying. This is, in part, a moral
attitude: life is preferable to whatever may follow it, and one
should not look forward to death unless he is in great pain.
This m oral attitude appears to be shared by The professional
people who work with or near the patients who die in our
1 Herman Feifel, Death, in Norman L. Farberow, Taboo Topics
(New York: Atherton Press, 1963), pp. 8-21.
PART
o n e : in t r o d u c t io n
hospitals. Although trained to give specialized medical o r nurs

ing care to term inal patients, much of their behavior towards
death resembles the laym ans. T he training that physicians and
nurses receive in schools of medicine or nursing equips them
principally for the technical aspects of dealing with patients.
Medical students learn not to kill patients through error, and
to save patients lives through diagnosis and treatm ent, but their
teachers emphasize very little, or not at all, how to talk with
dying patients, how or whether to disclose an impending
death, or even how to approach the subject with the wives,
children and parents of dying patients.2
Similarly, students at schools of nursing are taught how to
give nursing care to terminal patients, as well as how to give
post-m ortem care, but only recently have the psychological
aspects of nursing care been included in their training. Few
teachers talk about such m atters, and they generally confine
themselves to a lecture or two near the end of the course, some
times calling in a psychiatrist to give a kind of expert testi
mony. 3 Although doctors and nurses in training do have some
experience with dying patients, the emphasis is on the necessary
techniques of medicine o r nursing, not on the fact of dying
itself. As a result, sometimes they do not even know they are
treating or caring for a dying patient.
Although physicians and nurses may thus exhibit consider
able technical skill in handling the bodies of terminal patients,
their behavior to them otherwise is actually outside the prov
ince of professional standards. In hospitals, as in medical and
nursing schools, discussion of the proper ways to manage dying
patients tends to be only in strictly technical medical and nurs
ing terms. Also, staff members are not required to report to
These statements are the result of a secondary analysis of fieldnotes from a study of the University of Kansas Medical School; see
Howard S. Becker, Blanche Geer, Everett C. Hughes, and Anselm L.
Strauss, Boys in White (Chicago: University of Chicago Press, 1961) Jeanne C. Quint and Anselm L. Strauss, Nursing Students, As
signments, and Dying Patients. Nursing Outlook, Vol. 12 (January
1964); also Jeanne C. Quint, The Nurse Student and the Dying Patient
(New*York: The Macmillan Company, in press, 1966).
The Problem of Awareness of Dying
each other, or to their superiors, what they have talked about

with dying patients. As we will discuss later, they are account
able only for the technical aspects of their work with the
dying.4
Medical and nursing personnel commonly recognize that
working with and around dying patients is upsetting and some
times traumatic. Consequently some physicians purposely spe
cialize in branches of medicine that will minimize their chances
of encountering dying patients; m any nurses frankly adm it a
preference for wards or fields of nursing where there is little
confrontation with death. Those who bear the brunt of caring
for terminal patients tend to regard its hazards as inevitable
either one can take working with these patients or he cannot.
Physicians and nurses understandably develop both standard
ized and idiosyncratic modes of coping with the worst hazards.
The most standard mode recognized by physicians and nurses
themselves is a tendency to avoid contact with those patients
who, as yet unaware of impending death, are inclined to ques
tion staff members, with those who have not accepted their
approaching deaths, and with those whose terminality is accom
panied by great pain. As our book will attest, staff m embers
efforts to cope with terminality often have undesirable effects
on both the social and psychological aspects of patient care and
their own comfort. Personnel in contact with terminal patients
are always somewhat disturbed by their own ineptness in han
dling the dying.
The social and psychological problems involved in term i
nality are perhaps most acute when the dying person knows that
he is dying. F or this reason, among others, Am erican physicians
are quite reluctant to disclose impending death to their pa
tients,5 and nurses are expected not to disclose it without the
consent of the responsible physicians. Yet there is a prevailing
belief among them that a patient who really wants to know will
4 Anselm L. Strauss, Barney G. Glaser, and Jeanne C. Quint, The
Nonaccountability of Terminal Care, Hospitals, 38 (January 16, 1964).
pp. 73-87.
* Feifel, op. cit., p. 17.
p a r t o n e : in t r o d u c t io n
somehow discover the truth without being told explicitly. Some

physicians, too, maneuver conversations with patients so that
disclosure is m ade indirectly. In any event, the dem eanor and
actions of a patient who knows or suspects that he is dying
differ from those of a patient who is not aware of his terminality. The problem of awareness is crucial to w hat happens
both to the dying patient and to the people who give him medi
cal and nursing care.
From one point of view the problem of awareness is a
technical one: should the patient be told he is dying and what
is to be done if he knows, does not know, or only suspects?
But the problem is also a m oral one,-involving professional
ethics, social issues, and personal values. Is it really proper,
some people have asked, to deny a dying person the opportunity
to
his peace with his conscience and with his God, to
settle his affairs and provide for the future of his family, and
to control his style of dying, much as he controlled his style of
living? Does anyone, the physician included, have the right to
withhold such information? Someone m ust decide whether to
disclose, and when to disclose, but on whose shoulders should
this responsibility fall the physician, the family, or the patient?
The rationale for m aking such decisions, as well as for designat
ing who will make them , is not based on technical reasoning
alone but also on various ethical, m oral and hum an
considerations.
Both the hum an and the technical aspects of the awareness
problem are becoming increasingly momentous, for at least two
reasons. One is that m ost Americans no longer die at home.
Last year in the U nited States, 53 per cent of all deaths oc
curred in hospitals, and many more in nursing homes.6 The
family may be present m uch of the time while a person is dying,
but he is also surrounded by many strangers, however compas
sionate and technically skilled they may be. Dying away from
home is com pounded by a noticeable and im portant medical
Robert Fulton, Death and Self, Journal of Religion and Health,
3 (July 1964), p. 364.
trend: because medical technology has vastly improved, fewer

people are dying from acute diseases and more from chronic
diseases. M oreover, the usual duration of most chronic diseases
h is increased, so that terminal patients, in the aggregate, take
longer to die than they used to. They may spend this time in
the hospital, or return several times to the hospital while they
ire dying.
Dying away from home, and from a chronic disease, will
become m ore common during the next decades, m aking the
problem of awareness even more salient to everyone con
cerned. Hospitals are scientific establishments and staff mem
bers are expected to make com petent judgments about what is
wrong with patients and to assess their prospects for recovering
o r not recovering. F or this very reason more and m ore ter
minal patients will persist in asking questions, and in expecting
explicit, detailed answers, about the nature of their illness, how
long the hospitalization will last, and why, and in what shape
they will leave for home. Inevitably, more of them will disoover, or be told, the truth. (T he widespread improvement in
educational level will strengthen these trends.) A nd so it is
predictable that the problem of awareness will become more
nd more central to what happens as people pass from life to
death in Am erican hospitals. T hat problem is the principal
ubject of our book.
AW ARENESS A N D T H E STUDY O F IN T E R A C TIO N
This book will be directed to two audiences: those who are
concerned with dying in our hospitals (especially those who
must deal with it), and sociologists (especially those who study
social interaction). O ur book, then, can be read for its prac
tical details as well as for its theoretical aspects. Writing to such
double audience is only slightly hazardous, we believe, be
cause we are convinced th at sociologists cannot say anything
profoundly useful unless it is o f theoretical interest to sociolo
gists themselves.
PART
As noted, our book is based upon extensive field obser

vation and interviewing done at hospitals in the San F ran
cisco m etropolitan area. The research was part of a larger study
of how nursing and medical personnel give care to terminal
patients. In Appendix A, both the larger project and the field
work methods which underlie our conclusions in this book are
described briefly. In general, our project was focused on what
kinds of thing happen around patients as they lie dying in
Am erican hospitals. In this book, we began by narrowing that
focus to the following questions. W hat are the recurrent kinds
of interaction between dying patient and hospital personnel?
W hat kinds of tactics are used by the personnel who deal with
the patient? U nder what conditions of hospital organization do
these kinds of interaction and these tactics occur, and how do
they affect the patient, his family, the staff, and the hospital
itself, all o f whom are involved in the situations surrounding
dying? In finding answers to these questions, we discovered
that most variations could be accounted for by what each party
to the dying situation was aware of about the patient s fate.
Thus we further narrowed our focus to dealing with these
questions as they related to awareness: a powerful explanatory
variable.
These questions are of immense practical relevance to hos
pital personnel, who are confronted daily with the exceedingly
delicate and difficult task of caring for term inal patients. But
outsiders like ourselves, however observant, can contribute
little by merely narrating the difficulties experienced by hos
pital insiders, and even less by offering a few suggestions for
coping better with those difficulties. (W e are chary of prescrip
tions and evaluations, though in C hapter 14 we shall address
ourselves to how our theory of awareness bears upon issues of
practical understanding and control.) A sociologist contributes
most when he reports w hat he has observed in such a fashion
that his account rings true to insiders, but also in such a
fashion fliat they themselves would no t have written it. T h a t
il, most useful sociological accounts are precisely those which

Insiders recognize as sufficiently inside to be true but not so
"Inside that they reveal only w hat is already known. A nd
since insiders, especially those in stressful circumstances, do
not always agree with each other on all m atters, a sociological
narrative must also take this disagreement into account. The
Sociologists obligation is to report honestly but according
to his own lights.
These two factors how the report is written and what it
dds to what is already known by insiders also perm it the
Sociologist to contribute to the factual knowledge of his own
eraft. To the degree that the sociologist treats his data ab
stractly, he also has an opportunity to say something im portant
to his own colleagues. So, in this book, we address ourselves
to the problem of how to conceptualize and account for social
Interaction. This problem is a central sociological issue: vir
tually all the influential theorists in sociology M ax Weber,
Durkheim, Thomas, Mead, Parsons and Goffman, to name
but a few have grappled with the issue. Researchers, too,
lUlt often decide how to characterize and explain the inter
action that takes place within groups, organizations and insti
tutions.
a w a ren ess c o n t e x t s
Our own approach to social interaction and awareness has

been much influenced by our study of the situation of dying
people. When a patient is brought to the hospital irrevocably
Comatose, and diagnosed as dying, then there is no question
that he will ever become aware of that diagnosis. It is pos
sible that nurses and physicians may define this patients
Status differently, but such disagreement is unlikely. On the
Other hand, when a patient arrives sentient, and there is some
question about whether or not he is dying, then of course his
definition may diverge sharply from theirs, and theirs vary
10
PART
among themselves. W hat each interacting person knows of the

patients defined status,7 along with his recognition of the
others awarenesses of his own definition the total picture as
a sociologist might construct it we shall call an awareness
context. It is the context within which these people interact
while taking cognizance of it. Plainly, that context is complex,
and it may change over time, especially as the patients condi
tion worsens and as explicit or implicit messages on his condition
get through to him.
If we assume that a num ber of persons are party to the
interaction, then the logical combinations of who knows what
yield a great many different types of awareness context.8 For
P art II of this book, we have selected several that recur em
pirically-^and are immensely im portant for what goes on
around the patient when only two interactants are assumed.
Those two are the patient and the hospital staff. To simplify
m atters, we assume that the staff acts mainly as though all
its members share the same awareness. F or P art III, we do
not make the assumption of staff consensus, bu t discuss what
happens under various conditions of agreement and disagree
ment. W hen discussing the patients family in P art III, we
shall not assume consensus either.
Each chapter in P art II is organized around specific aware
ness contexts, each turning around the confrontation of patient
and hospital personnel and the characteristic modes of inter
action th at appear within the context.9 F o r instance, there is
TIn Chapter 2 we analyze me elements o f this definition of the dying
status.
Barney G. Glaser and Anselm L. Strauss, Awareness Contexts and
Social Interaction, American Sociological Review, 29 (1964), PP669-79. In this paper we consider the various dimensions that make
up the types considered in this book and that could make up thirty-six
possible types. These dimensions are (1) number of people involved;
(2) acknowledgment of awareness (pretense or not); (3) degree of
awareness (aware, suspicion, unaware); (4) object of awareness (in
our case a death expectation); and (5) awareness of other persons
awareness of object
These interactional modes may occur also in other types of aware
ness context, but in less empirically significant form or magnitude.
The Problem o f Awareness of Dying
JJ
the situation where the patient does not recognize his impending
death even though everyone else does. There is the situation
where the patient suspects what the others know and therefore
attempts to confirm or invalidate his suspicion. There is the
situation where each party defines the patient as dying, but
each pretends that the other has not done so. And, of course,
there is also the situation where personnel and patient both
are aware that he is dying, and where they act on this aware
ness relatively openly. We shall refer to these situations as,
respectively, the following types of awareness context: closed
awareness, suspected awareness, mutual pretense awareness,
and open awareness.
The im pact of each type of awareness context upon the
interplay between patients and personnel is profound, for
people guide their talk and actions according to who knows
what and with what certainty. As talk, action, and the accom
panying cues unfold, certain awareness contexts tend to evolve
Into other contexts. Said another way, interaction evolves or
develops; it does not remain static. O ur task will be to trace
typical patterns of interactional development, and to link them
with awareness contexts, according to the following paradigm
for the study of awareness contexts and interaction: (1 ) a
description of each awareness context, (2 ) the social structural
conditions that enter into the context, (3 ) the consequent
interaction, including various tactics and countertactics, (4 ) the
change of interaction from one type of awareness to another,
(5 ) the ways in which various interactants engineer changes
of awareness context, and (6 ) various consequences of the
Interaction for the interactants, for the hospital, and for
further interaction. Each chapter in P art II is organized ac
cording to the above design. The purpose of these chapters is
to show how our theory discriminates various interactional
modes and processes. Thus, they show that the theory works.
Part III, which then follows, is focused on various general
problems associated with awareness contexts: notably as they
bear upon the patient, his family and the hospital staff. These
12
PART
chapters develop models for understanding and explaining

problems of awareness of dying which will be useful to both
medical practitioners and sociologists. The analysis in each
chapter is carried on according to the prescriptions of the aware
ness context paradigm, but not with the same organization. They
are organized, rather, by the substantive issues of the problem
under consideration. Thus these chapters show our theory
applied to substantive issues. In P art IV we consider some
implications of our general analysis for nursing and medical
care and for awareness and interactional theory.
In so arranging our book, we hope to link together our
dual com m itm e n t to practical and theoretical affairs. Surely
we shall not have said all there is to say indeed not all we
ourselves have to say about dying in American hospitals. But
we hope that our account will help establish surer ground
than now exists for understanding the social-psychological
dim ensions of terminal care, as well as contribute to sociological
theory.
It will prove useful to note a few general features of aware
ness contexts in the dying situation as they contrast with other
situations involving the problem of awareness. O ther features
will be noted in succeeding chapters. First, unlike some secrets
that may be kept or disclosed, a m ans approaching death is
almost certain to be bad news to him when and if he learns
of it (unless he is very old or in tremendous p ain ). Conse
quently, the stakes involved in this particular awareness are
very high. F o r the staff, the stakes are also frequently high,
because of various reasons soon to be detailed. Second, the
Hying situation involves a single individual (the patient) who
is pitted against a group of others (the staff). Each staff
member usually acts as a group representative when interacting
with the patient. This situation can be contrasted with spy
dramas, where spies and counterspies, on both (o r even more
than tw o) sides, all belong to teams. Third, the signs of the
patients terminality are not always visible to him , and, if
visible, may not usually b e very meaningful to him. Their
meaning is not always apparent, either, to all other interactants.

Again this situation contrasts with some other awareness situa
tions where a single sign (say, dark skin) is likely to be
Interpreted specifically by most interactants ( race ). These
different features of different awareness situations suggest the
generality of awareness context as a concept. We shall de
velop this theme of generality in the closing chapter of this
book. Initially, it is enough to say that the hum an condition is
uch that m en often act without full surety of knowing one
anothers true identity. Action proceeds on faith, and on the
basis of canniness and tactics, as well as on m utual under
standing of motives and statuses.
SOCIAL O R D E R AND T H E
ANALYSIS O F IN T ER A C TIO N
Other sociologists have duly taken into account this phe
nomenon of problem atic identity, o r assumed its existence,
but few of them have analyzed awareness of identity and inter
action in detail. Some theorists (Simmel, Cooley, M ead,
Icheiser, Goffman and Davis) have written in detail about
awareness and consequent interaction, but generally there has
been little theoretical cognizance of either the structural con
text within which types of awareness occur, or the structure
of the awareness context itself. As a result, they have tended to
restrict their range in considering the problem of awareness.
Their characteristic theoretical perspectives and concerns may
have led them to focus on one or two types of awareness:
George M eads concern with the implications of m utual under
standing is a good example. O r a writers ideas m ay be directly
affected by the particular kinds of data with which he has
worked.10
Our own approach to the problem of interaction is much
affected by a set of convictions about the maintenance of social
order. Concerning social order, N orton Long, a political scien" See Awareness Contexts and Social Interaction," op. cit., for full
discussion.
14
PART
tist, recently wrote that there is a sociology of John Locke

and a sociology of Thom as Hobbes: the one emphasizes a
realm of stable unquestioned values underlying the social order,
and the other, the highly problematic character of social order.11
George M ead earlier rephrased the Hobbesian view by noting
that social changes occur constantly; he argued that the hum an
task is to bring those changes about in an orderly fashion,
to direct change so that it is not chaotic, or in H obbes language,
anarchistic.12 O ur version, which combines both views of this
overdrawn dichotomy, is that the bases of social order must
be reconstituted continually, m ust be worked at, both ac
cording to established values and with the purpose of establish
ing values to preserve order.13 The maintenance of social order,
whether in society, organizations, groups o r interaction, we
believe, turns about handling relatively unforeseen consequences
in this twofold manner.
O ur assumptions about social order have naturally lead
us to explore the developmental possibilities of interaction,
though our assumptions need not be accepted to legitimate this
aim as of immense sociological importance. In his recent Be
havior in Public Places, Erving Goffman analyzed interaction
in term s of the social regulations that subtly govern it and
m ake it the genuine embodiment of social o rd e r;14 in contrast,
11 The Political Act as an Act of Will, American Journal of Soci
ology, 69 (July 1963), pp. 1-6.
11 The Problem of SocietyHow We Become Selves, Movements
of Thought in the Nineteenth Century (Chicago: University of Chicago
Press, 1936), pp. 360-61. See also Herbert Blumer, Society as Sym
bolic Interaction, in Arnold Rose (ed.), Human Behavior and Social
Processes (Boston: Houghton Mifflin, 1962), pp. 179-192.
u For an analysis of working at social order, see Anselm Strauss,
Leonard Schatzman, Rue Bucher, Danuta Ehrlich, and Melvin Sabshin,
The Hospital and Its Negotiated Order, in Eliot Freidson (ed.), The
Hospital in Modern Society (New York: Free Press of Glencoe, 1964),
pp. 147-169; also by Strauss et al.. Psychiatric Ideologies and Institu
tions (New York: Free Press of Glencoe, 1964), pp. 292-315.
Erving Goffman, Behavior in Public Places (New York: Free
Press of Glencoe, 1963). For our contrasting view, see Anselm Strauss,
Mirrors and Masks (Chicago: Free Press of Glencoe, 1959), especially
the section on Self-appraisals and the Course of Action, and Inter
action, pp. 31-88.
15
we are cognizant of rules but are principally interested in

analyzing interaction in terms of its open-ended and prob
lematic character. R ather than focusing on interactional stability,
we shall be preoccupied in this book with changes that may
occur during the course of interaction. However, relatively stable
or purely repetitive interaction also demands analysis beyond
detailing the governing regulations: that is, one m ust also
consider the tactics sustaining that stasis which constantly is
subjected to change and unforeseen consequences; one must
combine both the rule and the working at basis of social
oMer. Thus, in our account of the interaction that centers
around dying patients, we shall be interested not only in the
social regulations and other structural conditions entering into
the interaction, but also in the tendency for interaction to move
out of regulated social bounds and into new interactional
modes.
Expectations of Death: A Problem in Social Definition
2
Expectations
of Death: A Problem in
Social Definition
The issue discussed in this chapter is basic to all discussions
of interaction: How do the interacting persons come to define
both themselves and others as interactants and how do they
m ake redefinitions, if necessary, as the interaction continues?
As it pertains to the dying patient, the question is how hospital
personnel come to see the patient as due to die within some
approximate time period, and how they define their status and
sometimes their selves in their relations with this patient.
The patient m ust do the same with them, in reverse, even if
he never discovers their true definitions of him.
Some years ago the Swedish social scientist G unnar M yrdal,
in his book about race relations in the U nited States, remarked
upon the importance of criteria, in this instance especially a
cinglp criterion, skin color, for defining statuses. H e writes:
I had once another experience which throws light cm the
same problem. . . . T he N.A .A .C.P. had, in 1939, their
an n u al convention in Richm ond, Virginia. I visited the
mmrings and took part in a boat excursion which ended
die convention. O n board I approached a group of offi
cers and crew (w hites) who held themselves strictly
apart, looking on the Association members who had
crowded their ship for the day with an unmistakable
Hihrtnrf- of superiority, dislike, em barrassm ent, interest
and friendly hum or. . . . I told them that I was a
stranger who by chance had come on the boat, just for
16
17
the excursion. . . . [They] entertained me for more than

an hour by telling me everything about the Negro and
the Negro problem in America. During the course of
our conversation I rem arked that there were apparently
a lot of white people, too, on the boat. A t first they just
laughed at my rem ark and insisted that all persons
present were Negroes. Some Negroes are so fair, they
told me, that only Southerners, who know them by life
long intimate association can distinguish them from
whites. I insisted and pointed to M r. W alter White, the
secretary of the N.A.A .C.P. and some other white
Negroes, and actually succeeded in drawing an acknowl
edgment that he and some other men (who I knew were
all Negroes) were, indeed, white upon closer observa
tion. One of my interlocutors went to have a closer look
at the persons I had pointed out, and came back and
confirmed authoritatively that they were indeed white.
There are some nigger lovers in the N orth and we
have a few down here, too, he com m ented.1
The whites first saw the passengers as Negroes primarily be
cause of skin color, and then misperceived some as whites
for the same reason. This imperious role of skin color in
American race relations was brought home recently when a
white journalist, John Griffin, ingeniously changed his skin
color and passed as Negro without once being challenged by
either southern whites or Negroes.2 Griffins passing seems
even more ironic in light of a rem ark by a young sociologist,
unaware of the book, that a white sociologist could not study
Negro social relations in depth because he could not pass as a
Negro both Negroes and whites would discover who he really
was.
But in most cases a single criterion rarely is the basis
for interactional definition. The complexity of the process of
making definitions in interaction will become clear in our discus
sion of the specifics of defining dying patients.
1Gunnar Myrdal, An American Dilemma (New York: Harper and
Bros., 1944), Vol. 1, p. 68.
'Jo h n Griffin, Black Like Me (Boston: Houghton Mifflin Co., 1960).
PART o n e :
18
in t r o d u c t io n
DIA GNOSIS AND PROGNOSIS AS

D E F IN IT IO N A L QUESTIONS
W hen a patient enters a hospital, one of the m ost im portant
initial questions is whats the diagnosis? This question is
n o less im portant when the patient is fated soon to die, unless

death is so imminent that diagnosis is meaningless. W hat is
done to and for most patients depends mainly on the answer
to the diagnostic (and its allied prognostic) question. M odem
hospitals are organized to insure a relatively speedy answer.
If diagnosis is uncertain, then additional diagnostic soundings
are typically made, and the course of the illness itself, during
the next few days, may prom pt more accurate answers.
From a sociological perspective, the im portant thing about
any diagnosis, whether correctly established or not, is that it
involves questions of definition. Even with a known terminal
patient, another im portant question how fast he will die
m ust still be determined. M any a hospital patient poses for the
staff both questions simultaneously: Is this patient going to
die here, a n d wheat? T he first o f these questions refers to un
certainty of death while the second refers to tim e of death.
E ven m ore specifically, let us say that certainty o f death means
the degree to which the defining person (phyacnan, nurse, or
even the patient him self) is convinced th at the patient will die.
L et us say th at tim e o f death means the expectation of either
( a ) when th e Certain death will occur, o r (b ) when the un
certainty abqpt death will be resolved. Tune units can be from
m inu tes to m onths, varying with the nature of the illness and
the patients location in the hospital. F o r example, on emergency
wards only a few m inutes m ay pass before it is known for
certain whether a patient will live. A bout prem ature babies,
nurses usually think o f death in term s o f hours o r a few days
a t m ost; while for cancer patients, th e tim e unit may bem onths.
In com bination, certainty and time yield four types of
death expectations: ( 1 ) certain death a t a known time,
19
(2 ) certain death at an unknown time, (3 ) uncertain death

b u t,
-tmiiB r a ^ ^ r t K e ' question will be resolved, and
(4 ) uncertain death and unknown time when the question
will be resolved. As we shall show later, these types of ex
pectations have varying effects on the interaction of participants
in the dying situations.
ESTABLISHING EX PEC TA TIO N S O F D EA TH
Anyone may read the medical signs and draw his own
conclusions, the terminal patient included. But in American
hospitals, the attending physician is the only one who can
legitimately define the patients condition, because of his pro
fessional expertise and the professional mandate that he be
medically responsible for the patient. Ordinarily, only he may
tell patients that they are dying. U nder extraordinary conditions,
nurses tell patients or relatives directly, but this is not the
usual practice. (The physicians problems in reading the medical
signs correctly will not concern us here, though it is worth
noting that uncertainty in medical diagnosis and prognosis, and
the need to train young physicians to manage it, has produced
a special literature.3)
Nurses m ust also correctly assess whether the patient is
dying and when he will die. To m ake those assessments is
often no easy matter. In forming their expectations, the nurses
may rely cm their own reading of cues how the patient looks
and acts, what his charts report about him as well as on cues
Talcott Parsons, The Social System (New York: Free Press of Glen
coe, 1951), pp. 447-466; Ren6e Fox, Training for Uncertainty, in
Robert K. Merton, et al. (eds.), The Student Physician (Cambridge:
Harvard University Press, 1957), pp. 207-241; Ren6e Fox, Experiment
Perilous (New York: Free Press of Glencoe, 1959), see index on
page 262 for diverse sources; Fred Davis, Uncertainty in Medical
Prognosis, American Journal of Sociology, 66 (July, 1960), pp. 41-47;
Fred Davis, Definitions of Time and Recovery in Paralytic Polio
Convalescence, American Journal of Sociology, 61 (May 1956), pp.
382-587; Julius A. Roth, Timetables (Indianapolis: Bobbs-Merrill,
1963); and Howard S. Becker, Blanche Geer, E. C. Hughes, and An
selm L. Strauss, Boys in White (Chicago: University of Chicago Press,
20
PART
flashed, perhaps unwittingly or obliquely, by the doctor. Some

times they also receive direct information from the doctor;
typically, they trust this source m ore than their own individual
or collective reading of cues, though of course the reverse
may be true if they are experienced nurses and
the doctor
seems inexperienced, incompetent, or not well acquainted with
the case. Sometimes the cues are so obvious that the physician
needs to say little o r nothing. So, although the most legitimate
source for forming death expectations is the physician, the
nurses also observe cues constantly.
Doctors vary considerably as to whether they give nurses
a legitimate basis for death expectations. ( The doctor may or
may not tell us a patient is critical. We decide. They expect us
to use our beans. ) It is unusual for a nurse to ask the doctor
directly,
but there m ay be an implicit understanding that he
will tell her, or she can hint that he should tell
her or give
her cues. H e may tell h er obliquely at first, but more directly (
later when the patient nears death. In one Catholic hospital
a nurse told us, There is no formal declaration that a patient
is terminal. Sometimes a doctor will tell the nurse, but usually
you just pick it up. She added: If the doctor knows that a
patient is going to die, I prefer to know it. This nurse wants
legitimate expectations, but she is forced to rely mainly on
cues. Sometimes nothing much needs to be said, as when a
cancer patient who has returned often to the service returns
now in obviously critical condition, and everyone, residents
and nurses alike, agree that this is the last time. Sometimes
the cues are as explicit as words might be, as when a feeble
patient who still seems to have a chance to live is taken off
intravenous blood injections.
In some hospitals, the doctor is required to deal with the
uncertainty dimension by putting the patient on a critically,
dangerously or seriously ill list, or by including that information
on an admission card. A patients appearance on such a critical
list initiates m any activities. In one veterans hospital, for
example, families are allowed to stay around the clock with
21
the patient; if relatives are not on hand they are sent a wire
stating that Y our (kin) has been put on the seriously ill list,
please come at once. The physician then has a talk with the
family, while social workers and the chaplain drop in to help.
To be sure, medical personnel at this hospital are reluctant
to initiate all this activity by formally indicating the patients
Impending death, and they wait until the last mom ent to put
patients on the critical list. Paradoxically, the nurses therefore
must work longer with cues unless given explicit information
by the doctor.
Two principal types of cues that can be read by the nurses
are the patients physical condition and the tem poral references
made either by themselves o r the medical staff. Physical cues,
ranging from those that spell hope to those that indicate im
mediate death, generally establish the certainty aspect of death
expectations. Tem poral cues, however, have many reference
points. A m ajor one is the typical progression of the disease,
against which the patients actual m ovement is m easured (he
ll going fast o r lingering ). A nother is the doctors ex
pectation about how long the patient will remain in the
hospital. F o r instance, one patients hospitalization was lasting
longer than the short while anticipated by the physician. Work
schedules also provide a tem poral reference: nurses adjust their
expectations according to whether the patient can continue
being bathed, turned, fed, and given sedation regularly. All
llich references pertain to the tem poral aspect of dying; that
ll, to how long the patient is expected to live.
Because physical cues are generally easier to read, and their
presence helps to establish some degree of certainty about dying,
temporal cues are rather indeterminate when the physical ones
I t absent. The patient m ay die sometime o r at any tim e.
As both types of cues accumulate, they m ay support each other;
for example, as a patients condition becomes m ore grave his
hospitalization grows longer. But physical and tem poral cues
Qan also cancel each other: thus, an unduly long hospitalization
n be balanced or even negated by increasingly hopeful physical
22
PART
cues. W hen cues cancel each other, nurses can use the more
hopeful one (he is going home sooner than expected) to balance
o r deny the less hopeful (he looks b ad ). As physical and
tem poral cues accumulate faster and become more severe, they
become harder to deny, and the expectation of death is gradually
more firmly established.
A bout the patients own expectation of death we shall have
much to say in succeeding chapters. H ere we need note only
the most general sources of his expectations. The m ost im
peccable source is the physician himself, who may choose to
announce directly you have little chance to live or your
disease is fatal. However, as noted in C hapter 1, American
physicians very infrequently m ake such announcements. M uch
m ore frequently they drop gentle, oblique references, relying
on the patients willingness to read those references correctly
but if he chooses to ignore their sombre meaning, then physi
cians assume that the patient really does not wish to confront
the fact of his own death.4
Besides the physicians word, what else can the patient
depend on if he does want to know his fate? If one were to
write a set of directives for such a patient it would go as fol
lows: A part from making your physician tell you, listen care
fully to what the staff says about you. Listen carefully also to
anything the medical or nursing personnel may tell you,
obliquely, about your condition; for sometimes they flash cues,
perhaps knowingly or perhaps unwittingly, about you. And do
not forget that you can sometimes force m ore cues out of the
staff members; by being canny and clever, you can catch them
unawares. M any patients, of course, figure out for themselves
such methods for discovering what the staff knows. In addition,
they can attem pt to read the physical cues in their own bodies.
These are sometimes unambiguous enough to be read correctly
by very aged patients or patients with chronic retrogressive
diseases. Again, as with the nurses themselves, tem poral cues
4 This topic of disclosure to patients by doctors is taken up in detail
in Chapter 8.
23
are less dependable but operate in conjunction with the other

cues just mentioned.9
C H ANG ING EX PEC TA TIO N S
We shall leave, for later pages further references to the
patient s definitions of his condition and turn now to ways in
which expectations of a patients death may change.
Nurses definitions of the patients illness status that is,
their expectations affect their behavior toward him. There
fore, the mom ent when their expectations change is significant.
For example, even when the 'physicians cues imply that the
patient is doomed, as when he stops blood transfusions, the
nurses may still not be absolutely sure that the implied predic
tion is accurate. They m ay not lower their levels of alertness
or reduce their effort to save the patient. They may say of the
patient, as one nurse did, If he comes out of it, well work
on him. H e only has to give us the slightest cue. Since the
doctor has said nothing official, even nurses who believe the
patient is dying can still give him an outside chance and stand
ready to save him. They remain constantly alert to countercues.
Everybody is simply waiting, said one nurse. If the doctor
had indicated that the patient would die within the day, nurses
would have ceased their constant watch for countercues and
reduced their efforts to save him, concentrating instead on giving
comfort to the last, with no undue prolonging of life. Thus,
the nurses changing expectations can have considerable impact
on their care of dying patients.6
* For a book length discussion of how one type of patient may sur
vive in a hospital with the aid of sociological knowledge see Julius A.
Roth, Handbook for Tuberculosis Hospital Patients or How to Survive
Ike Treatment (Mimeo: April, 1963). Much of the discussion in our
book on awareness could also be written as a hand book of prescrip
tions to patients on how to cope with the hospital organization, the
staff and the medical rigors of dying, in order to bring ones life to a
close in a more satisfying way than often allowed by medical treat
ment, staff and hospital organization.
Chapters 11, 12 and 13 examine in detail the effects of changing
expectations on medical care, nursing care, and on 'the nurse,
respectively.
24
PART
The changing expectations of nurses and physicians actually

m ap out the patients changes of status. M any patterns of
status passage, 7 with typical rates of movement, are well
known. A classical pattern is the lingering patient: he is certain
to die, but when he will do so is unknown and he does not
die for some while. On one cancer ward we studied, an alltoo-typical sequence of expectations for the lingering patient
ran the gamut of various stages of determ inancy: from original
prognosis of certain death but uncertain time, through the weeks
when the patient began obviously to decline, to the time that
his precise time of death finally became relatively certain. O f
course, before the final decline takes place, such patients may
alternate between hospital visits and periods at home. Often
the nurses feel that a lingering patient is taking more time
than is proper, because there is really no hope for him. (In
this sense, even an unknown time period has limits.)
A patient expected to die on time, bu t who suddenly begins
to recover slightly o r to linger the short-term reprieve pattern
can cause problems for nurses, family, physicians, and hos
pital administrators. H ere is an example: one patient who was
expected to die within four hours had no money, but needed
a special m achine during his last days. A private hospital, at
which he had been a frequent paying patient for thirty years,
agreed to receive him as a charity patient. H e did not die
im m ediately but started to linger indefinitely, even to the point
where there was some hope that he might live. The money
problem, however, created much concern among both his family
members and the hospital administrators. Paradoxically, the
doctor had continually to reassure both parties that this patient,
who actually lasted six weeks, would soon die; that is, to try
to change their expectations back to certain to die on time.
A nother pattern, which may be called the vacillating pat
tern, is a variation on the short-term reprieve. The patient
alternates from certain to die on time to lingering. The
Barney G. Glaser and Anselm L. Strauss, Temporal Aspects of
Dying as a Non-scheduled Status Passage, American Journal of
Sociology (June, 1965).
25
alternation may occur sufficiently often to cause stress among

family members and hospital personnel. Whenever the patient
genuinely starts to fade, the nurses may call the family, for
as one nurse said, If you do not call the family and the
patients die, thats wrong. The family members arrive for
their last look at the dying man, but he begins to linger. They
finally leave saying, please call us again when he begins to
die. Family and nurses may go repeatedly through this stressful
cycle.8 The physician and chaplain may also be affected. Changes
in the activities and moods of the various participants are thus
linked with this vacillating pattern, and the nurses typical mood
is relief when at last they can forecast the end of the unusual
lingering.
In contrast, some patients are the focus of what is com
monly known in the hospital as heroics ; that is, speedy and
all-out attem pts to stave off death. Patients on emergency, pre
mature baby, and intensive-care units sometimes are the recip
ients of such heroic measures. (So are patients who are deemed
especially valuable: extreme cases being Dr. W ard of the
recent British political scandal and a famous Soviet scientist
who has been saved repeatedly through heroic hospital meas
ures.) Heroic measures signify uncertain prognosis, which will
be quickly resolved, with an appropriate change in staff ex
pectations.
The two extremes tow ard which patients may progress are
either getting well or being certain to die at a specific time,
but the nurses and physicians may have other intermediate
expectations for a given patient who leaves the hospital. He
may have arrived with uncertain prognosis, but been sent home
diagnosed as cancerous: certain to die, but when is quite
unknown. The prognoses of more puzzling cases m ay be un
certain on both counts.
The staff may be surprised by unexpected changes in the
expected passage of patients toward death. Am ong the most
Chapters 9 and 10 focus on the changing expectations of the
family.
26
PART
surprising changes are the sudden death, or onset of death,

of a patient who previously had prognoses of doubtful certainty
and time. Of course, the surprise is greatest when there has
been no death expectation whatever, as when seemingly healthy
patients die. W hen such patients die on the operating table, the
resulting im pact is tremendous. The patients death is too sud
den: there is no time to prepare oneself for it. (In one instance,
a surgeon much admired by his nursing stall shocked them with
an unexpected loss on the operating table. R um ors of negligence
were rife, until autopsy showed that the m an had died of
unanticipated natural causes.) In less traum atic instances, when
expectations are revised as the patient moves toward death, but
his progress turns out to be unusual, personnel may experience
a disquieting feeling of having missed certain steps.
T he dying process can become exceedingly complex, for
doctors, nurses, family and patient can have different death
expectations based on different sources, or they may interpret
the same cues differently. The total combinations of these
differential expectations make up the what we referred to in
the first chapter, in our definition of the awareness context
around the patient.9
W e wish to underscore the theoretical step we have taken from
two important papers by Fred Davis (see Note 3), each of which
brought out the notion of differential perceptions. In his medical prog
nosis article, Davis discussed the differential perceptions of certainty of
prognosis held by doctor, patient, and family, and in his Definitions of
Time . . . he discussed differential perceptions of time of recovery.
In our study each participant defines the dying patient situation in
terms of both certainty and time. This yields more complex, but more
determinant, sets of differential perceptions than Davis deals with ex
plicitly, and these, in turn, have led us to the concept of awareness
context.
Part Two
Types of
Awareness
Contexts
3
Closed
Awareness
In Am erican hospitals, frequently the patient does not recognize
his impending death even though the hospital personnel have
the information. This situation can be described as ^ closed
awareness context. Providing the physician decides to keep
the patient from realizing, or even seriously suspecting, what
his true status is, the problem is to maintain the context as a
closed one. With a genuinely comatose patient, the staff mem
bers naturally need not guard against disclosure of his terminal
condition. As an interactant, the comatose person is what
Goffman has called a non-person. 1 Two nurses caring for
him can speak in his presence without fear that, overhearing
them, he will suspect o r understand what they are saying
about him. Neither they nor the physicians need to engage in
tactics to protect him from any dread knowledge. A nd of
course with term inal babies, no precautions against disclosure
are needed either. But with conscious patients, care m ust be
taken not to disclose the staffs secret.
CO N TR IB U TIN G STRU CTU RA L CONDITIONS
There are at least five im portant structural conditions which
contribute to the existence and maintenance of the closed
awareness context.2 First, most patients are not especially
1 Erving Goffman, The Presentation of Self in Everyday Life (Edin
burgh, Scotland: University of Edinburgh, 1956), pp. 95-96.
* Cf. Barney Glaser and Anselm Strauss, Awareness Contexts and
Social Interaction, American Sociological Review, 29 (October, 1964),
pp. 669-679.
29
30
PART
t w o : t y p e s o f a w areness c o n tex ts
experienced at recognizing the signs of impending death. Of

course, a patient who has been in an auto accident and is
injured terribly, although still conscious, may recognize how
close death is. H e may also recognize it if he is himself a
physician or a nurse, or if as a chronic hospitalized patient
he has encountered fatal signs in dying comrades.3 B ut most
Americans have not had such opportunities to witness rehearsals
for their own deaths.
A second structural condition is that A m encan physicians
ordinarily do not tell patients outright that death is probable or
inevitable. As a num ber of studies have-shown, physicians find
medical justifications for not disclosing dying status to their
patients.4 F o r instance, D r. D onald Oken has recently demon
strated th at a chief reason offered is clinical experience. 5
The physician states from clinical experience, that when one
announces terminality to a patient, he is likely to go to pieces ;
one m ust therefore carefully judge whether o r not to tell after
sizing up the individual patient. In actual fact, Oken notes,
the clinical experience is not genuinely grounded experience,
but a species of personal mythology. Generally the experience
consists of one or two unfortunate incidents or even incidents
recounted by colleagues and this effectively cuts off further
clinical experimentation to discover the possible range of conse
quences of disclosure.
Am ong several articles of faith in professional ideology
which support not disclosing terminality, is the belief that
patients really do not wish to know whether they are dying.
If they did, then they would find out anyhow, so there is no
* Ren6e Fox, Experiment Perilous (New York: Free Press of
Glencoe, 1959).
Herman Feifel, Death in Norman L. Farberow, Taboo Topics
(New York: Atherton Press, 1963), p. 17.
* Donald Oken, What to Tell Cancer Patients: A Study of
Medical Attitudes, Journal of the American Medical Association,
175 (April 1, 1961), pp. 1120-28. See the bibliography cited m his
paper for further studies on physicians tendency not to announce ter
minality. See also our further discussion in Chapter 8 of physicians
announcements.
Closed Awareness
31
sense telling them directly. Presumably some, patients do not

wish to know their fates, but there is no really good evidence
that all wjsITto reinain in blissful ignorance or that these pa
tients will find out while in the hospital. And there is some
good evidence that they do wish to know.6 Quite possibly also,
physicians, like other Americans, shy away from the em bar
rassment and brutality of making direct reference to another
person about his impending death. They also undoubtedly would
rather avoid the scene that an announcem ent of impending
death is likely to precipitate.
A third stm ctural condition is that families also tend J o
guard the secret. Family members sometimes may reveal it, but
in our own study we never witnessed deliberate disclosure by
a family member. One psychiatrist told us, however, that he
had disagreed with his fathers physician about the wisdom of
not telling his father, and so had made the disclosure himself.
(Family members, of course, usually confirm what the physician
has already announced.) A n interesting contrast is the usual
practice in Asian countries, where the extended kin gather
around the hospital death bed, two o r more days before death
is expected, openly indicating to the patient that they are there
to keep him com pany during his passage out of life.
A fourth structural condition is related both to the organ
ization of hospitals and to die commitments of personnel who
work within them. O ur hospitals are admirably arranged, both
by accident and by design, to hide medical information from
patients. Records are kept out of reach. Staff is skilled at with
holding information. M edical talk about patients generally
occurs in far-removed places, and if it occurs nearby it is
couched in medical jargon. Staff members are trained to discuss
with patients only the surface aspects of their illnesses, and,
as we shall see, they are accustomed to acting collusively around
patients so as not to disclose medical secrets.
A fifth structural condition, perhaps som ewhat less apparent,
Eighty-two percent of Feifels sample of sixty patients wanted to
be informed about their condition.
32
PART
t w o : t y p e s o f a w areness co n tex ts
is that ordinarily, the patient has no allies who reveal o r help

him discover the staffs knowledge of his im pe^di^gjjeath; Not
only his family but other patients withhold that information,
if they know. W hen a patient wants less distressing information,
he may readily find allies among the other patients; but when
he lies dying, the other patients follow ordinary rules of tact,
keeping their knowledge to themselves o r at least away from
the doomed person. There may be exceptional incidents: the
parents o f a dying teenager told neighbors about his condition,
and these neighbors told their son who in turn spilled out this
inform ation to the patient himself. Among adults, such failures
of secrecy ordinarily do not occur.
While together these structural conditions contribute to
the occurrence and maintenance of the closed awareness con
text, a change in any one condition may precipitate a change
to another type of context. F o r example, the physician may
decide to tell the patient which leads to an open awareness
context; or a patient begins to suspect something amiss because
he begins to learn some of the indicators relevant to increas
ingly grave illness which leads to a suspicion awareness
context. In a later section of this chapter we shall discuss how
changing structural conditions may transform the closed aware
ness context.
T H E PA T IE N T S F IC T IO N A L B IO G R A PH Y
A newly hospitalized patient hassom e -basic, questions
about his condition. H e wants to know, if h e does not know
already, just how ill he is, and whether he is going to get
better or worse, how much better or worse* and how quickly
he will progress or retrogress. In short, he ordinarily is pas
sionately concerned with his' own sick statusr-rlie wishes to
know what m ay usefully be referred to as his future biog
raphy.
The patient who is dying but who has not yet discovered
o r been told of his terminality faces a peculiar problem in get
Closed Awareness
33
ting an accurate assessment of his condition from the medical

and nursing personnel. Can he trust the doctor, can he trust
the nurses and aides, to yield him a true account of his sto ry ?7
Or will their account be tinged by deceit and perhaps thor
oughly false? A false account is a real possibility, although
the patient may not recognize it as such. So trust is im portant
in his transactions with the staff members.
Developing trust is also basic in the staffs transactions
with the patient. To keep the patient unaware of his terminality,
the staff members must construct a fictional future biography,
for him, and they must sustain his belief in that biography by
getting and keeping his trust.
W hat is involved in winning the patients trust? First,
consider a child who is old enough to know about death but
still too young to doubt that his elders will properly care for
him during his illness. The essence of his relations with adults
is that he takes their actions at face value. W hat they are doing
to him, and what they m ake him do is, as they explain, to make
him get better. With children of this age, hospital personnel
need not be so concerned about betraying their own behind-thescenes knowledge. Such suspicions as the child may have will
be more I dont trust them to help m e rather than I dont
trust them because they are not telling me the tru th about my
condition. Consequently, personnel on pediatric wards report
no great danger of unwarily letting their younger patients know
they are dying. Indeed, they report th at less control of ones
face in a childs presence is needed, as he is not so apt to draw
conclusions from an expression.
With adults the m atter of trust in physicians and nurses is
7 Research in hospitals where psychotherapy is practiced indicates
the patients collectively develop a philosophy that one must have faith
In ones doctor. C/. William Caudill et at., Social Structure and the
Interaction Process on a Psychiatric Ward, American Journal of Psy
chiatry, 22 (1952), pp. 314-334; also Anselm Strauss et al.. Psychiatric
Ideologies and Institutions (New York: Free Press of Glencoe, 1964),
pp. 271-273. But compare with the frequent distrust of doctors found
In TB hospitals: Julius Roth, Timetables (Indianapolis: Bobbs-Merrill,
1963).
34
PART TWO: TYPES OF AWARENESS CONTEXTS
rather more complicated. Trust does not arise automatically;

it m ust either be part of the history of a particular relationship
or it m ust be earned. Furtherm ore, once earned it m ust be
m aintained. The importance of keeping trust can be seen in
the dilemma of an unaware patient who has long placed trust
in a particular physician, but whose suspicions have now been
aroused by some incident or remark. Can the m an still be
trusted in this new domain, trusted not only to pull me
through my illness but to tell me the worst? In general,
the physician and others m ust not arouse any suspicion by
their words and actions that they are concealing knowledge
about terminality. They must not even seem to fail in giving
honest answers to any questions that might touch upon termi
nality. W hen staff members succeed in acting convincingly,
the patient accepts their account of his future biography as
accurate, or at least as accurate as they know how to make it.
ASSESSMENT M A N A G EM EN T
To sustain the unaware patients belief in their version of
his future biography, the staff members m ust control his assess
ments of those events and cues which might lead him to suspect
or gain knowledge of his terminality. Their attem pts at managing
his assessments involve them in a silent game played to and
around him ; during which they project themselves to him as
people who are trying to help him get better, or at least to
keep him from getting worse. They must be sufficiently com
m itted to this game, and sufficiently skilled at it, not to give
it away. Their advantage is that they can collaborate as a
team, sometimes a very experienced one, against an opponent
who, as noted earlier, has ordinarily not m uch experience in
discovering o r correctly interpreting signs of impending death,
and who is usually without allies.
Though the staffs explanations of his condition initially
may seem convincing to the unaware patient, and though he
may greatly trust the staff, he may begin to see and hear things
Closed Awareness
35
that arouse his suspicions about his condition. Inevitably he

wonders about, and requires explanation for, a host of events:
Why is he given certain treatments o r exposed to certain tests
and procedures? Why is he moved from one room to another?
Why has one more physician dropped into his room to examine
him? Why is he not getting better m ore quickly? H e wants
answers to questions stimulated by various events even when
he does not in the slightest suspect their true significance.
The staff members usually hasten to give him reasonable
interpretations of those events. Frequently they even offer
interpretations before he requests them, because they calculate
he may otherwise secretly suspect the real meanings of those
events. Their interpretations may be intentionally incorrect or
only partly correct if they feel that a correct one will po
tentially or actually reveal terminality to the patient. Thus,
many of their interpretations are m eant to mislead him. Of
course a correct interpretation is offered when the staff member
judges it will neither disclose n o r arouse suspicions of termi
nality.
A patient near death, for instance, may be moved to a
special space or room, but is informed that this move is only
to permit more intensive care. A patient may even be sent home
to die on request of his family, but his discharge from the hos
pital is explained in quite other terms. Occasionally the physician
will give a patient an incorrect or incomplete diagnosis to
explain his symptoms as in one instance when a patient was
discovered, through a diagnostic operation, to have incurable
cancer of the pancreas but was sent home with a diagnosis of
diabetes and given instructions in the use of insulin. If the
patient believes that he has taken a tu rn for the worse, or at
least is not improving, staff members are ready with reasonable
answers that are m eant to mislead him. A t the very least, certain
of his symptoms m ust be explained away, discounted, as sig
nifying something much less alarming than they appear to him.
Sometimes the physician warns him of anticipated symptoms,
discounting them in advance. The physician and the nurses
36
PART
may also go out of their way continually to reassure him that

things will turn out all right, and indeed are coming along
fine. They will fabricate, hint at, or suggest favorable prog
ress. Nurses especially compare the patients condition, and
his progress, with that of other patients they have known (in
cluding themselves), thus suggesting reasons for optimism. If
the patient asks the nurses Am I going to die? as occasionally
patients do without even real suspicion, they refer him to his
physician for an answer ( Have you asked your physician? ),
or change the subject without answering the question, or turn
it aside with a stock answer (for instance, We all have to
go sometime ). W hen a patient asks the same question of the
physician, the latter may simply lie.
The staff members also use tactics intended to encourage
the patient to m ake his own interpretations inaccurately opti
mistic. They will comment favorably on his daily appearance,
hoping he will interpret their comments optimistically. Some
comments are downright misrepresentations and others are
ambiguous enough to be misread easily. Staff members will
attem pt to establish a m ood consonant less with terminality
than with things are not so bad even if not yet better. They
also practice a sleight of hand, like magicians, drawing the
patients attention away from a dangerous cue by focusing
his attention on an innocent one. They raise false scents, some
times displaying elaborate interest in the symptoms he himself
brings up for consideration. The physician may even put on
diagnostic dramas for him, sending him for irrelevant tests. A
certain am ount of conversation with him about his imminent,
or eventual, return hom e may occur; and his own reading of
physical and tem poral cues that represent progress to him
may be supported. There is also what one might call a socia
bility shield, o r conversation that circumvents disclosure;
during this conversation, personnel carry on as if the entire
situation were quite normal. With such interactional devices,
the staff simultaneously projects their own trustworthy medical
Closed Awareness
37
identities, conveying to the patient that he is not expected to

die.
In addition to all these devices, other techniques reduce
cues that might arouse the patients suspicion. Space is care
fully managed, so that talk about him occurs away from his
presence. If a nurse believes h er involvement with, or sadness
about, the patient might give the secret away, she may move
quickly outside his visual range. She may even request an
assignment away from him. Possibly revealing cues are also
reduced by decreasing the time spent with the patient. Personnel
who fear that they may unwittingly disclose something m ay
remain with the patient very little, or choose to work on his
body rather than talk much with him. They may keep tabs on
his physical condition by popping in and out of his room, but
thereby keep conversation at a minimum. Sometimes, when
the patient is extremely close to death but there is nothing
much to be done for him, staff members tend to go no farther
than the doorway. If the patient becomes genuinely comatose,
nurses or aides can again circulate freely in the patients room.
The disclosing cues are minimized m ost subtly by reducing
the range of expression and topic. The face can be m a n a g e
so as to minimize the dangerous cues it conveys; hence the
conventionally bland or cheerful faces of nurse and physician.
A less obvious mode of minimizing such cues is to censor and
select conversational topics. Staff members, for instance, steer
conversation away from potentially revealing subjects and
toward safer ones. They may talk, especially, about the rela
tively straightforward meanings of procedures being used on
the patients body, providing those meanings are not revealing
One aspect of talk that personnel are cautious about is
its present-future orientation. W hen a patient is defined as
certain to die within a few days, nurses tend to focus their
conversation with, him on. the immediate . present, discussing
such m atters as current doses of medication for pain relief
or other topics relevant to his comfort. But if they do not antici
38
PART TWO: TYPES OF AWARENESS CONTEXTS
pate death for some time, then they will extend the temporal
range implied in their talk. One nurse thus said, before leaving
for a weekend, See you next week. A nother told a patient
that he would need another X ray in two weeks. Similarly, they
talk about blood tests to be done next week, or about the
familys impending visit. One young nurse told us that she
used to chat with a young patient about his future dates and
parties, but that after discovering his certain and imminent
death, she unwittingly cut out all references to the distant
future. It is not easy to carry on a future-oriented conversation
without revealing ones knowledge that the conversation is, in
some sense, fraudulent, especially if the speaker is relatively
inexperienced.
Staff members, again especially if they are inexperienced,
m ust also guard against displaying those of their private
reactions to him and to his impending death as might rouse
the patients suspicions of his terminality. F or instance, young
nurses are sometimes affected by terminal patients of their own
age whose deaths become standing reminders of their own
potential death. ( I found . . . that the patients who concerned
me most when they died were women of my own age. . . . )
Identification of this kind is quite common, and makes more
difficult the staff members control of their behavioral cues.
Their reactions to the patients social loss 8 can also be
revealing. In our society, certain values are highly esteemed
among them youth, beauty, integrity, talent, and parental and
marital responsibility and when a terminal patient strikingly
embodies such values, staff members tend to react to the
potential or actual loss to his family o r to society. B ut such
reactions m ust not be allowed to intrude into the fictionalized
future biography that is directly and indirectly proffered to
the patient. Since personnel tend to share a common attribution
of social value, that intrusion is quite possible unless they keep
Barney Glaser and Anselm Strauss, The Social Loss of Dying
Patients, American Journal of Nursing (June 1964), pp.
Closed Awareness
tight control over their reactions. Although the staffs control

is not always fully conscious, it is no less real.
Finally, because the patient is usually in contact with many
staff members during each day, his suspicions of terminality
may be aroused if they do not give consistent answers to his
questions or if their behavior yields him inconsistent cues. So
they m ust strive for consistency. But if he queries any incon
sistency, then the staff members m ust offer him reasonable
interpretations to explain away the apparent inconsistency.
FROM CLOSED TO O T H E R AW ARENESS CONTEXTS
Inherently, this closed awareness context tends tow ard
Instability, as the patient moves either to suspicion or full
wareness of his terminality. The principal reasons for the
Instability of closed awareness require only brief notation, as
they have already been adum brated. First, any breakdown in the
structural conditions that make for the closed awareness con
text may lead to its disappearance.9 Those conditions include
the physicians decision not to tell the patient, the familys
supporting agreement, and the tactful silence of other patients.
For example, such a break occurs repeatedly at one Veterans
Administration hospital we studied, where patients who are
eventually judged to be hopelessly terminal are sometimes asked
whether they would care to become research subjects and re
ceive medication that might prolong their lives.
Some unanticipated disclosures or tip-offs, stemming from
Organizational conditions, can also occur. The staff members
Who work in emergency rooms in hospitals need to be espe
cially careful in controlling their facial surprises or immediately
*
Catholic practice of giving the sacrament requires calling in a
priest prior to possible death, and this may function as an announcejnent to the dying patient. But non-Catholics probably need to be told
lhat this is done only to insure that the patient does not die without
IlM sacrament. If he recovers, whether to live for many years or merely
ays, then he may receive the sacrament once or perhaps several times
Igaln.
40
PART
compassionate distress when suddenly confronted by obviously

dying patients, often the battered victims of accidents caused
by others carelessness. Staff members control in these in
stances is not always successful. O n other hospital services,
rotation of nursing staff may bring the patient into contact
with new persons who do not always know of his terminality
or who have not been coached properly in die history of how
the staff has conducted itself in die presence of this p atien t
Even the daily shift carries hidden dangers. F or instance, the
doctor may inform the day .shift that his patient is dying, but
unless specific measures are taken to transmit this information
to the night shift, the patient may be cared for by personnel
not sufficiently alert to the dangers of unwitting disclosure. This
danger is especially grave when the patient is worried about
certain new symptoms.
New symptoms understandably are likely to perplex and
alarm the patient; and the longer his retrogressive course, the
more difficult it becomes to give him plausible explanations,
though a very complicated misrepresentational dram a can be
played for his benefit. Even so, it becomes somewhat more
difficult to retain the patients trust over a long time. W hen
a patients physical symptoms become compelling, sometimes
he can force nurses into almost an open admission of the truth.
Not to hint at, o r tacitiy adm it to, the truth would be to risk
losing his trust altogether. So it is not unknown for the staff
to snow a patient under with drugs as he nears death, partly
to reduce his suffering (and perhaps their ow n) and partly to
reduce the likelihood that finally he will correctly read the
fateful signs. Indeed, sometimes a race with time occurs, there
being some question whether the patient will die, or at least
become comatose, before he becomes -suspicious or aware of
his terminality.
A nother threat to closed awareness, closely linked with
retrogressive physical symptoms, is that some treatm ents make
little sense to a patient who does not recognize that he is
dying. Just as many polio patients will not take full advantage
Closed Awareness
of rehabilitation programs because they do not anticipate being

cured, so a terminal patient may refuse a medicine, a machine,
an awkward position o r an inconvenient diet.10 To accomplish
her task, the nurse may have to hint at the extreme seriousness
of her patients condition, hoping that he will get the point
just long enough for her to complete the procedure. But if this
tactic works, it may then also stimulate an attem pt by the
patient to discover what is really happening to him. The staff
then must redouble its assessment management. F o r instance,
the nurse who alarmed him m ust immediately assure him that
he has interpreted her words incorrectly but she m ay not be
wholly or at all successful. The physician can also have prob
lems when explaining to his patient why he must undergo
certain additional treatments, tests and operations, or why he
needs to return to the hospital for a while.
A t times, moreover, a patient may be unable to cope with
his immensely deteriorating physical condition unless nurses
interpret that condition and its symptoms to him. T o do this,
nurses may feel forced to talk of his dying. Not to disclose
at this desperate point can torture and isolate the patient, which
runs counter to a central value of nursing care, namely to
make the patient as comfortable as possible. Similarly, the
physicians inaccessibility may force nurses to disclose the
truth in order to do something immediately for, or with the
patient.
The danger that staff members will give the show away by
relaxing their usual tight control also increases as the patient
nears death, especially when the dying takes place slowly. F or
instance, concerned personnel will continually pause at the
patients door, popping their heads into his room to see how he
Is or if he is still alive. Sometimes a patient very close to death
will be given more privileges because of simple compassion
for him (like permitting him to eat previously denied favorite
foods or to take an automobile ride or even to return home
10 Cf. Fred Davis, Uncertainty in Medical Prognosis, American
Journal of Sociology, 66 (July 1960), p. 45.
42
PART
t w o : t y p e s o f aw areness c o n tex ts
perm anently); occasionally this can cue him into suspicion

awareness. The staff will also sometimes relax their guard
when they believe the patient is too far gone (physically dazed,
comatose, or senile) to understand what would otherwise be
revealing cues o r downright disclosing conversations but he
may be sentient enough to understand.
This last set of conditions brings us to the question of
whether, and how, personnel actually may engineer a change
of the closed awareness context. F o r instance, a physician may
drop hints to the patient that he is dying, giving the information
gradually, and hoping that eventually the patient will compre
hend the finality of his medical condition. Nurses have so often
confided to us that they hoped certain patients were really
aware of impending death that we can be sure they have
signaled, however ambiguously, to such patients, especially when
the latter are well beloved and the nurses can see no good
reason why they should not be prepared for death even though
the physicians choose not to tell. The true situation can be
deliberately conveyed to a patient by facial expression, by
carefully oblique phrasing of words, or merely by failure to
reassure him about his symptoms and prognosis. And of course
the family members may occasionally signal or hint at the
dreadful truth. Indeed, when the family actually knows the
truth, the hazards to m aintaining closed awareness probably
are much increased, if only because kin are m ore strongly
tem pted to signal the truth.
CONSEQUENCES O F CLOSED AW ARENESS
So many discrete consequences seem to flow from this
closed awareness context that we m ust restrict our attention
to a few of its consequences for the patient, his family, the
hospital personnel, and the hospital itself. We shall return to
this discussion of interactional consequences in later chapters;
here we shall merely initiate it.
We can approach the consequences of closed awareness by
Closed Awareness
43
touching on a contrasting situation. In her book Experim ent

Perilous, Renee Fox has described a small research hospital
whose patients recognized their own inevitable terminality be
cause that was why they were research patients.11 Death was
an open and everyday occurrence. Patients could talk fam iliarly
to each other, about their respective fatal conditions, as well as
to the staff members. Various consequences seemed to flow
from this generally open situation, including the following:
Patients could give each other support; and the staff could
support patients. Patients could even raise the flagging spirits
of the staff! From their deathbeds, they could thank the physi
cians for their unstinting efforts and wish them luck in solving
their research problems in time to save other patients. They
could close their lives with proper rituals, such as letter writing
and praying. They could review their lives and plan realistically
for their families futures. All these various possibilities, avail
able to aware patients in the open awareness situation, are of
course not available to unaware patients in the closed awareness
situation.
When patients are kept unaware of their terminality, then
other consequences emerge. Since the unaware patient believes
he will recover, he acts on that supposition. H e may often be
extremely cooperative with physicians and nurses because he
believes faithful following of their orders will return him to
good health. H e talks and thinks as if his period of illness is
only an interruption of norm al life. Thus he may convert his
sick room into a tem porary workplace, writing his unfinished
book, telephoning his business partners, and in other ways
carrying on his work somewhat as usual. H e carries on his
family life and friendship relations, also, with only the inter
ruption necessitated by tem porary illness. And of course, he
can plan, if plans are called for, as if life stretched away before
him. Since, in reality, it does not, other consequences may
attend his acting as though it did. H e may work less feverishly
on his unfinished book than if he knew time was short, and
11 Ren6e Fox, op. cit.
44
PART
t w o : ty pes o f aw areness co n tex ts
so fail to finish it. He may set plans into operation that make
little sense because he will soon be dead, and the plans will
have to be undone after his death. Also, the unaware patient
may unwittingly shorten his life because he does not realize that
special care is necessary to extend it. Thus he may not under
stand the necessity for certain treatm ents and refuse them.
Unaware cardiac patients, as another instance, may even destroy
themselves by insisting upon undue activity.
A word or two should be said about the next of kin. It is
commonly recognized that it is in some ways easier for the
family to face a patient who does not know of his terminality,
especially if he is the kind of person who is likely to die
gracelessly. And if an unaware person is suddenly stricken
and dies, sometimes his family is grateful that he died without
knowing. On the other hand, when the kin m ust participate
in the non-disclosure drama, especially if it lasts very long, the
experience can be very painful. W hat is more, family members
suffer sometimes because they cannot express their grief openly
to the dying person; this is especially true of husbands and
wives who are accustomed to sharing their private lives. Other
consequences for the family of a patient who dies without
awareness are poignantly suggested in the following anecdote:
The dying m ans wife had been informed by the doctor, and
had shared this information with friends, whose daughter told
the patients young son. The son developed a strong distrust
for the doctor, and felt in a way disinherited by his father
since they had not discussed the responsibilities that would
fall to him in the future (nor could they). The father, of course,
could do nothing to ameliorate this situation because he did
not know that he was going to die; and so, this closed aware
ness situation was, perhaps unnecessarily, made more painful
and difficult for the family members.
We have already indicated in detail what difficulties the
closed awareness context creates for the hospital staff, especially
for the nurses. Nurses may sometimes actually be relieved
when the patient talks openly about his demise and they no
Closed Awareness
45
longer have to guard against disclosure. By contrast, the closed

context instituted by the physician permits him to avoid the
potentially distressing scene that may follow an announcement
to his patient, but subjects nurses to strain, for they must spend
the most time with the unaware patient, guarding constantly
against disclosure. The attending physician visits the patient
briefly and intermittently, sometimes rarely.
On the other hand, under certain conditions, nurses prefer
the closed context. Some do not care to talk about death with
a patient, especially a patient who does not accept it with
fortitude. An unaware person is sometimes easier to handle
anyway, for precisely the reason that he has not given up,
or is not taking death badly or hard. (A t one county
hospital where we observed both nurses and interns, it was
remarkable how little strain they experienced in their frequent
contact with extremely ill, but unaware, terminal patients.)
Nonetheless, as we shall show later in more detail, the closed
awareness situation prevents staff members from enjoying
certain advantages that accompany a patients resigned or
joyousmeeting with death.
As for the hospital itself, the im portant consequences of
closed awareness derive mainly from the consequences for the
staff. Unaware patients who die quickly tend to do so without
fuss, so the hospitals routine work is delayed less. On the other
hand, as one can readily imagine, a patient who moves ex
plosively and resentfully from an unaware to a highly suspicious
or fully aware state is relatively disruptive. B ut these are only
transitory consequences; the long-run consequences are most
important.
The most crucial institutional consequence has already been
mentioned: because American physicians choose not to tell
most patients about terminality, the burden of dealing with
unaware patients falls squarely and persistently upon the nursing
personnel. Quite literally, if subtly, this considerable burden
is built into the organization of the hospital services that deal
with terminal patients. A nother social structural condition in-
46
PART
t w o : ty pe s o f aw areness c o n tex ts
trinsic to the functioning of American hospitals also increases

the nurses burden. We refer to the nursing staff s commitment
to work relatively closely with and around patients. Again,
this structural condition can be better appreciated if one thinks
of the contrast in Asian hospitals, where the family clusters
thickly and persistently around the dying patient. A corollary
of this familial clustering is that the nursing personnel can
remain at a relatively great emotional distance from, and spend
relatively little time with, the patient. The enormously high
patient-to-personnel ratio increases the probability of great
distance and little contact. Although American nurses are
sometimes criticized for a propensity to anchor themselves at
the nurses station, they do spend more time with fewer pa
tients, including those who are dying unaware.
Finally, we must ask about the consequences of closed
awareness for continued interaction with the patient. Perhaps
little need be said about this aspect of the closed awareness
context. It should be abundantly clear that closed awareness
can change smoothly and easily, or explosively and brutally,
into another type of awareness context, depending on how
the closed awareness was managed, and on the conditions under
which the terminal patient discovered what the staff members
were doing to or for him. A t the m ost painful extreme, he
can feel betrayed; the happiest outcome is that he feels grateful
for their protection and genuine sensibility.
4
Suspicion Awareness:
The Contest for Control
How does a patient who suspects that he is dying try to verify
his suspicions while others, who recognize that he suspects,
simultaneously attem pt to negate his suspicions? The hallm ark
of this awareness context is that the patient does not know,
but only suspects with varying degrees of certainty, that the
hospital personnel believe him to be dying. The consequential
interaction to run a bit ahead of the story can be described
metaphorically as a fencing match, wherein the patient is on
the offensive and staff members are carefully and cannily on
the defensive. U nder conditions of closed awareness there was
little contest for interactional control; now a contest between
patient and staff is characteristic.
This particular awareness context should be distinguished
from two others involving suspicion, neither of which we shall
analyze but which require mention. These are both closed
suspicion contexts. One type is closed because, though the
patient is suspicious, the staff does not recognize his suspicion.
A patient may choose not to reveal his suspicion, and manage
that game for some while, but only rarely can he maintain this
silent and difficult pose for any considerable time without
either being discovered or confirming his suspicions about his
true status. Perhaps a more im portant type of awareness con
text, for our purposes, occurs when staff members wonder
but are not sure whether he really does suspect the worst. Or,
they may wonder whether he really knows but is choosing
not to reveal his knowledge. In either event they are careful
47
48
PART
t w o : t y p e s o f a w areness c o n t ex t s
not to show their own suspicions, unless they specifically wish

to make him suspicious o r fully aware. These awareness con
texts where the patient covertly suspects his true status, or
where staff only suspects that he .may suspect or even k n o w wili be touched on occasionally to highlight im portant features
of open suspicion awareness context, in which staff is certain
that the patient is suspicious.
C O N TR IB U TIN G STRU CTU RA L CONDITIONS
We indicated some of the structural conditions for this
type of awareness in C hapter 3; here we shall discuss them in
more detail. If any structural condition that ordinarily sustains
a closed awareness context happens to be absent, the patient
may begin to wonder about his real condition. F or instance, as
noted earlier, m ost patients are not very knowledgeable about
the signs of terminality when they first enter the hospital, but
a patient who is somewhat knowledgeable (a nurse, for ex
am ple) may be much m ore alert to the implications of certain
physical symptoms. Similarly patients who have been chronically
ill not only may recognize genuine and alarming bodily changes
when they occur, but also may be extremely shrewd about the
significance of new procedures to which they are introduced.
Especially in the county hospitals, where the impoverished aged
are sent to die, terminal patients frequently if not indeed char
acteristically recognize their condition when not too far gone
to note it. A nd physicians often admit to cardiac patients before
last-chance operations how slim are the chances for their
survival; so that during intensive care after the operation,
their suspicions of terminality are high and often accurate
despite reassuring statements by staff members.
Closed awareness is usually sustained by the physicians
doing nothing that will arouse the patients suspicion of term i
nality. B ut he may quickly decide not to withhold telltale in
formation altogether but to hint at the truth, so th a t his patient
begins to catch on. A nd because the very names of certain
Suspicion Awareness: The Contest For Control
49
diseases notably cancer are so alarming to laymen, a physi

cian who names such a disease, even though he disclaims its
fatality, may sharply arouse the patients suspicions. In Japan,
one of the authors once interviewed a patient at the Tokyo
Cancer Hospital. This patient explained that he did not know
what was wrong until he was brought to the hospital; then,
seeing its name carved above the entrance, he realized with
horror that probably he had a fatal disease.
The structural conditions that ordinarily maintain closed
awareness may change during the patients hospitalization, and
so change the context to one of suspicion. One change lies in
a failure of organizational efficiency such as inadequate com
munication among daily shifts so that the patient begins to
understand that something is m ore radically wrong than he,
in his innocence, had thought. In addition, the interaction itself
may bring him to the brink of suspicion: family or staff mem
bers may unwittingly flash cues, or adequate answers may not
be forthcoming when his symptoms begin to grow frightening.
And of course, after spending some days in the hospital, the
patient may become more knowledgeable about what is going
on around him, either because he is now more alert physically,
less dazed; or because he is now better acquainted with staff
and hospital. F o r instance, he m ay realize that he is on an
intensive care unit for very sick patients. Besides, the very
structure of the institution tends eventually to arouse the pa
tient s suspicion (especially if he recognizes his bodily deteriora
tion) so far as he comes to understand, if he did not already,
that the hospital is organized not to give him information but
rather to withhold most of it or to dole it out in carefully
measured am ounts.1
The difficulty we do not say the impossibility of keeping
the patient in the dark about his terminality is recognized in
American hospitals, in a myth to which hospital personnel
1 Hans O. Mauksch, Becoming a Nurse: A Selective View, The
Annals of the American Academy o f Political and Social Science
(Medicine and Society), March 1963, pp. 88-98.
50
PART
t w o : t y pe s o f a w areness co n tex ts
commonly subscribe: they believe that most terminal patients

really know that they are dying, whether o r not they openly
display that knowledge. They reason that patients m ust know
because, given a combination of alarming symptoms, the hos
pitalization itself, and the many telltale cues afforded by various
institutional arrangements, it is impossible to keep the knowl
edge from them. Add to this that the physician may tell the
patient indirectly, by his tone of voice o r his lack of optimism.
So staff members reason that even if a patient does not talk
about his impending death or test bystanders to validate his
suspicions, nevertheless he recognizes that he is dying. This
m yth is supported by a quasi-psychiatric common sense notion
that the patient knows way down deep even if he doesnt
know consciously.
Just because personnel apply this line of reasoning to certain
patients does not mean that they apply it to all and by no
means all patients do suspect their terminality. B ut when the
hospitalization period is long enough and the patient sufficiently
sentient, continued interaction with others is likely to lead
to suspicion of terminality if not, indeed, to full awareness.
W hen he shows this suspicion, either by conversation or less
explicitly and deliberately by gesture, then an open suspicion
context exists.
T H E CO N TEST F O R C R U C IA L IN FO R M A TIO N
Once a patients suspicions of terminality have been aroused,
for whatever reason, his problem is to confirm them .2 W e shall
ignore for the present the situation in which the term inal pa
tient is so determined to avoid confirming his suspicions that
* Sometimes patients suspect or believe themselves terminal when
they are not. Then the staff tries to dispel the error by direct state
ments and disconfirming evidence. In one instance we observed, the
patient remained suspicious, and continued testing the staff members
long after they flatly told her she was merely ill. This kind of case
highlights the difficulty a patient can have because of his relative igno
rance of medicine, and when he believes he cannot trust his physician or
nurses. The difficulty, of course, is heightened when his illness necessi
tates medical procedures which he finds frightening or threatening.
51
he refuses to read correctly even very clear signs. But when

a patient defined as terminal begins to suspect that definition,
ordinarily he attempts to obtain some confirming evidence.
His m ajor problem comes down to this: he m ust either detect
or elicit signs that will confirm his suspicions. He needs crucial
information, but how can he get it?
The patients actual resources are exceedingly slim. Con
trast his situation with two suspicion situations that com
monly occur outside hospitals. Consider a wife who suspects
that her husband has a mistress. She too must either detect
or elicit confirming signs. Among her resources are the follow
ing: To begin with, she usually knows her husband far better
than the patient knows his nurses, aides, and physician. She
has more right, and certainly more occasion, to engage him in
lengthy conversations during which she may detect or elicit
confirming cues. Also she is physically mobile, so she can
shadow his movements. She may pay someone else to follow
him and question other people about him, or she may ask
those questions herself. She may also enlist the efforts of in
formal teammates she has friends who may observe his
movements or seek cues through conversation with him. And
though she may desperately want the truth immediately, her
need is actually much less urgent than that of many terminal
patients.
Another situation, where the stakes for suspected and
crucial information are high, is that of organized espionage.
In the real world of international relations, one spy system
plays against many others, but for simplicitys sake let us say
that one system is matched against a single other. Among the
well-known tactics for getting ordinarily inaccessible informa
tion are the following: Either you must get men inside the
other system, or people inside must betray information to
you. Some insiders may voluntarily and freely offer their
services because of animus against their government, as did
Czechs within Austrian officialdom during World W ar I. You
may have to buy insiders services; then you must contact
52
PART
t w o : t y pe s o f a w areness c o n tex ts
them or they must contact you. Im portant officials may reveal

confidences to your spies, not recognizing their true identity.
Also the use of shadows is usual: put your m en inside the
country if not actually within its government. You also set traps
for the enemy, whereby he reveals information, whether un
wittingly or with belated recognition. Some trapping strategies
are based on yielding enough m inor information to catch the
enemy off guard; other methods of lowering his guard involve
implementing a policy that conveys your lack of suspicion of
his own line.*
It is interesting that Erving Goffman, writing about the
problems of informational control faced by teams in work
establishments, such as restaurants and stores, has noted that
the audience (clients) for the team s perform ance must not
acquire destructive information about the situation that is being
defined for them . . . a team must be able to keep its secrets
and have its secrets kept. 4 Goffman suggests some types of
person who are privy to the team s secrets, listing precisely
the types th at are potentially dangerous to any spy system.
Among them are the informers, the spotters, the gobetweens and the confidantes.
Now, com pare the dying patients situation with the spy
system. H e has no team and no allies, but he faces an organized
team. Except under unusual circumstances, he must rely on
his own senses and actions and often his movements are quite
restricted to detect crucial signs or to discover outright con
firmation. H e cannot plant allies within the hospital staff, and
only rarely can he purchase information from staff members.
Of course, occasionally someone will consciously betray the
secret out of sympathy for him, but not because he wants
to betray the staff. (The teen-ager who was told about his
8 For illustrations of such methods, cf. Richard Collier, Ten Thou
sand Eyes (New York: Dutton and Co., 1958); Allen Dulles, The
Craft of Intelligence (New York: Harper and Row, 1963); Sanche de
Gramont, The Secret War (New York: G. P. Putnams Sons, 1962);
and Oreste Pinto, Spy Catcher (New York: Harper and Bros., 1956).
4 The Presentation of Self (Edinburgh, Scotland: University of
Edinburgh Press, 1956), in Chapter 4, Discrepant Roles, pp. 87, #.
53
terminality by his friend is a nice example of an informal

team mem ber not kept in line by his own commitment to
silence.) The patient, then, has to use his own eyes and ears.
He does not even have the great resources available to any
spy system for properly interpreting such bits of information
as he is able to obtain. Furtherm ore, he must outwit the staffs
countertactics, if he is to compete efficiently in this contest.
Otherwise he will have to depend principally on sheer accident
to bring him crucial information.
THE PA T IE N T S IN Q U IR Y AN D TACTICS
The patients objective is to get true indicators of his sus
pected status; that is, he wants validating cues that tell him for
sure. H e may attem pt to obtain the crucial information first
hand, by sneaking a look at his medical charts or by trying to
overhear staff conversations. H e may also directly query his
physician or the nurses, Am I going to die? not necessarily
expecting an honest answer but hoping th at his earnest desire to
know will elicit at least a revealing one. Nurses find themselves
confronted from time to time with these queries, which may be
asked suddenly to catch them off guard.
But if a patient receives a negative answer, or if he chooses
not to ask directly, then he may make various kinds of indirect
queries. A simple approach is to announce that he is dying in
order to see how staff members react. A patient may hint that
he is dying without making a straightforward announcement,
again seeking to elicit significant cues. H e may engage a nurse
or physician in conversation about his symptoms, pressing very
hard, then listening intently for the meanings around the edges
of their answers and comments. In certain illnesses, as in can
cer, a great many medical tests may be given; these test situa
tions afford opportunity for conversational gambits. Shrewdest
of all are the conversational snares a patient can set for unwary
personnel, to get them unwittingly to reveal w hat they know or
contradict implicitly what someone else has told him. Those
54
PART
slips and contradictions can tu rn out to be im portant bits of

information.
All these tactics are predicated on the assumption that the
patient is purposely trying to elicit validating cues about his
true status. Frequently, however, the suspicious patient acts
spontaneously, without conscious design, and accidentally hap
pens to bring out the cues. Unwittingly, he can ask em barrass
ing questions, or cry or complain or get angry, or spontane
ously announce he thinks hes dying, and the way personnel
respond to these may help to confirm suspicion.
M uch of the patients success, however, stems from pounc
ing on cues not elicited at all but flashed accidentally. The
hospital milieu continually manufactures potential cues that
arouse and sustain, and perhaps eventually confirm, the pa
tients suspicion. Once he imagines that others are in collusion
to keep something from him, their every word or gesture can
become significant. F or instance, on services where people work
in close quarters, as on intensive care units, a patient often
sees staff members huddled in conversation, and he may over
hear something that touches off, maintains, o r confirms, his
suspicion. Similarly, a patient who has had surgery may anxi
ously scan the staffs gestures and actions for revealing messages
about his possible terminality. The very intensity with which a
nurse cares for him, or the fact that emergency procedures are
later performed on him, may further stimulate his suspicion.
The am ount of time the physician spends or does not spend
with him; the nurses silence, or her chattering: even these
opposite cues may be read identically and pessimistically. Spa
tial cues, as when a patient is moved to another room, can be
read with increased suspicion. So can tem poral cues, as for
instance, when a patient remains long hospitalized after having
expected an early discharge.
But the im portant point about these cues is that the patient
must interpret them, just as he must interpret those he gains
by design. H ere lies his big problem. Unless a remark o r an
action is so clear that it tells the true story immediately, the
55
patient must construct his own version of its meaning. O rdi

narily a suspicious patient does not have sufficient medical
knowledge to read the many signs that may be available. He
must put two and two together. In more technical language,
he must build his cues into a systematic assessment, putting
together a variety of indicators, which together represent an
accurate and complete answer to his basic query. A patient
cannot escape this problem of assessment unless the physician
ends his wondering with an explicit announcement, or unless
the equivalent confirmation is obtained from medical charts
(but these rarely signify impending death with certainty) or
from an overheard staff conversation. To make the m atter even
more complicated, the patients inquiry has two parts: Am I
dying, and if so, when? Consequently, even when he becomes
certain he is dying, the patient still has to discover how much
longer the medical authorities expect him to live. We shall,
however, reserve the tem poral aspects of the inquiry for
Chapter 6.
TH E STAFF: TACTICS AND C O U N TERTACTICS

When a patient conceals his suspicions, so that the staff
continues to believe him unaware, staff members are at a
greater disadvantage in controlling potentially confirming cues
than when they have some inkling of his suspicions.
The first problem for the staff, then, is to determine whether
the patient does suspect. Ordinarily, this problem turns out to
be simple, for a patient tends to reveal anxieties and conjectures
through his actions. Still, occasionally staff members wonder,
without knowing for certain, whether he might not suspect or
even know the truth. A fter such a patient dies, especially if
the staff was unusually devoted to him, debate about whether
he suspected or knew can be endless and unresolved, sometimes
continuing long after death.
Once staff members decide that a patient actually or prob
ably suspects his terminality, then they must act accordingly.
56
PART
t w o : t y pe s o f a w areness co ntex ts
The fact that personnel designate the patients obvious con

firmation-seeking tactics a s 'a kind of testing, bespeaks their
own need for shrewd and careful stratagems. In general, the
tactics are similar or identical to those in the closed awareness
context, but the staff members now recognize the situation as
more delicate. They are skating on thinner ice because the pa
tient suspects his terminality.
If the suspicious patient asks why he is being moved to a
room nearer the nursing station, or is subjected to a given
procedure, the nurses give him reasonable answers. If the pa
tient asks whether he is going to die, they inquire if he asked
his physician, or remark bluntly, I dont know, Im not a
doctor thus signaling that the patient should not ask them,
at least. Accompanying such conventional m anagements is a
controlled face and manner. Two nurses once spelled out how
this is done; they happened to be speaking about managing the
patients suspicious relatives, but they might just as well have
been talking about a patient:
F i r s t n u r s e : A stem face, you dont have to communi
cate very m uch verbally, you put things short and

formal. . . . Yes, very much the nurse.
S e c o n d n u r s e : Be tender but dont. . . .
F i r s t n u r s e : Sort of distant, sort of sweet.
S e c o n d n u r s e : T alk about everything but the condition
of the patient.
F i r s t n u r s e : And if you do communicate with them,
when you are not too much the nurse, you could
talk about all kinds of other things; you know, care
fully circling the question of death.
Confronted by the patients more oblique references to his
dubious future, a nurse may choose to ignore the conversa
tional gambit, or deny that anything but a roseate future
awaits him. As a general personal rule, she m ay evade the
patients conversational testing entirely, by paying strict
attention to her chores and working attentively on the patients
57
body. She may divert the patient by chattering away, or when

the going gets rough escape the room altogether.5 B ut most of
all, the nurses rely on acting as though the patient were ill but
not dying, seeking to convey their impatience with his sus
picions. Typically they conduct themselves in a cheerful, bland,
or brisk manner. It is not surprising that some patients grow
angry or, from the nurses viewpoint, ornery accusing the
staff of stringing them along, or giving them the runaround.
In all of this, teamwork is essential. F o r example, the
nurses customarily share some of the their observations about
the suspicious patient, so that all can present more or less the
same line to him. Nurses generally need to develop both com
mon and personal tactics more than the physicians do, because
they spend more time with the patient and are more likely to be
asked extensive indirect queries. (Some patients, perhaps espe
cially those of lower socioeconomic status, do not like to ap
proach their physicians too directly, out of consideration for
their high status.)
In their contests with suspicious patients, nurses are sus
tained by a num ber of resources. As noted earlier, the paucity
of the patients resources favors the staff members in keeping
the secret. So does the hospital organization, both in its spatial
arrangement and in more abstract ways, as when nurses in
voke the hospital rules to counter a suspicious patient. Nurses
also take advantage of the hospital organization when they
avoid spending time with the patient by doing some other work,
by spending time with other patients, or by charting back at
the nursing station. Nurses also can call upon professional rules,
as when they counter a direct query with Have you asked your
doctor? o r pretend that they are not in on his medical story.
More subtly, they can also rely to some extent on rules of con
ventional tact: when a nurse indicates that she does not wish
to converse because she is busy, the patient may be constrained
simultaneously by the institutional requisite of nurses work
* Cf. Jeanne Quint, Institutionalized Practices of Information Con
trol," Psychiatry (May 1965).
58
PART
and the implicit imperative of shes in no m ood for talk. In

short, personnel have various rights due their status which the
patient may feel bound to honor.
Yet those rights are not always granted automatically. They
m ust sometimes be enforced, at least for certain suspicious
patients and at critical moments. Young nurses and new aides
m ust not only learn the common tactics but also develop per
sonal techniques for getting their rights honored. The common
tactics are picked up relatively quickly from more experienced
personnel, but the development of personal techniques can
present problems. Young nursing students reveal this clearly as
they work around suspicious patients. As students, nurses are
taught how to care for patients, physically and sometimes psy
chologically, but rarely if ever are they taught how to handle
conversation about death with patients.6 Young nurses have
difficulty with a num ber of conversational gambits, questions,
and topics. Suspicious patients can complain, ask embarrassing
questions, indicate feelings of hopelessness, and be curious
about diagnosis or prognosis. Like anyone, a young nurse may
be more o r less competent in such situations. Students com
monly find the testing done by suspicious terminal patients
very difficult at first; they have no professionally or personally
satisfactory way to handle it. R ather quickly, however, they
seem to develop the tactics described above, learning to control
interaction largely by asserting their professional rights.
A C H A R A C TER IZA TIO N O F T H E IN TERA CTIO N
The concept of rights does not completely cover what
transpires in this context of open suspicion. Perhaps one should
say that each staff mem ber may have to lay claim to his
status during interaction with the patient. We use the term
claim rather than expectations, because expectations are
more characteristic of highly structured interaction, especially
Cf. Jeanne Quint, The Nurse Student and the Dying Patient (New
York: Macmillan, forthcoming, 1966).
59
ceremonies and rituals. If you expect others to act appropriately

toward yourself, but you and they share no firm understanding
about this, then your expectation is more suitably viewed as a
claim.7 In the sometimes strenuous interactional contest be
tween patient and staff, no clear rules express honored rights;
the nurse (o r aide or physician) must claim a given professional
status (which includes working as she sees fit) and she must
invite, persuade, or even force the patient to accept that claim.
When an aide tells a patient who tries to probe about possible
terminality, Honey, just banish those m orbid thoughts, she
is saying, in effect, Do not force me to talk about that sub
ject; she is telling the patient to behave himself in her
presence by remaining silent about a subject she does not wish
to discuss with him.
It is worth exploring the assertion that staff members must
invite, persuade, and even force a patient to accept their
claims, and that they must also deny his claim to be told about
his terminality. They deny his claim by refusing his invitations
to talk, to drop hints or even to disclose, and by rejecting his
efforts to force them into unwanted interactional stances, as
when he demands a straight answer or attempts to trap them
into a revealing cue. In turn, the patient must either accept or
attempt to deny the claims they make upon him. The stakes
in this game involve not merely information or status in the
hospital setting, but personal and professional identity.8 For
precisely this reason, staff members as well as patients tend to
be under considerable strain.
The contrast with closed awareness interaction is immedi
ately apparent, for then there is little contest in maintaining
Anselm Strauss, Mirrors and Masks (New York: Free Press of
Glencoe, 1958), p. 87.
Nelson Foote has phrased this matter of stakes succinctly: Fully
to analyze and understand the participation of any person in an episode
of interaction, account must be taken of wliat was authentically at stake
for him in that situation. . . . See his Concept and Method in the
Study of Human Development, in M. Sherif and M. Wilson (eds.),
Emerging Problems in Social Psychology (Oklahoma City: University
of Oklahoma Duplicating Service, 1957), p. 36.
60
PART
t w o : t y pe s o f aw areness co n tex ts
the awareness context. It is precisely the contest aspect of the

open suspicion context that calls for the strategems outlined
above. This contest is in contrast with what happens when the
suspicious patient determinedly attem pts to deny rather than
confirm his suspicions. Then he searches for, o r elicits, denial
cues rather than validating cues. If the staff members recognize
that he is searching for denial cues, then ordinarily they play a
supporting game rather than participate in the interactional con
test described above.
C HANGES IN T O O T H E R AW ARENESS C O N TEX TS
The open suspicion context is a very unstable one. As we
rem arked earlier, patients may evince various degrees of sus
picion. In general, the m ore suspicion, the fancier the footwork
staff m embers m ust use to avoid final confirmation at best
they may only succeed in keeping him suspicious or perhaps
lessening his suspicions. This suggests not only that the degree
of suspicion can change over time but that change may fluctuate
according to the nature of the cues the patient picks up from
time to time and his interpretation of them. Unless, however,
a patient can be kept oscillating between more and less sus
picion, the open suspicion awareness context tends to be con
verted into other types. This tendency occurs even though
patients have relatively limited resources available for con
firming their conjectures; since the possibility that staff may
unwittingly drop cues more than compensates for this de
ficiency.
A patient of course is sometimes lulled into believing that
his suspicions were completely unfounded, and that he is going
to get better or at least not worse. This interactional outcome
(a closed awareness context) is not unknown to hospital
personnel, especially when a patient earnestly wishes not to
learn the truth about his condition or when they carry their
collusive tactics to the point of sending him home, even though
they really expect him soon to return to the hospital. A more
61
common situation, however, is that staff members believe the

patient has been fooled when in reality he merely refrains
from voicing his partly or fully confirmed suspicions. (T he first
is a covert suspicion context, and the second a variant of the
closed context.)
But the two awareness contexts into which open suspicion
is most frequently transform ed are open awareness (where
everyone openly displays knowledge of term inality) and mu
tual pretense awareness (where both staff and patient know
that the other knows, but nothing is said openly). We shall
discuss these contexts in later chapters; here we wish only to
note briefly that the patient can become aware through staff
members purposeful confirmation of his suspicions. There are
occasions when someone knowingly engineers, o r attempts to
engineer, the patients full awareness of his impending death.
The rules pertaining to indirect disclosure are ambiguous
enough to perm it staff members not to disclose directly, but
to drop sufficient cues for pretty fair confirmation when they
feel the patient really should become aware. As we noted
earlier, the staff members will sometimes wonder whether a
patient suspects his terminality and will feed him subtle con
firmatory cues, hoping he will signal back. Thus a chaplain
told us he had intentionally read to a dying patient a psalm
that referred to death. The patient, who was in a very weak
ened condition, seemed to press back slightly in return, but the
chaplain did not know whether this response represented a
genuine message.
C ertain patients are probably m ore likely to get their con
jectures confirmed in these indirect but intentional ways. A l
though we have no direct evidence on this point, certain
patients undoubtedly can exert m ore powerful claims for
disclosure on the staff the very old, the well liked, the
V IP s, and certainly those who seem very eager to know
about their terminality and seem of sufficiently strong character
to take that knowledge. In any case, the circumstances under
which patients suspicions are confirmed intentionally supple-
62
pa r t t w o : t y p e s o f aw areness c o ntex ts
m eat the m ore frequent circumstances of unwitting indirect

disclosure.
CONSEQUENCES O F SUSPICION AW ARENESS
The im portant consequences of the suspicion awareness
context are easy to imagine. Concerning patients: they may
become irrevocably comatose, or die, without confirming their
suspicions. All the consequences of closed awareness noted in
C hapter 3 then obtain, unless they are negated by prudent
actions taken just in case like drawing up wills. O n the
other hand, patients can be made frantic by the evasiveness of
physician and nurses, when they suspect they are being given
the runaround. Oscillation between more and less suspicion
can alternately lower the patients spirits o r raise them and
affect his actions accordingly. The quality of nursing care
given such patients certainly the quantity, at least is affected
sometimes adversely and perhaps sometimes for the better.
Concerning family members: as anyone who has participated
in this collusive game of silence around the suspicious patient
is aware, it may engender a fantastic psychological strain,
especially if played for very long. Family members are also
subject to the various advantages and disadvantages in dealing
with a not fully aware relative, such as those associated with
his open acceptance or fear of death.
The nurses, of course, bear the brunt of the hospital situa
tion. Of course, they have more experience in this game than
family members and usually are far less involved in the patients
fate; nevertheless they may experience considerable strain.
Nurses tell us that they are often greatly relieved when the
patient learns the truth, for then one need no longer be so
gingerly in conversation nor keep ones guard so high. The
nurses reference to testing itself reveals how significant that
activity of the suspicious patient is for their own work. M ain
taining suspicion may protect the nurses from distressing scenes
in d embarrassingly graceless dyings; but when the patient is
63
so well protected, when he recognizes very late perhaps in

the very last moments of life that he is dying, then his be
havior is all the more difficult for the nurses to cope with. At
any rate, the nurses supporting game does help the attending
physician, since it relieves him of a good part of the inter
actional m anagement of a suspicious patient.
A few consequences for the hospital itself should be noted.9
A patients open suspicion produces not only the division of
labor between physicians and nurses we have already remarked
upon, but also a collective mood of tenseness that quite literally
gets built into the tem porary structure of a hospital service.
When the patient reacts against the line they are handing m e,
that reaction also becomes part of the tem porary structure.
More perm anent is the kind of informal training of new nurses
and aides that a rough suspicious patient provides. And
again, the general organization of work is affected by the need
to handle the patients open suspicions (often more than one
patient at a tim e), to hold them at arm s length, as expressed
in the tactic of being around these sometimes troublesome
patients as little as possible.10
Finally, open suspicion awareness affects continued inter
action. The way that staff members handle the patients sus
picions, whether they confirm or deny them, affects whether the
patient will choose later to play a mutual pretense game or
speak openly of his coming death. If he chooses the latter, there
arises the question of his image of death and his reactions
to it these are the effects to which we shall devote the next
two chapters.
* Here and elsewhere, we leave to the reader any judgments as to
whether a particular consequence or set of consequences is good or
"bad for the hospital.
10Compare this situation, however, with one of mutual suspicion that
sometimes existed in Chinese prison camps during the Korean War,
when each American prisoner began to suspect each other of inform
ing" to the enemy. See Albert Biderman, March to Calumny (New
York: The Macmillan Co., 1963), pp. 52-53.
The Ritual Drama of Mutual Pretense
65
go silently about his own business.1 Yet, as we shall see, some

times the m utual pretense context is neither silent nor un
negotiated.
5
The Ritual Drama
of Mutual Pretense
W hen patient and staff both know that the patient, is dying
but pretend otherwise when both agree to act as if he were
going to live then a context of m utual pretense exists. Either
party can initiate his share of the context; it ends when one
side cannot, or will not, sustain the pretense any longer.
The mutual-pretense awareness context is perhaps less
visible, even to its participants, than the closed, open, and
suspicion contexts, because the interaction involved tends to
be more subtle. On some hospital services, however, it is the
predominant context. One nurse who worked on an intensive
care unit rem arked about an unusual patient who had announced
he was going to die: I havent had to cope with this very
often. 1 may know they are going to die, and the patient knows
it, but (usually) hes just not going to let you know that he
knows.
Once we visited a small Catholic hospital where medical
and nursing care for the many dying patients was efficiently
organized. The staff members were supported in their difficult
work by a powerful philosophy that they were doing every
thing possible for the patients comfort but generally did
not talk with patients about death. This setting brought about
frequent mutual pretense. This awareness context is also pre
dom inant in such settings as county hospitals, where elderly
patients of low socioeconomic status are sent to die; patient
and staff are well aware of imminent death but each tends to
64
The same kind of ritual pretense is enacted in many situa

tions apart from illness. A charm ing example occurs when a
child announces that he is now a storekeeper, and that his
m other should buy something at his store. To carry out his
fiction, delicately cooperative action is required. The m other
must play seriously, and when the episode has run its natural
course, the child will often close it himself with a rounding-off
gesture, o r it may be concluded by an intruding outside event
or by the mother. Quick analysis of this little game of pretense
suggests that either player can begin; that the other must then
play properly; that realistic (nonfictional) action will destroy
the illusion and end the game; that the specific action of the
game m ust develop during interaction; and that eventually the
make-believe ends or is ended. Little familial games or dramas
of this kind tend to be continual, though each episode may
be brief.
F o r contrast, here is another example that pertains to both
children and adults. A t the circus, when a clown appears, all
but the youngest children know that the clown is not real. But
both he and his audience m ust participate, if only symbolically,
in the pretense that he is a clown. The onlookers need do no
more than appreciate the clowns act, but if they remove them
selves too far, by examining the clowns technique too closely,
Robert Kastenbaum has reported that at Cushing Hospital, a
Public Medical Institution for the care and custody of the elderly in
Framingham, Massachusetts, patient and staff members frequently have
an implicit mutual understanding with regard to death . . . institutional
dynamics tend to operate against making death visible and a subject
of open communication. . . . Elderly patients often behave as though
they appreciated the unspoken feelings of the staff members and were
attempting to make their demise as acceptable and unthreatening as
possible. This observation is noted in Robert Kastenbaum, The
Interpersonal Context of Death in a Geriatric Institution, abstract of
paper presented at the Seventeenth Annual Scientific Meeting, Geronto
logical Society (Minneapolis: October 29-31, 1964).
66
PART
t w o : t y pe s o f a w areness c o ntex ts
let us say, then the illusion will be shattered. The clown must
also do his best to sustain the the illusion by clever acting, by
not playing too far out of character. Ordinarily nobody
addresses him as if he were other than the character he is
pretending to be. T hat is, everybody takes him seriously, at
face value. A nd unless particular members return to see the
circus again, the clowns perform ance occurs only once, be
ginning and ending according to a prearranged schedule.
O ur two simple examples of pretense suggest some im
portant features of the particular awareness context to which
we shall devote this chapter. The make-believe in which patient
and hospital staff engage resembles the childs game much
more than the clowns act. It has no institutionalized beginning
and ending com parable to the entry and departure of the clown;
either the patient or the staff must signal the beginning of
their joint pretense. Both parties m ust act properly if the
pretense is to be m aintained, because, as in the childs game,
the illusion created is fragile, and easily shattered by incon
gruous realistic acts. B ut if either party slips slightly, the
other may pretend to ignore the slip.2 Each episode between
the patient and a staff member tends to be brief, but the mutual
pretense is done with terrible seriousness, for the stakes are
very high.
21. Bensman and I. Garver, Crime and Punishment m the Factory,
in A. Gouldner and H. Gouldner (eds.), Modem Society (New York:
Harcourt, Brace and World, 1963), pp. 593-96.
3 A German communist, Alexander Weissberg, accused of spying
during the great period of Soviet spy trials, has written a fascinating
account of how he and many other accused persons collaborated with
the Soviet government in an elaborate pretense, carried on for the
benefit of the outside world. The stakes were high for the accused
(their lives) as well as for the Soviet. Weissbergs narrative also illus
trated how uninitiated interactants must be coached into their roles
and how they must be cued into the existence of the pretense context
where they do not recognize it. See Alexander Weissberg, The Accused
(New York: Simon and Schuster, 1951).
67
IN ITIATING T H E PR ET EN SE
This particular awareness context cannot exist, of course,
unless both the patient and staff are aware that he is dying.
Therefore all the structural conditions which contribute to the
existence of open awareness (and which are absent in closed
and suspicion awareness) contribute also to the existence of
mutual pretense. In addition, at least one interactant must
indicate a desire to pretend that the patient is not dying and
the other must agree to the pretense, acting accordingly.
A prime structural condition in the existence and m ain
tenance of mutual pretense is that unless the patient initiates
conversation about his impending death, no staff m em ber is
required to talk about it with him. As typical Americans, they
are unlikely to initiate such a conversation; and as professionals
they have no rules commanding them to talk about death with
the patient, unless he desires it. In turn, he may wish to initiate
such conversation, but surely neither hospital rules nor common
convention urges it upon him. Consequently, unless either the
aware patient or the staff members breaks the silence by words
or gestures, a m utual pretense rather than an open awareness
context will exist; as, for example, when the physician does not
care to talk about death, and the patient does not press the
issue though he clearly does recognize his terminality.
The patient, of course, is more likely than the staff mem
bers to refer openly to his death, thereby inviting them, ex
plicitly or implicitly, to respond in kind. If they seem unw illing,
he may decide they do not wish to confront openly the fact of
his death, and .then he may, out of tact or genuine empathy
for their em barrassm ent o r distress, keep his silence. H e may
misinterpret their responses, of course, but for reasons sug
gested in previous chapters, he probably has correctly read
their reluctance to refer openly to his impending death.
Staff members, in turn, m ay give him opportunities to speak
of his death, if they deem it wise, without their directly or
PART
68
obviously referring to the topic. B ut if he does not care to

act o r talk as if he were dying, then they will support his
pretense. In doing so, they have, in effect, accepted a com
plementary assignment of status they will act with pretense
toward his pretense. (If they have misinterpreted his reluctance
to act openly, then they have assigned, rather than accepted, a
complementary status.)
Two related professional rationales permit them to engage
in the pretense. One is that if the patient wishes to pretend,
it may well be best for his health, and if and when the pretense
finally fails him, all concerned can act m ore realistically. A sec
ondary rationale is that perhaps they can give him better medi
cal and nursing care if they do not have to face him so openly.
In addition, as noted earlier, they can rely on common tact to
justify their part in the pretense. Ordinarly, Americans believe
that any individual may live and die as he chooses, so long
as he does not interfere with others activities,or, in this case,
so long as proper care can be given him.
To illustrate the way these silent bargains are initiated and
maintained, we quote from an interview with a special nurse.
She had been assigned to a patient before he became terminal,
and she was more apt than most personnel to encourage his
talking openly, because as a graduate student in a nursing
class that emphasized psychological care, she had more time to
spend with her patient than a regular floor nurse. H ere is the
exchange between interviewer and nurse:
I n t e r v i e w e r : Did he talk about his cancer or his
dying?
N u r s e : W e ll, n o , h e n e v e r ta lk e d a b o u t it. I n e v e r
h e a r d h im u s e th e w o r d c a n c e r . . . .
I n t e r v i e w e r : Did he indicate that he knew he was
dying?
N u r s e : Well, I got that impression, yes. . . . It wasnt
really openly, but I think the day that his roommate

said he should get up and start walking, I felt that
he was a little bit antagonistic. H e said what his
69
condition was, that he felt very, very ill that

moment.
I n t e r v i e w e r : He never talked about leaving the
hospital?
.
N u r s e : Never.
I n t e r v i e w e r : Did he talk about his future at all?
N u r s e : Not a thing. I never heard a word. . . .
I n t e r v i e w e r : Y ou said yesterday that he was more or
less isolated, because the nurses felt that he was hos
tile. But they have dealt with patients like this many
many times. You said they stayed away from him?
N u r s e : Well, I think at the very end. You see, this is
what I m eant by isolation . . . we dont communi
cate with them. I didnt, except when I did things
for him. I think you expect somebody to respond to,
and if theyre very ill we dont . . . I talked it over
with my instructor, mentioning things that I could
probably have done; for instance, this isolation, I
should have communicated with him . . .
I n t e r v i e w e r : Y ou think that since you knew he was
going to die, and you half suspected that he knew
it too, or more than half; do you think that this
understanding grew between you in any way?
N u r s e : I believe so . . . I think its kind of hard to say
but when I came in the room, even when he was
very ill, hed rather look at me and try to give me a
smile, and gave me the impression that he accepted
. . . I think this is one reason why I feel I should
have communicated with him . . . and this is why I
feel he was rather isolated. . . .
From the nurses account, it is difficult to tell whether the
patient wished to talk openly about his death, but was re
buffed; o r whether he initiated the pretense and the nurse
uccepted his decision. B ut it is remarkable how a patient can
flash cues to the staff about his own dread knowledge, inviting
the staff to talk about his destiny, while the nurses and physi
cians decide that it is better not to talk too openly with him
about his condition lest he go to pieces. The patient, as
remarked earlier, picks up these signals of unwillingness, and
JQ
PART t w o : t y p e s o f a w a r e n e s s c o n t e x t s
the m utual pretense context has been initiated. A specific and

obvious instance is this: an elderly patient, who had lived a
full and satisfying life, wished to round it off by talking about
his impending death. The nurses retreated before this prospect,
as did his wife, reproving him, saying he should not think or
talk about such morbid matters. A hospital chaplain finally
intervened, first by listening to the patient himself, then by
inducing the nurses and the wife to do likewise, or at least to
acknowledge more openly that the m an was dying. H e was
not successful with all the nurses.
The staff members are more likely to sanction a patients
pretense, than his familys. The implicit rule is that though the
patient need not be forced to speak of his dying, or to act as
if he were dying, his kin should face facts. A fter all, they will
have to live with the facts after his death. Besides, staff mem
bers usually find it less difficult to talk about dying with the
fam ily Family members are not inevitably drawn into open
discussion, but the likelihood is high, particularly since they
themselves are likely to initiate discussion or at least to make
gestures of awareness.
Sometimes, however, pretense protects the family m em ber
temporarily against too much grief, and the staff members
against too immediate a scene. This may occur when a relative
has just learned about the impending death and the nurse
controls the ensuing scene by initiating tem porary pretense. The
reverse situation also occurs: a newly arrived nurse discovers
the patients terminality, and the relative smooths over the
nurses distress by tem porary pretense.
T H E PR ET E N SE IN TER A C TIO N
An intern whom we observed during our field work sus
pected that the patient he was examining had cancer, but he
could not discover where it was located. The patient previously
had been told that she probably had cancer, and she was now
at this teaching hospital for that reason. The intern s examina
72
tion went on fo r some time. Yet neither he nor she spoke

about what he was searching for, n o r in any way suggested
that she might be dying. We mention this episode to contrast
it with the more extended interactions with which this chapter
is concerned. These have an episodic quality personnel enter
and leave the patients room, or he occasionally emerges and
encounters them but their extended duration means that
special effort is required to prevent their breaking down, and
that the interactants must work hard to construct and m aintain
their mutual pretense. By contrast, in a formally staged play,
although the actors have to construct and maintain a perform
ance, making it credible to their audience, they are not required
to write the script themselves. The situation that involves a
terminal patient is much more like a m asquerade party, where
one masked actor plays carefully to another as long as they
are together, and the total dram a actually emerges from their
joint creative effort.
A masquerade, however, has m ore extensive resources to
sustain it than those the hospital situation provides. M asquer
aders wear masks, hiding their facial expressions; even if they
"break up with silent laughter (as a staff m em ber m ay break
down with sym pathy), this fact is concealed. Also, according
to the rules ordinarily governing masquerades, each actor
chooses his own status, his character, and this makes his
role in the constructed dram a somewhat easier to play. H e may
even have played similar parts before. But terminal patients
usually have had no previous experience with their pretended
ntutus, and not all personnel have had much experience. In a
masquerade, when the dram a fails it can be broken off, each
actor moving along to another partner; but in the hospital
the pretenders (especially the patient) have few comparable
opportunities.
Both situations share one feature the extensive use of
props for sustaining the crucial illusion. In the masquerade,
the props include not only masks b u t clothes and other eostuming, as well as the setting where the masquerade takes
72
PART
t w o : ty p e s o f aw areness c o n tex ts
place. In the hospital interaction, props also abound. Patients

dress for the part of not-dying patient, including careful atten
tion to grooming, and to hair and makeup by female patients.
The term inal patient may also fix up his room so that it looks
and feels just like hom e, an activity that supports his enact
ment of normalcy. Nurses may respond to these props with
explicit appreciation how lovely your hair looks this m orn
ing or even help to establish them, as by doing the patients
hair. We remem ber one elaborate pretense ritual involving a
husband and wife who had won the nurses sympathy. The
husband simply would not recognize that his already comatose
wife was approaching death, so each morning the nurses care
fully prepared her for his visit, dressing her for the occasion
and m aking certain that she looked as beautiful as possible.
The staff, of course, has its own props to support its ritual
prediction that the patient is going to get well: thermometers,
baths, fresh sheets, and meals on time! Each party utilizes these
props as he sees fit, thereby helping to create the pretense
anew. But when a patient wishes to dem onstrate that he is
finished with life, he may drive the nurses wild by refusing to
cooperate in the daily routines of hospital life that is, he re
fuses to allow the nurses to use their props. Conversely, when
the personnel wish to indicate how things are with him, they
may begin to om it some of those routines.
During the pretense episodes, both sides play according
to the rules implicit in the interaction. Although neither the
staff nor patient m ay recognize these rules as such, certain
tactics are fashioned around them, and the action is partly
constrained by them. One rule is that dangerous topics should
generally be avoided. The most obviously dangerous topic
is the patients death; another is events that will happen after
wards. O f course, both parties to the pretense are supposed to
follow the avoidance rule.
There is, however, a qualifying rule: Talk about dangerous
topics is permissible as long as neither party breaks down.
Thus, a patient refers to the distant future, as if it were his
73
to talk about. He talks about his plans for his family, as if

he would be there to share their consummation. H e and the
nurses discuss todays events such as his treatments as if
they had implications for a real future, when he will have
recovered from his illness. A nd some of his brave or fool
hardy activities may signify a brave show of pretense, as
when he bathes himself or insists on tottering to the toilet
by himself. The staff in turn permits his activity. (Two days
before he returned to the hospital to die, one patient insisted
that his wife allow him to travel downtown to keep a speaking
engagement, and to the last he kept up a lively conversation
with a close friend about a book they were planning to write
together.)
A third rule, complementing the first two, is that each actor
should focus determinedly on appropriately safe topics. It is
customary to talk about the daily routines eating (the food
was especially good or b ad ), and sleeping (w hether one slept
well or poorly last night). Complaints and their management
help pass the time. So do m inor personal confidences, and
chatter about events on the ward. Talk about physical symptoms
is safe enough if confined to the symptoms themselves, with
no implied references to death. A terminal patient and a staff
member may safely talk, and at length, about his disease so
long as they skirt its fatal significance. And there are many
genuinely safe topics having to do with movies and movie
stars, politics, fashions with everything, in short, that signifies
that life is going on as usual.
A fourth interactional rule is that when something happens,
or is said, that tends to expose the fiction that both parties
are attempting to sustain, then each must pretend that nothing
has gone awry. Just as each has carefully avoided Calling atten
tion to the true situation, each now must avert his gaze from
the unfortunate intrusion. Thus, a nurse may take special pains
to announce herself before entering a patients room so as not
to surprise him at his crying. If she finds him crying, she may
Ignore it or convert it into an innocuous event with a skillful
74
PART
t w o : ty pe s o f aw areness co n tex ts
com m ent or gesture much like the tactful gentleman who,

having stumbled upon a woman in his bathtub, is said to have
casually closed the bathroom door, murmuring Pardon me,
sir. The m utuality of the pretense is illustrated by the way a
patient who cannot control a sudden expression of great pain
will verbally discount its significance, while the nurse in turn
goes along with his pretense. O r she may brush aside or totally
ignore a m ajor error in his portrayal, as when he refers spon
taneously to his death. If he is tem pted to admit impulsively
his terminality, she may, again, ignore his impulsive remarks
or obviously misinterpret them. Thus, pretense is piled upon
pretense to conceal or minimize interactional slips.
Clearly then, each party to the ritual pretense shares respon
sibility for maintaining it. The m ajor responsibility may be
transferred back and forth, but each party m ust support the
others tem porary dominance in his own action. This is true
even when conversation is absolutely minimal, as in some
hospitals where patients take no particular pains to signal
awareness of their terminality, and the staff makes no special
gestures to convey its own awareness. T he pretense interaction
in this case is greatly simplified, but it is still discernible. W hen
ever a staff m em ber is so indelicate, or so straightforward, as
to act openly as if a terminal patient were dying, or if the
patient does so himself, then the pretense vanishes. If neither
wishes to destroy the fiction, however, then each m ust strive
to keep the situation norm al. 4
T H E TR A N SITIO N T O O PEN AW ARENESS
A mutual pretense context that is not sustained can only
change to an open awareness context. (E ither party, however,
may again initiate the pretense context and sometimes get
4 A close reading of John Gunthers poignant account of his young
sons last months shows that the boy maintained a sustained and deli
cately balanced mutual pretense with his parents, physicians and nurses.
John Gunther, Death Be Not Proud (New York: Harper and Bros.,
1949). Also see Bensman and Gerver, op. cit.
75
cooperation from the other.) The change can be sudden, when

either patient or staff distinctly conveys that he has perm a
nently abandoned the pretense. O r the change to the open
context can be gradual: nurses, and relatives, too, are familiar
with patients who admit to terminality more openly on some
days than they do on other days, when pretense is dominant,
until finally pretense vanishes altogether. Sometimes the physi
cian skillfully paces his interaction with a patient, leading the
patient finally to refer openly to his terminality and to leave
behind the earlier phase of pretense.
Pretense generally collapses when certain conditions make
its maintenance increasingly difficult. These conditions have
been foreshadowed in our previous discussion. Thus, when the
patient cannot keep from expressing his increasing pain, or
his suffering grows to the point that he is kept under heavy
sedation, then the enactment of pretense becomes more difficult,
especially for him.
Again, neither patient nor staff may be able to avoid
bringing impending death into the open if radical physical
deterioration sets in, the staff because it has a tough job to
do, and the patient for other reasons, including fright and panic.
Sometimes a patient breaks his pretense for psychological rea
sons, as-when he discovers that he cannot face death alone, or
when a chaplain convinces him that it is better to bring things
out into the open than to remain silent. (Sometimes, however,
a patient m ay find such a sympathetic listener in the chaplain
that he can continue his pretense with other personnel.) Some
times he breaks the pretense when it no longer m akes sense in
light of obvious physical deterioration.
Here is a poignant episode during which a patient dying
with great pain and obvious bodily deterioration finally aban
doned her pretense with a nurse:
There was a long silence. Then the patient asked, After
I get home from the nursing home will you visit me?
I asked if she wanted me to. Yes, M ary, you know we
76
PART
t w o : t y p e s o f aw areness
CONTEXTS
could go on long drives together. . .

She had a far
away look in her eyes as if day-dreaming about all the
places she would visit and all the things we could do
together. This continued for some time. Then I asked,
Do you think you will be able to drive your car again?
She looked at me, M ary, I know I am daydreaming;
I know I am going to die. Then she cried, and said,
This is terrible, I never thought I would be this way.
In short, when a patient finds it increasingly difficult to
hang onto a semblance of his form er healthy self and begins
to become a person who is visibly dying, both he. and the staff
are increasingly prone to say so openly, whether by word o f
gesture. Sometimes, however, a race occurs between a patients
persistent pretense and his becoming comatose or his actual
death a few m ore days of sentience or life, and either he or
the staff would have dropped the pretense.
Yet, a contest may also ensue when only one side wishes'
to keep up the pretense. When a patient openly displays his
awareness but shows it unacceptably, as by apathetically giving
up, the staff or family may try to reinstate the pretense.
Usually the patient then insists on open recognition of his own
impending death, but sometimes he is persuaded to return to
the pretense. F or instance, one patient finally wished to talk
openly about death, but her husband argued against its
probability, although he knew better; so after several attem pts
to talk openly, the patient obligingly gave up the contest. The
reverse situation may also occur: the nurses begin to give the
patient every opportunity to die with a maximum of comfort^
as by cutting down on norm al routines thus signaling th at he
should no longer pretend, bu t the patient insists on putting up
a brave show and so the nurses capitulate.
We would complicate our analysis unduly if we did m ore
than suggest that, under such conditions, the pretense ritual
sometimes resembles Ptolemys cumbersomely patched astro
nomical system, with interactants pretending to pretend to
pretend! We shall only add that when nurses attem pt to change
77
the pretense context into an open context, they generally do

this on their own and not because of any calculated ward
standards or specific orders from an attending physician. And
the tactics they use to get the patient to refer openly to his
terminality are less tried and true than the more customary
tactics for forcing him to pretend.
CONSEQUENCES O F M U TU A L PR ET E N SE
For the patient, the pretense context can yield a measure
of dignity and considerable privacy, though it may deny him
the closer relationships with staff members and family mem
bers that sometimes occur when he allows them to participate
In his open acceptance of death. A nd if they initiate and he
accepts the pretense, he may have nobody with whom to talk
although he might profit greatly from talk. (O ne term inal patient
told a close friend, who told us, that when her family and
husband insisted on pretending th at she would recover, she
suffered from the isolation, feeling as if she were trapped in
cotton batting.) F or the family especially more distant kin
the pretense context can minimize em barrassm ent and other
Interactional strains; but for closer kin, franker concourse may
have many advantages (as we shall discuss further in C hapter
6.) Oscillation between contexts of open awareness and m utual
pretense can also cause interactional strains. We once observed
a man persuading his m other to abandon her apathy she had
permanently closed her eyes, to the staffs great distress and
"try hard to live. She agreed finally to resume the pretense,
but later relapsed into apathy. The series of episodes caused
some anguish to both family and patient, as well as to the
nurses. When the patient initiates the m utual pretense, staff
members are likely to feel relieved. Yet the consequent stress
of either maintaining the pretense o r changing it to open aware
ness sometimes may be considerable. Again, both the relief and
the stress affect nurses more than medical personnel, principally
because the latter spend less time with patients.
78
PART
t w o : t y pe s o f a w a r en ess co n tex ts
But whether staff o r patient initiates the ritual of pretense,

m aintaining it creates a characteristic ward m ood of cautious
serenity. A nurse once told us of a cancer hospital where each
patient understood that everyone there had cancer, including
himself, but the rules of tact, buttressed by staff silence, were
so strong that few patients talked openly about anyones con
dition. The consequent atmosphere was probably less serene
th an when only a few patients are engaged in m utual pretense,
but even one such patient can affect the organizational mood,
especially if the personnel become involved with him.
A persistent context of m utual pretense profoundly affects
the m ore perm anent aspects of hospital organization as well.
(This often occurs at county and city hospitals.) Imagine what
a hospital service would be like if all term inal patients were
unacquainted with their terminality, or if all were perfectly
open about their awareness whether they accepted or rebelled
against their fate.5 W hen closed awareness generally prevails
the personnel must guard against disclosure, but they need
not organize themselves as a team to handle continued pretense
and its sometimes stressful breakdown. Also, a chief organiza*
tional consequence of the m utual pretense context is that it
eliminates any possibility that staff members might work with
patients psychologically, on a self-conscious professional basis.
This consequence was strikingly evident at the small Catholic
hospital referred to a few pages ago. It is also entirely possible
that a ward mood of tension can be set when (as a former
patient once told us) a num ber of elderly dying patients con*
tinually communicate to each other their willingness to die,
but the staff members persistently insist on the pretense that
the patients are going to recover. On the other hand, the p r e
vailing ward mood accompanying mutual pretense tends to be
more serene or at least less obviously tense than when
open suspicion awareness is dominant.
For a description of a research hospital where open awareneN
prevails, with far-reaching effects on hospital social structure, see Renejj
Fox, Experiment Perilous (New York: Free Press of Glencoe, 1959).
The Ambiguities
of Open Awareness
Whenever both staff and patient know that he is d y in g , and
acknowledge it in their actions, the context is one of open
awareness. (W e have touched several times on the structural
conditions that contribute to the existence or appearance of
open awareness.) But openness does not eliminate complexity,
and, in fact, certain ambiguities associated with two properties
of the open awareness context are inevitable. The first the
time and m anner of death will occasionally be relevant to
our discussion. The second property, however, will be central
because it pertains directly to awareness o f death itself.
First, even when he recognizes and acknowledges the fact
of terminality, the patients awareness is frequently qualified
by his ignorance or suspicion about other aspects of his death.
Thus, a patient who knows that he is dying may be convinced
that death is still some months away. Staff members m a y then
conceal their own knowledge of the time th at death is expected
to occur, even though they may refer openly to the fact that
it is expected. Similarly, they may keep secret their expectation
that the patient is going to deteriorate badly, so long as he is
unaware of this contingency. Only rarely will staff members
attempt to make the patient fully aware about time or mode of
death if they judge the time to be sooner than the patient
expects or the m ode unpleasant.
O f course certain structural conditions may occasionally be
present to make the patient m ore aware o f these subsidiary
aspects of his impending death. H e may be a physician who
80
PART
t w o : t y p e s o f aw areness c o n t ex t s
is familiar with the course of his disease, for example. And

patients who have the same disease are often kept together,
so th at each may observe a kind of rehearsal of his own fate
by watching others who are closer to death. American patients,
however, can seldom negotiate successfully enough with their
physicians to obtain good information directly about the mode
of dying when it is likely to be unpleasant; indeed, sometimes
they find it difficult to secure an understanding that the physi
cian will ever be candid about the probable time of death.
The awareness, therefore, of subsidiary features of termi
nality may be of different types than the openness about death
itself. Those other types can change, or be maneuvered, in
various directions: for instance, closed awareness about time
of death can be transform ed into suspicion about time of death;
or suspicion about mode of death can be transform ed into
closed awareness about mode of death. Open awareness about
terminality itself, however, stays quite stable. Although patients
may think of the time of dying as farther away than indicated
by the warnings they have received, or they may choose to
ignore the prospect of living as a vegetable before finally
dying, they display little tendency to forget or repress the
knowledge that they are going to die.
The second ambiguous property of the open awareness
context is another new uncertain element which exists even
though both parties can refer relatively openly to the patients
terminality. The patient may wish to die in certain ways: with
out pain, for instance, with dignity, or perhaps in private. A t
the same time, the staff has its own ideas about the way patients
ought to die, involving not merely the physical aspects of dying,
but also its moral and stylistic aspects. M uch of the consequent
interaction depends on the initial congruence of their respective
ideas, as well as on the congruence that may develop before
the patients death occurs. But the existence of open awareness
about terminality makes this confrontation of persons possible,
and so makes im portant the discrepancy between their ideas
on how to die.
Divergent expectations about appropriate ways of dying
The Ambiguities of Open Awareness
81
reflect in part the common tendency, in American hospitals,

for staff and patients to come from different class and ethnic
backgrounds. Although the patients may come from various
strata, and the staff may also be heterogeneous in both class
and ethnic membership, the majority of the staff is frequently
unlike the majority of patients. In city and county hospitals,
for instance, the patients are relatively impoverished, while
nurses and physicians are middle class. In hospitals supported
by wealthier patients, the nurses are lower in socioeconomic
origin and often ethnically different from most patients. Al
though the degree of difference between patient and staff back
grounds varies in various hospitals, probably it is often great
enough to produce rather incongruent perspectives on dying. An
additional source of discrepancy is that deeply inculcated profes
sional and institutional norms separate staff from patients. To
the extent that dying is a medical phenomenon, doctors and
nurses necessarily have a view of death unlike the laymans.
An example from M alaya illustrates, though in rather an
extreme form, the kind of divergence possible. M alayan hos
pitals are staffed almost exclusively with Chinese and Indian
professionals. When a Chinese patient is near death, his kinfolk
flood the space around his bed and no staff mem ber is sur
prised by their ritual preparations for his death or by their
loud lamentations after death occurs. B ut the M alayans special
beliefs about death and the body make them highly suspicious
of hospitals and the non-M alayans who run them. O n the
somewhat unusual occasions when M alayans permit a very sick
kinsman to enter a hospital, they are quick to whisk him from
the strangers hands when he seems to be dying. Naturally, the
medical staff greets this action with disapproval or paternal
amusement. The staffs reaction is based predominantly on
professional attitudes but also upon ethnic perspectives. In
Japan, however, where the population is culturally homogene
ous, the norms affecting mode of dying seem more uniform than
in either M alaya or the United States, and there is probably less
divergence between patient and staff perspectives.
82
PART
t w o : ty p e s o f a w areness c o n t ex t s
STYLES O F DYING
Once a patient has indicated his awareness of dying, the
most im portant interactional consequence is that he is now
responsible for his acts as a dying person. He knows now that
he is not merely sick but dying. He must face that fact. So
ciologically, facing an impending death m eans that the
patient will be judged, and will judge himself, according toi
certain standards of proper conduct. These standards, pertaining to the way a m an handles himself during his final hours and
to his behavior during the days he spends waiting to die, apply
even to physically dazed patients. Similarly, certain standard*
apply then to the conduct of hospital personnel, who must
behave properly as humans and as professionals. The bare
bones of this governed reciprocal action show through the
conversation between a nurse and a young dying girl. The
nurse said, Janet, Ill try as hard as I can ; and then when
the youngster asked whether she was going to die, the nurse
answered, I dont know, you might, but just keep fighting.
A t first glance, staff members obligation to a dying patient
(regardless of specific awareness context) seems simple enough.
If possible, they must save him; if not, then they m ust give
proper medical and nursing care until he dies. But so simple!
a conception of their obligation ignores the fact that the patient;
deserves their obligation because he has been defined as dyings
Generally he need not actively seek definition of this status^
but need only be defined medically as dying (though on rare:
occasions a patient may have to convince his physician th at
he really is dying). On the other hand, ethical and social a t
well as medical judgments enter into questions such as when;
to try to save a patient and when to give up, how much toi
prolong life when death is certain or the patient is already
comatose, and so on. If such nonmedical judgments are rele
vant, then logically some patients must act so that they do
not deserve as much care as do patients who behave properlyy
83
Consequently, certain patients may receive better medical

and nursing care, not because the staffs obligation is greater,
but because they are defined implicitly as more deserving.
In an open awareness context, patients defined as less
deserving risk the additional judgment that they are acting with
purpose. Since they know that they are dying, their improper
behavior cannot be interpreted as a consequence of ignorance.
Patients known to be aware of death have two kinds of obliga
tion: first, they should not act to bring about their own deaths.
Second, there are certain positive obligations one has as a
dying patient. Let us examine these two aspects of acceptable
behavior in the face of death.
Occasionally patients who discover that they are about to
die attem pt suicide. The staff takes a relatively charitable atti
tude toward suicidal acts if the patient has suffered a great
deal. Otherwise they view such an act as either incompre
hensible o r requiring additional explanation of a psychiatric
kind. (In fact, if the attem pted suicide survives he is likely to
be sent as at an emergency service where we observed to a
psychiatric service. The presumption is that such a person is
mentally ill; patients who attem pt suicide while hospitalized also
get psychiatric attention.) The American attitude that suicidal
persons do not deserve respect or extensive care is rather
starkly expressed on emergency services in our hospitals. If the
suicide arrives already dead, staff members may display little
sympathy for him. If he is alive, they will try to save him, but
with traces of contem pt and disgust. When they fail to save
him, the absence of sadness and dismay is conspicuous. Some
times, when a suicide seems rather far gone, the staff may
sharply curtail its efforts to save him.
In addition to suicides, other types of relatively undeserving
terminal patients arrive at our emergency services: accident
cases known to have caused the accident by carelessness or
drunkenness, drunkards picked up dying from the streets, and
the stabbed, slashed, or battered victims of brawls. Staff mem
bers tend to express distaste and even enmity for all these,
84
PART
whether they are dying or not. From the staffs perspective,

these patients are in the hopsital only because they have acted
improperly. Those who die have broken an implicit pact with
society not to die this way or perhaps a pact to live properly
so as not to die this way. W here the majority of such patients
are drawn from lower socioeconomic strata, or are Negroes,
Puerto Ricans, or other minority group members, then the
staff tends to disapprove of their style of life generally, which
reinforces staff attitudes toward suicide, drunkenness and
violence.
M ore subtle instances of this judging of impropriety
include a patients failure to validate his suspicion that he is
ill. Thus, two interns indirectly berated an undeserving mastec
tomy (breast cancer) patient after her operation by standing
outside her door and talking, in deliberately loud voices, about
her foolishness in not visiting a physician earlier. Other patients
who, after hospitalization, give up by not struggling to stay
alive are guilty of the same kind of lapse. Surrender is justifiable
if death is certain, or if suffering is intense and the battle
for life almost hopeless. But sometimes the staff can find no
justification for the patients giving up. One nurse said about
a cancer patient, Oh, that patient, I still think he didnt have
to die. She saw no good physiological reasons for the patient
to have died just then he simply went ahead and did it. This
particular patient was difficult anyway, even pretending to
fall into comas, so that the nurse understandably found his
behavior, including his dying, quite unjustifiable.
But such patients are considered far less reprehensible than
the suicides or the fatally careless automobile drivers. It is
also instructive to compare them with an 81-year-old woman
found half dead on the street. She was resuscitated in the
hospital but she wished to die, and when the physicians insisted
on blood transfusions, she kept the nurses on the run by
repeatedly attempting to remove the transfusion tube from her
arm. The nurses were upset about her doing this, although they
85
were also disturbed by the physicians insistence on rather

senselessly prolonging her life.
Geriatric patients like this woman have almost earned the
right to die. Indeed, staff members almost literally say so. Truly
deserved deaths are designated by such statements as It is
just as well he died, Its a blessing he died, or He had
nothing to live for. These comments are applied to very wornout geriatric patients, terribly deteriorated patients, and pre
mature infants who might be brain-damaged, to inescapably
comatose patients, and those who are fully sentient but quite
painfully dying. These patients all have a right to die quickly,
and if sentient cannot be blamed for not struggling against
death. Even their occasional acts of suicide may elicit sympathy
rather than adverse moral judgments.
The second aspect of dying improperly is failure to fulfill
ones obligations as a dying patient. People are supposed to
live correctly while dying, providing they understand that they
are dying, but Americans have no clear rules for this behavior.
No script has been carefully laid out for the dramatis personae.
Structural reasons why the script is unclear include lack of
pecific training in ways to die; Americans generally have had
few opportunities to observe death firsthand. (Unlike most
European and Asian countries, the United States in this cen
tury has had no w ar on home territory, which has drastically
reduced the likelihood of civilians seeing actual scenes of
death.) The recent trend toward dying in hospitals and other
Institutions, away from home, has further reduced opportunities
to see people die. In addition, the heterogeneity of the American
population mitigates against consistent and clear-cut societal
directives: even the standards that pertain to the more ritual
istic aspects of death whether the body should be on display
at the funeral parlor or whether burial o r cremation is prefer
able are somewhat hazy. Such standards as exist for behavior
in the face of death are even m ore conflicting and vague. Hos
pital personnel, as we have noted, have had little or no explicit
86
PART
t w o : ty pes o f a w areness c o n t ex t s
training for their roles in the dram a of death, an d like other

A m ericans they have differing opinions as to exactly how a
patient should face death.
Nevertheless, in our hospitals staff members do judge the
conduct of dying patients by certain implicit standards. These
standards are related to the work that hospital personnel do,
as well as to some rather general American notions about
courageous and decent behavior. A partial list of implicit
canons includes the following: The patient should m aintain
relative composure and cheerfulness. A t the very least, he
should face death with dignity. He should not cut himself off
from the world, turning his back upon the living; instead he
should continue to be a good family member, and be nice
to other patients. If he can, he should participate in the ward
social life. H e should cooperate with the staff members who
care for him, and if possible he should avoid distressing or
embarrassing them. A patient who does most of these things
will be respected. He evinces what we shall term an acceptable
style of dying, or, more accurately, an acceptable style of
living while dying.
An example of this style was described by the nurse who
said: Until he became unconscious he was always cheerful.
H ed have a smile for you and never complained. While he
was still able to hed help out on the ward, doing work. This
case was especially hard on all of us. . . . His m other was
like him, never complaining. Staff members are also familiar
with the typical grandm other of the service: the wonderful
elderly lady who is so cheerfully reconciled to death, and who
is so nice to talk with that even the interns and residents drop
in to spend time with her. A psychiatrist might point out that
the grandmothers teach the youngsters how to banish anxiety
about their own deaths; to a sociologist, the significance of the
grandmothers is that their style of dying approaches the ideal
The contrasting pattern what the staff defines as unact
ceptable behavior in aw are dying patients is readily illustrated;
F o r instance, physicians usually honor requests f o r additional
87
(or consultants) opinions, but object to shopping around

for impossible cures: as one physician warned a wife about
her dying husband, faced with death. . . .He might want to
find quacks to tell him they could cure him . 1 O ther types
of unacceptable behavior emerge vividly from our field notes.
Thus; the next quotation involves a patients serious failure to
cooperate in his medical care, and it shows the extremes to
which a physician will go to get such cooperation:
The patient had been moving his arm around a lot so
his intravenous needle was in danger of coming out; he
is very testy at all such rigmarole. The doctor got irri
tated, apparently, at his lack of cooperation and said
that if he took that needle out of his arm, hed die. The
nurse: T hats what the doctor said to his face that
this is what is keeping you here; you pull it out and
youll die.
There are less dram atic instances of unacceptable dying. Thus:
Miss Jons mentioned that M r. Jam es was giving a great
deal of trouble today. They had been trying everything
on him, and nothing had worked. He was refusing all
kinds of things, such as medication, pills, and food,
and having trouble getting him to take his temperature.
They had tried to be persuasive and now there was a
nurse who was trying just the opposite. Miss Jons
giggled and said it wasnt getting her anywhere either.
. . . T o this other nurse I said, Tough customer, isnt
he? She looked stolid and said, He just wants to be
ornery. Then added, H es been ornery all day.
Then there are patients who do not face dying with fortitude
but become noisy or hysterical:
Miss Smith came into the nursing station saying, Im
sick of insight. According to her, this woman patient
gets scared and theres nothing you can do, and then
1Lael T. Wertenbaker, Death of a Man (New York: Random House,
1957), p. 56.
88
PART
t w o : t y pe s o f aw areness c o ntex ts
she gets more scared. . . . A t the staff meeting, Miss

Smith said that this lady gags more and belches, and the
nurse finally decided there was no physical reason for
this. The nurse put on a big dram atic act as she said
this because in a big stew about it. She had spoken to
the patient and told her it was just doing her harm and
would h urt her more. L ater I heard this lady belching
pretty loudly, and the nurses just looked at each other.
O ther patients are defined as making excessive demands and
it does not especially help the nurses to know that patients can
m ake those demands precisely because they know they are
dying.
She finally came back [to the hospital] for the last time
and knew that this was the end and kept saying, Im
getting closer and closer; Im dying, dying. Do some
thing for m e. She wanted somebody by h er all the time
as a security. . . . Shed praise us and things like that.
Then if wed say something to the opposite effect, or not
answer right away, then shed turn against us. She was
just scared all the time. . . . She was a great m ental
strain on me. . . . My reaction to her was quite hostile
at times because I didnt know how to handle her and
I got angry with her. I felt she should accept a lot of
these things and she didnt. . . . I guess some feel that
in their life they have things that they set out to accom
plish more o r less, so that death is easier for them than
others.
The nurses last comment clearly enunciates the doctrine that
people ought to face their deaths properly, though some do
not. In addition, her remarks help to clarify why nurses may
feel anger as well as guilt when a patient emphasizes their
inability to give adequate care or com fort while he is dying.
Crying or sobbing is one thing but when patients wail,
cry out in terror, complain, accuse you of doing nothing, or
refuse to cooperate in their medical or nursing care, that is
something else again. Perhaps m ost unnerving to some staff are
patients who become apathetic; they sometimes seem hostile
89
and even reproachful. A change of this kind is particularly

disturbing in a patient who previously has been cheerful and
likable. We remember a young cancer patient who returned
to the hospital in just such an apathetic condition, thereby
causing widespread consternation among the nurses. W hen one
day he recovered his spirits slightly and started to fight with
the nurses over minor matters, as he had previously, they were
delighted. Apathy in such a patient is perceived by the staff
not merely as giving up but also as an implicit message that
he is finished with everything and everyone including the staff.
For instance, a previously m entioned courageous woman who
finally gave up, signifying this by permanently closing her eyes,
so distressed her family, as well as her private nurses, that a son
finally flew in from across the country. H e persuaded his mother
to open her eyes again, because her withdrawal was hurting
everybody so.
In general, then, the staff appreciates patients who exit
with courage and grace, not merely because they create fewer
scenes and cause less emotional stress, but because they evoke
genuine admiration and sympathy, as well as feelings of pro
fessional usefulness. It is difficult to admire a patient who
behaves improperly even though one can sympathize with his
terrible situation. W hen a patient knows he is dying, he must
present some kind of dying self to the world. People cannot
help judging him, even if by diverse and not altogether explicit
standards. Occasionally a patient provides such a model of
courage and fortitude that the staff remembers him with ad
miration long after his death. In the reactions of staff members
is not only respect for a great hum an being but gratitude for
being allowed to participate in the near-perfect dram a of his
dying.
TH E M A N A G EM EN T O F A C C EPTA B LE DYING
Patients who already know how to die by reasoned judg
ment, or by intuition, or by sensing the staffs implicit standards
90
pa rt t w o : t y p e s o f a w areness c o n tex ts
make life easier for the staff. Staff members need not work
at getting these patients to behave properly, since they take
considerable responsibility for dying properly if not with
dignity, then with minimal loss of self-control. (F o r instance,
one American nurse described an old-world patient who closed
off his life with great ceremony: he called his family to his
bedside, so that he could say farewell, then thanked the nurse
for all her care, and finally closed his eyes to sleep and die.)
The staff members need only align their own conduct with the
patients particular but acceptable style of facing death.
Patients who do not die properly, on the other hand,
create a m ajor interactional problem for the staff. The problem
of inducing them to die properly gives rise, inevitably, to a
series of staff tactics.2 Some are based on the patients under
standing of the situation: staff members therefore command,
reprimand, admonish and scold. Thus one nurse spoke firmly
to patients, or relatives, when they began to make too much
noise at the approach of death. A nother scolded a patient, If
you can do all that, why you can drink fluids too. If youd
eat and drink, then the doctor wouldnt have to stick needles
in your legs. And patients are told to stop incessant crying,
It doesnt help you and it doesnt make me feel any better.
Threats are also used:
I asked what they did about such patients generally, and
she went into a lovely imitative act: she put on a firm
voice and appearance and told the imaginary patient
just what would happen to him if he didnt take what
ever he was supposed to take, or do whatever he was
supposed to do. If you dont do this, then these conses Getting the patient to die properly is, of course, also a problem
in the suspicion awareness and mutual pretense contexts. In the latter
situation, however, the tactics characteristic of open awareness can also
be used; or the staff can occasionally shatter the pretense, but then
quickly return to it. When a patient is suspicious of his terminality^
inducing him to die properly is managed less directly and fully than
when the patient is aware. Consequently, we have chosen to discuss
acceptable dying mainly in connection with the open awareness
context.
91
quences will follow, physically and medically, is what

she got across, saying also and then well have to put
a tube down your stomach.
But these negatively toned tactics commands, reprimands,
threats are supplemented, and often overshadowed, by others
through which personnel attem pt to teach patients how to die
properly. In effect, the nurses signal what they define as unac
ceptable behavior and point out what is acceptable. Sometimes,
when she is angry enough or intimate enough with the patient,
a nurse can talk at some length with him, and this talking,
perhaps by accident rather than design, may help teach the
patient how to die. But generally the teaching process consists
of fragments of interaction coaxing, brief coaching, and the
like rather than extended coaching sessions. Teaching usually
consists of something like this: Well, you just have to have
patience for a while, we just cant do everything all at once
that you want us to. Y ouve got to try, Mrs. Smith, rather
than this:
A nother time I got the feeling the patient wanted her
children to cry about it because shed moan and groan
in front of them. . . . I got rather angry with her because
I had talked to her so many times before about dying
. . . and she was beginning to accept a lot of it; and
she was trying to prepare her daughters for it, and then
she would turn around and want them to feel sorry for
her. I missed my chance that one day. I was going,
really, to tell her.
Nurses may call on higher authority or on allied forces
to remind or help teach the patient what is required. Thus the
physician may be asked to command the patient to complain
less or cry less loudly. In one hospital we observed that the
nurses gladly delegated to a skillful chaplain the task of
breaking through to frightened, hostile or silently apathetic
patients. Psychiatrists sometimes perform similar functions.
Where open awareness is the prevailing context, the other
92
PART
patients as a group can also be relied on to exert implicit

and sometimes explicit controls on unacceptable dying.3
Of course, staff members may ask the patients relatives
to help manage their dying kinsman, but in most cases their
unsolicited cooperation is depended on for encouraging ac
ceptable dying. F or instance, one m an actually did some of the
nurses work because he was not sympathetic to his wifes
withdrawal. He made her get out of bed and walk; he jollied
her, o r ordered her to eat when she didnt want to eat. A nurse
described another patient who spoke angrily to her; his wife
would intercede: D ont be like that, dont talk to the nurses
like that. A nother nurse rem arked: I encourage the relatives
sitting there holding their hand it may help.
W hen the patients kinsmen visit the hospital they too, of
course, are expected to behave properly in the face of dying
and death. Although we shall consider the familys reactions
at some length in later chapters, we will note here that kinsmen
who fail to conform to the staffs notions of propriety are likely
to be coached, reprim anded o r even subjected to disciplinary
measures. A rather extreme instance of this was observed once
on the cancer service of a teaching hospital. Family members
spoke loudly and clearly to each other about a dying ado
lescent in his presence; they laughed often, and carried on
generally. The nurses felt that this was causing the boy great
distress, and, unable to control the situation, they finally
appealed to the physician. H e banned everyone but the parents
from the room, explaining his action to the parents.
Visiting n u rse s. sometimes comment unfavorably on the
households they visit in term s of the members behavior in the
face of death: the family m ay behave worse than the patient,
or vice versa. Thus one harassed nurse exclaimed: I had
another case that was just the opposite the patient was very
nice and the person taking care of her, her daughter, was
awful. A nother nurse who had genuine problems of manage Ren6e Fox, Experiment Perilous (New Yofk: Free Press of Glen
coe, 1959), pp. 139-181.
93
ment rem arked that: The m other was irritable and swore
at both the daughter and the R N s. Hospital personnel do
not usually have to contend with difficult families as often or
as directly as visiting nurses, but when they do, they may en
counter the same problems of management.
O ur discussion of the style of dying would be incomplete
if we failed to note that the patients own conceptions of the
proper way to die may contradict staff expectations. When
a patient has his own clear conception of propriety, staff mem
bers may have difficulty inducing him to accept or conform to
their ideas. Occasionally, for example, a patient displays so
much composure as to seem unnatural: The hospital staff
reacted to W erts behavior, which changed in no particular
from his norm al behavior, with awe and awkwardness. I had
hoped they would react with easy frankness toward him. In
stead the opaqueness of the relationship between patient and
medical attendants seemed to increase. . . . One . . . young
nurse . . . said to me, B ut hes so natural. 4
One patient we observed greatly unnerved the nurses be
cause he insisted upon signing his own autopsy permit. Others,
as already noted, simply want to die in private, alone with
themselves and perhaps with their God. In fact, some patients
would rather die at home, not because of the familiar sur
roundings, but because then they can minimize interference
with their own managem ent of death. In the hospital, the
nurses tend to see gestures toward privacy as a rejection of
themselves, and find them difficult to understand or tolerate.
And when a hospital patient commits suicide, he represents at
its widest the potential gulf between staff and patient concep
tions of the proper approach to death, for suicide epitomizes the
possibilities for rationally managing ones own death so as to
remove it from staff control altogether. Thus, staff members
may sometimes condone a suicide b u t rarely can fully approve
of it.
4 Wertenbaker, op. cit., p. 66.
94
PART
t w o : t y pe s o f a w areness c o n tex ts
T H E P A T IE N T N EG O TIA TES
The interaction between patient and staff members is also
greatly affected by the patients negotiations with them. Every
sentient patient, whether terminal or not, wants to get certain
things or to have certain events occur; he coaxes, wheedles,
bargain#, persuades, hints, and uses other forms of negotiation.
A patient who is aware that he is dying may attem pt to negoti
ate explicitly about matters pertaining to his own dying.5 F or
what he gains, he can, in return, offer specific cooperation with
the staff, or a generally acceptable style of dying; or even,
negatively, threaten a less acceptable style if they will not
negotiate properly.
M uch of the daily negotiation centers around the pacing
and minimizing of such routines as bathing, eating, taking
medication or undergoing treatments. Understandably, a nurse
wants these routines scheduled, and if they are medically im
portant, she wants them done properly. B ut a weary or de
pressed patient may wish to delay or completely avoid some
routines. Thus, when a cancer patient suggests later cooperation,
the nurses may not urge him to eat or bathe or have his
tem perature taken; or they may cater to his wishes to the
extent of offering to do something for him ( How about a
bath now? ) rather than announcing that something will now
happen.
A patient sometimes rejects a nurses offer, perhaps by
saying that he is not going to get out of bed anymore and thus
implying clearly that his end is near. O r, foreseeing his end,
a patient may bargain to avoid some routine. ( H e kept telling
me he doesnt need a bath, hes not going to be around very
long anyway, and I dont care what you do to me, and Im
going to die anyway. ) H e may negotiate for a delay in pro* Negotiation about dying may occur also in the suspicion awareness
or mutual pretense contexts. We have chosen to discuss negotiation,
however, in relation to open awareness contexts because their negotia
tions are especially explicit,
95
cedure. ( The nurse said hes a t a stage where h es slowing

certain procedures. She wanted to give him a bath and he
said, Well, wait til after the doctor comes and examines m e,
and then he pleaded for another half hour. ) A patient who
died shortly thereafter was described by a nurse: I sympa
thized with him when he said, Dont feel like eating, and
said, Its not gonna make that much difference. In general,
then, a considerable amount of daily negotiation occurs around
the pacing, minimizing and the avoiding of routine.
Negotiation may also involve the time that nurses spend
with the patient and the am ount of talk that occurs. Again,
the nurses tend to take their cues from the patient. If they
feel he doesnt want to talk, they do not force him to do so:
There were times when I knew that he just didnt want to
talk. We kind of sensed it, so we talked when he felt like talking,
and we didnt when he didnt want to . This assessment of
the patients desire to talk may not always be correct but it
is based on a reading of cues from the patient. A patient may,
of course, declare openly whether he wishes to talk. A social
worker noted that some patients are quite verbal about dying
and openly discuss, for example, W hat can I do about my
wife? On the other hand, some patients say quite plainly that
they are finished talking forever and turn their faces to the
wall. Staff members may respect that wish spontaneously or,
feeling that the patient has rejected their good intentions, retreat
in despair or hostility.
The death scene itself may be the subject of negotiation.
Sometimes a patient wants a staff mem ber to stay with him,
perhaps because he simply desires not to be alone, or perhaps
because he is afraid to die alone: He talked to me the day
he died. H e wanted me to stay there by him all the time, be
cause he said, Im going. He got the attention, even though
we were busy. Patients may beg personnel to call relatives
when death is imminent, just as relatives may wring that same
promise from personnel. A patient may also negotiate with his
physician for permission to die at home. H e may try to elicit
96
PART
t w o : t y pe s o f aw areness co ntex ts
the physicians promise not to let him live needlessly long, not
to perm it him to die in excessive pain, and to let him know
when death is quite near. And of course patients negotiate
principally with kinsmen to have things done after death has
taken place.
It is worth noting also that a person who is dying in an
unacceptable m anner faces difficulties when he attempts to
gain his ends through negotiation. Staff members understandably
have less patience for the negotiation and altogether less empa
thy with him. One woman, for instance, behaved so badly
for so many weeks that the nurses gradually took to spending
as little time with her as possible. Shortly before she died, the
chaplain entered her room and found her in considerable pain.
She said, Im looking for the nurses. The chaplain, when
later describing the scene, remarked that Id been pushing
the nurses on this just as hard as I could and was wondering
how far I could go on pushing them. H e helped to rearrange
the patients body on the bed so as to reduce her pain. Then
he and she prayed together. As he left the room, he asked, Is
there something else you would like the nurses to do for you?
She said, Yes, see that one comes immediately. The chaplain
thought to himself that he wasnt sure whether he could manage
that particular feat. In the hallway he found a nurse and, put
ting his arm around her, he asked, How about doing me a
favor? Would you go in and take care of M rs. Plum immedi
ately? She answered, Yes, Ill do it for you as a favor.
NON -M ED ICA L N EG L IG EN C E AND PR ID E
From the staffs point of view, negotiation involves much
more than simply deferring to patients wishes. If the staff
plays ball with a patient, of course, then he is more likely
to die in a m anner acceptable to them. Less obvious in the
negotiation is that they may develop an emotional commitment
to fulfilling their end of bargains made with patients. If they
succeed, they can remember the patient, after his death, with
97
pride, but if they fail o r perceive themselves as having failed

him, then they are likely to feel negligent.
This kind of negligence is not the same thing as medical
or procedural negligence. The professional canons that prescribe
reasonably effective medical and nursing care also encourage
professional pride when a job is well done, and feelings of
negligence when the job is badly done (whether o r not negotia
tion is involved). Patients who might have been saved by
more skillful surgery, or whose lives might have been usefully
extended by better nursing care, stimulate feelings of medical
and nursing negligence. This phenom enon is quite different
from what happens when personnel reckon that they have
failed patients in hum an, not technical, ways.
There are, of course, many possibilities for failure in meeting
human needs of patients who are unaware of their terminality.
One nurse, for instance, remarked, Something that bothers
me is the lack of prayers for the [unaware] dying. The chaplain
sees them when they are put on sick list, but not at the end. . . .
Im wondering if maybe we get hard and callous. Inherent in
the open awareness context, however, are certain additional
possibilities for failing or succeeding in these hum an responsi
bilities including those related to negotiation. A n aware patient
can make certain demands involving hum ane considerations
that he would not make if unaware. F o r example, a patient
who, sensing one evening that he would die that very night,
tried to persuade the evening nurse to call his wife. Believing
that he was wrong, she refused. Pressed, she called a staff
physician, who sustained her opinion. W hen the patient would
not yield, she got additional backing from her supervisor; but
finally, she called the wife. The patient died soon after his wife
arrived. The nurses assent saved her from feeling negligent
because of failure to render hum an assistance, though had she
continued to deny his request she would not have been negli
gent because of any professional error.
A nurse may m ake an explicit promise that she will be
there when the patient dies and if she fulfills the promise,
98
PART
then she can feel that she has done her hum an duty. If she
fails him, she is likely to feel somewat negligent. Promises may
be extended to relatives also. One nurse sent a husband to the
coffee room, promising that if his wife began to die, he would
be notified in time, but she failed to notify him quickly enough,
and afterwards felt very badly.
Negotiations need not take such an explicit form as exacting
and giving a promise. Nurses, thus, sometimes feel negligent
when they cannot converse adequately with patients who wish
to talk about their impending deaths, whether the desire to
talk is announced or only signaled. Here is an unusual but
illuminating example:
We had a m an on a pacemaker. We had him in inten
sive care for five weeks. We were probing his chest, four
times a day; he was arresting. . . . He developed an
infection. As long as they keep draining it, hell live.
But its deep and eventually going to get him. I ran out
of things to say. I had to sit and watch him and pound
his chest for five weeks. . . . I was very awkward be
cause he was very intelligent and very much aware of
what was going on. . . . Every day I had to come in
and say Good morning, how are you today? We
talked, we used to read, it was just terrible. I was never
so much involved with a patient.
Even physicians sometimes betray similar feelings of negli
gence or unease by the very brusqueness with which they reject
invitations to talk while making medical examinations of patients
who are fully aware of their own terminality.
A chaplain, whom we observed, derived great satisfaction
(because of having few feelings of negligence) from his ability
to converse with dying patients about their oncoming deaths
as well as about their post-death affairs. His satisfaction is
highly instructive for two reasons. He could respond to patients
invitations to talk and, indeed, draw them into conversation
better than could the nursing and medical personnel. And he
was wonderfully able to wed his professional standards, with
99
their emphasis on giving spiritual and psychological comfort

rather than medical care, with his sense of the humane of
what was a proper hum an death. The same unity of professional
and hum an standards occurs when physicians and nurses are
able to meet patients requests for hum an rather than purely
medical assistance. The special hazard of an open awareness
context is that the patient can make requests pertaining to his
dying. Not only does this possibility add to the num ber of
possibilities for hum an failure, but it also presents intrinsic
difficulties, for, as we have indicated, personnel may prefer to
avoid references to death.6 They may sometimes feel negligent
because they cannot manage this aspect of terminal care.
One of the complicating features of the open awareness
context is that negotiations instituted by the patient may catch
staff members in crossfires of professional and human responsi
bilities, with consequent inescapable feelings of negligence. The
Instance noted earlier of the patient who begged the nurse to
call his wife also illustrates how a nurse may be required to
choose between satisfying a patients request and following
ordinary institutional and professional practice. Even medical
and human responsibilities, however, may sometimes be in
compatible.
A patient may request his physician not to prolong his life
In the event that he becomes comatose, but the physician, guided
by the medical canon that life should be preserved as long as
possible, may ignore that request, or even renege on his promise
to the patient when the time comes to honor it. But the
physician cannot make such decisions without some turmoil.
Nurses whom we have observed have also often reflected
considerable strain, not always knowing whether it was better
' We predict that the strong trend in nursing education toward incul
cating standards of nursing care based on greater attention to inter
personal relations, and to the patients psychological needs, will
make nurses increasingly vulnerable when they work with patients who
are aware of terminality. We have observed this increased vulnerability
In young nurses, but even more so in graduate student nurses who
were learning the intricacies of giving psychological care to patients
who happened to be aware of their own impending deaths.
100
PART
h a t the patient be kept alive or allowed to die. W hen a physician

:an gain enough support from a professional rationale, he may
escape internal conflict, and consequent feelings of negligence,
jver his medical decisions. F o r instance, once we observed
nedical and hum an standards in especially sharp conflict
between a mother, pleading that her child be allowed to die, on
humane grounds, and a physician, who made his decision on
medical grounds to keep the infant alive under oxygen just
as long as was possible. The physician bluntly told the mother
that a physicians job was to keep life going as long as possible,
and that the decision to do so was his alone, not hers.
AW ARENESS O F C ONTIN GENCIES
We noted earlier that an open awareness context might
be qualified, because though the patient may know about his
terminality he may not know about or suspect other features
of his impending death, such as its time or mode. Of course,
the staff members may not be any better informed, since some
contingencies such as a lingering death or terrible bodily
deterioration may be quite unpredictable. On the whole,
however, the staff members have a much better chance of
knowing or suspecting these matters than the patient, and
certainly earlier than he does. The patients dawning suspicion
or final awareness of such contingencies noticeably affects
interaction in the open awareness context.
When the staff members judge that the contingencies will
not seem unduly unpleasant, then they may reveal them tq
the patient, especially if he asks about them specifically ot
envisions less pleasant contingencies. F or instance, if he queries
anxiously and repeatedly whether his death will be painful, they
may assure him that it will not, emphasizing that his kind of
disease generally does not entail great suffering. If he presset
his physician for approximate time of death and suggests afl
earlier date than the physician believes is probable, then thfc
physician may assure him that he is grossly underestim ating
101
how long he may yet live. Naturally if the patient remains

suspicious, the physician may have to redouble his efforts to
convince him, that is, to make him truly aware of the real, or
probable, contingency.
A patient who becomes aware of these contingencies may
take them into account in managing his own dying. H e need not
panic long before his death, but, knowing he has some months
to live, prepare to meet death with fortitude. He need not
have fantasies about terrible bodily deterioration, but can
prepare for death assured he will be in relative control of his
faculties and limbs. He need not fear that he will pass from
life screaming with pain (which might tem pt him to seek an
earlier and easier way o u t).
But when the staff judges that contingencies associated
with a patients death will seem unpleasant or upsetting, they
ordinarily attem pt to conceal what they know about them.
They find it m ore difficult, however, to conceal their knowledge
from a patient who is already cognizant of his terminality than
from one who is not. F or instance, the staff commonly paces
the medication given to a dying patient so that his increasing
pain can be kept under control, but a patient awaiting his own
death is sometimes likely to become aware of his increasing
medications and wonder about them in connection with his
impending death. Similarly, a patient aware of his terminality
is likely to read signs of bodily deterioration in terms of such
possibilities as: Will I go completely and disgustingly to pieces
physically before I die? Will I go to pieces sufficiently to dis
grace myself as I lie dying? Will I become a vegetable before
I die?
When a patient suspects such contingencies, or begins to
recognize their passage into actual fact his pain increases,
or he is obviously close to death since his strength is ebbing
rapidly then this additional awareness may greatly affect his
behavior. H e may become terrified at the prospect of increasing
pain. He may berate the physician who assured him that pain
could be kept under control, but who has long since realized
102
PART
his inability to do so. Indeed he may fear actual desertion by

the physician if and when the physician believes nothing m ore
can be done for him. The patient may also utilize his new
knowledge to manage his dying according to his own concep
tions of proper dying. F or instance, one patient drove her
physician into suggesting a last-ditch operation that might relieve
her increasingly intolerable pain. Yet she herself had to decide
whether to endure the pain or to risk an operation that she
visualized as possibly making her into a brainless vegetable.
She chose to undergo the operation, preferring the latter alterna
tive to dying with unendurable pain. A nother elderly patient,
previously mentioned, who knew she was being kept only
temporarily alive on m achinery, kept the nurses hopping
because she insisted on removing the blood transfusion tube
leading into her arm. She had chosen to die sooner rather than
later, if she could manage the earlier death.
Physicians and nurses are usually prepared to contend
with the patients eventual suspicion or recognition of con
tingencies, since they have long since seen them as probabilities
or certainties. F o r many of the patients responses they may
have routine tactics. Thus, nurses may attem pt to reassure
him that his deterioration will not get much worse, or they
may urge the physician to snow the patient with drugs
when death is imminent and the patient is screaming with
pain or fear. B ut just as the patients expectations of his own
behavior may go awry in the face of the actual contingencies,
so may the staff members behavior. There are no routine reac
tions when a patient, recognizing that he is dying, right now,
from suffocation, pleads with the nurse to save me!
Often the patient, however, prefers to initiate mutual pre
tense about contingencies of which he is aware. Although
speaking openly about his terminality, he does not make refer
ence either to the time or mode of his death. And he may
prefer to keep silent about his own reactions to that known
time or mode of death. One patient we observed, for instance,
was so frightened about his obviously imminent death (obvious
103
both because of his deterioration and the nurses reactions to

it) that his chin could be observed quivering. A nurse was
asked about the quivering: Well, hes that way all over, inside.
Do the nurses talk to him about it? No, but weve given
drugs for it, so we didnt talk about this. In short, the m ans
reactions to his immediately expected death were understood
by the nurses, but by mutual consent neither they nor the
patient referred to them.
CONSEQUENCES O F O PEN AW ARENESS

A few points about the consequences of open awareness,
though already noted in previous chapters, are worth em pha
sizing here. Awareness of impending death gives the patient
an opportunity to close his life in accordance with his own
ideas about proper dying. H e may finish im portant work,
make satisfying farewells, give gifts to his friends, and leave
detailed plans for his family and estate. The advantages of
openness are also implicit in a chaplains account of how he
and a patient together had reconciled the wife, the children,
and the patients parents to his inevitable death. With or
without staff intervention, a wife and husband may achieve
more amicable relations when both know that he is dying
and can talk about themselves in light of that event. In addition,
the aware patient who dies acceptably stands a good chance
of getting staff members to cooperate with him in the manage
m ent of his dying.
But open awareness has disadvantages for the patient too.
O ther people may not approve of the patients way of managing
his death, and may attem pt to change his ideas or even subvert
his management. Also, a patient may not be able to close off
his life usefully, and with dignity, either because he cannot face
death itself or because he suspects or knows its associated fea
tures. A n aware patient, therefore, may be unable to face
death with equanimity, dying with m ore anguish a n d less
dignity than he might if h e w o e unaw are of his terminality.
104
PART
He may question, with anguish, Why must it be me who

must die? Some patients even question what their lives were
all about as they lie dying. A nd some must decide whether to
accept imminent death or perhaps to prolong their lives some*
what through surgery. Such decisions are often agonizing for
the patient, and his agony while deciding will affect his family
and the hospital personnel.
A patient who makes it easy for himself also makes his
death easier for his family, So far as awareness is accompanied
by acceptable dying, it reduces the strain that would otherwise
be imposed on kinsmen by closed or suspicion awareness. If
the patient is able to attain a kind of psychological closing for
his life before he dies, his kinsmen may be able to share his
satisfaction at the close, and they may treasure their experience
for the rem ainder of their lives.7 On the other hand, some
people cannot sustain conversation with a patient who knows
that he is going to die. A nd if the patient dies in agony or
terror, or otherwise unacceptably, his family would doubtless
have preferred to have kept him unaware of his terminality.
M any staff members, especially nurses, prefer the open
awareness context. As one nurse said, I like . . . it best.
Im more comfortable when I can talk about it with the
patients. Nurses may derive genuine gratification from working
with a patient who allows them to participate in his confronta
tion of death. M any a nurse remembers with satisfaction how
she sat with a patient a few hours before his death, as he
talked about his past life. She has not only provided him with
an empathetic listener, bu t she has been brought into his
ceremonial preparation for passing out of life. W hether or
not they are present during the final hours, m any nurses do
derive pleasure from actively helping a patient face death
properly. While they cannot always give much help to patients
who are recoiling from thoughts of death, they can take com
fort from their own small, and sometimes impulsively com
passionate, actions: The patient reached out to the night nurse.
7 Cf. Wertenbaker, op. cit., and Maurice Goudeket, Close to Colette
(New York: F arrar, Straus and Cudahy, 1937).
The Ambiguities o f Open Awareness
105
She instinctively moved away. The patient said, I want some

body to love m e. The nurse realized w hat was happening, so
she took the patient in her arms and let her cry. It was the
crying of real despair.
O f course, patients who do not die acceptably make staff
members work more difficult than it would have been in a
closed awareness situation, creating, in addition, an emotional
strain, particularly for the nurses. The open awareness context,
in fact, does increase the probability that nurses will be sub
jected to stressful sessions with patients who wish to talk about
their deaths, and openly bem oan their dying. M ost nurses
we have observed cannot take these sessions. They develop
tactics for minimizing their im pact and the probability of their
occurrence; for instance, a nurse who has a knack for talking
with dying patients is drafted to work with those who wish to
talk about their deaths.
What about the hospital itself, how is it likely to be affected
by open awareness? The answer depends on several matters
discussed in this chapter. F o r instance, when patients and
staffs ideas about proper dying are congruent, and where both
can act in accordance with those ideas, then maximum condi
tions exist for what might be called graceful dying. The
research hospital described by Renee Fox, where doomed
patients cooperated manfully with physicians in their usually
hopeless efforts to check progressive deterioration, was char
acterized by high morale and graceful dying. Even patients
cooperated in each others dying dram as.8 In this case, there
actually seemed to develop at least implicit norms for how to
help one another die properly. Also negotiation by both pa
tients and physicians was perhaps exceptionally governed by
fairly explicit agreements developed between them. On most
hospital services, a lower proportion of patients will, in the
staffs eyes, die with acceptable style; th at is, there will be a
greater discrepancy between patients ideas and staff members
ideas about proper dying.
A t most hospitals there also tends to be less openness about
Renee Fox, op. cit.
106
PART
t w o : t y p e s o f aw areness c o n t ex t s
the subsidiary features of dying time and mode than at the

hospital described by Renee Fox. Consequently, the organiza
tion of patient care will be affected by the staffs attempts to
keep those secrets. The staffs work m ust also be organized
to face the consequences that appear when the patient who
finally becomes aware of them then grows angry at being kept
in the dark, or wishes then to engage in a ritual of m utual pre
tense about the circumstances of his dying.
7
Discounting
Awareness
In each of the foregoing chapters, we have described ways in
which others consider a patients actual or potential awareness
of his impending death, in combination with their own aware
ness. But imagine, now, certain situations in which the patients
awareness might be discounted. The hopelessly comatose pa
tient, of course, comes most quickly to mind, but awareness is
also discounted in less obvious cases. O ur discussion in this
chapter is intended to highlight, by contrast, the behavior that
occurs because people usually must take awareness into account.
CA TEG O R IC A L DISCOUNTING
With certain categories of patients, staff members may act
without taking awareness into account. We shall consider four
such categories, beginning with prem ature babies.
Most premies who die are recognized as doomed at time
of birth, or even before. Because an infant cannot know that
it is dying, the nurses on premie services need not take aware
ness into account. They can talk, in the infants presence,
about the time it is likely to die, o r how its parents feel about
its death, or say that they themselves feel its just as well
the infant will die. They can act as if they neither wanted nor
expected it to live, because of its physical immaturity. With
other premies, who at first are expected to live but then take
a turn for the worse, nurses work desperately to prevent
death, sometimes becoming very devoted to the dying baby.
Again the nurses behavior can be quite open, rather than
107
108
PART
masked, as with a sentient adult; what is said and done in any

infants presence need not be affected by consideration of his
awareness. Similar discounting of awareness can occur in the
presence of very young children.
With the parents, of course, awareness must be taken into
account. When the parents openly expect their baby to die, the
nurses can play openly into that expectation, but when the
parents express the hope that the baby will survive, then
parental awareness creates the same problems for nurses as that
of sentient adult patients. Thus, uneasy parents may telephone
the head nurse, who has to handle their suspicion that the
babys condition has become worse than was initially reported
to them. The death of a baby who has been frankly expected
to die is usually announced to its mother, who then tells the
father; but when death is unexpected, the psysician often tells
the father.
A second class of patients with whom awareness can be
discounted is the hopelessly comatose. Personnel are warned
against talking freely in the presence of any patient who is only
temporarily unconscious or has recently become comatose,
since the unconscious one may come to sufficiently to under
stand conversation, and the comatose may remain sentient
enough to understand its drift. Only when comatoseness is un
questionably permanent can staff members completely discount
awareness. Then they tend to act in the patients presence as
if he were a non-person, 1 talking freely even about things
that would m atter to him if he were sentient, just as if he were
not there. This does not m ean that the patients body does not
receive excellent medical and nursing care; but even though
the patient is regarded as if he were still somewhat of a person,
perhaps once liked or respected, considerations of awareness
no longer affect the interaction around him. A fter all, he can
no longer know, or discover, that he is dying. Socially he is
already dead, though his body remains biologically alive.
1 For a discussion of the non-person, see Erving Goffman, Presen
tation o f Self (Edinburgh, Scotland: University of Edinburgh, 1956),
pp. 95-96.
Discounting Awareness
109
Because a socially dead person cannot be aware, some

rather peculiar behavior may occur in his presence. F o r ex
ample, a comatose terminal patient, who happened to be a
young nurse herself, began to lose her hair because of her
particular malady. The nurses, distressed by the way she would
look at her funeral, began to prepare the living corpse for its
later public appearance. They worked assiduously on her hair
but finally, after giving up that task as hopeless, swathed her
head in a turbari. A nother instance of unusual discounting be
havior, one we have mentioned before in another context, oc
curred because of admiration for the patients husband. Refus
ing to accept his wifes comatoseness, he treated her at each
visit as if she were socially alive. The empathetic nurses took to
preparing the patients body for his visits. They fixed her hair
and face so that she would look as pretty as possible. They
dressed her properly. A nd when they were in her room with
the husband, they acted as if she were not what they knew
her really to be. (T heir methods of sustaining this odd pretense
with the husband, as well as their reasons for doing it, were
very different than if they had been trying to persuade a sen
tient patient to dress nicely, so that she would feel more like a
woman. Then the patients awareness is directly taken into
account.)
A third category of term inal patients with whom aware
ness is discounted includes those who are senile rather than
comatose. Where there are many such patients, as in state
mental hospitals and in nursing homes for the aged, death is
not noticed by the senile patients because of their senility. The
dying patient himself can be treated by the staff as if he were
ignorant of his own impending death, although he may be com
pletely conscious, or alive, about such matters as bathing and
eating. No additional discounting takes place when such a dying
patient also becomes comatose, though the staff now has the
additional burden of bathing and feeding him. A nother point:
although the hospital personnel may be very fond of certain
senile patients, o r of patients who later become senile, and
may treat them as if they were much alive, this can be done
110
PART
t w o : ty p e s o f a w areness c o n t ex t *
without acting as if the patient could ever become aware of

his terminality.
When senile patients are consigned to such homes and hos
pitals, family members, too, sometimes tend to treat the patients
as if they were already dead as socially dead as if they were
hopelessly comatose and do not bother to visit them. O ther
families treat the senile as being alive but perhaps not all
here. These families periodically visit their unfortunate kins
man (just as parents visit children who were never all here ),
but even they do not act as if the dying patient were aware
of his forthcoming death.
A fourth group of dying patients whose awareness may be
discounted is those whose awareness matters little to the staff
because of the way they view the patients characteristics. Staff
members respect these patients so little that they give less
thought to the control of cues that might disclose terminality.
M ost strikingly, this kind of discounting occurs on the emer
gency wards of city or county hospitals where, as we have
already observed, middle class professionals confront lower
class patients under conditions encouraging the staff to view
patients as inexcusably negligent in the preservation of their
own lives. Partly because so many lower class patients are
brought in as accident cases, or as victims of razor slashings
or other forms of violence, staff members tend frequently to
disapprove of them. When such patients are brought dying to
the emergency ward, staff members talk about them without
much regard for what they may see, hear, or sense. Staff dis
regard for awareness is particularly evident in the way suicide
cases are handled, for suicides, regardless of social class origin,
are especially likely to arouse anger and tendencies toward
overt disrespect. A genuine effort to save the suicides life may
be accompanied by talking to him, or in his presence, as if
he deserved to die.
111
SITUATIONAL DISCOUNTING
Three types of situations also exist in which awareness is
often discounted. The first involves the conscious use of space
to prevent the changing of suspicion or closed awareness into
open awareness, so that staff members can ignore the patient,
as when they talk about his terminality beyond the range of
his hearing. This use of space can be hazardous, since the
patient accidentally may overhear the conversation or take
special pains to overhear it. Yet a spatial barrier very often
allows personnel to discount the patients awareness.
A second type of situation occurs when patients who are
fully aware of their own terminality behave acceptably and
permit others partly to discount their awareness. Conversation
can flow easily because they do not make death the sole topic
of conversation. N ot every word and gesture need touch on
the mutual awareness of the impending death, though after a
few minutes the m atter of terminality may emerge again, en
tering directly or obliquely into the interaction. Even when a
patient who does not accept his death easily converses with a
staff member, terminality can be on the far edge of the patients
consciousness and not appreciably intrude into their conver
sation. And, of course, many tasks that staff members perform
in the patients presence are so remote from m atters of ter
minality that any incidental conversation taking place at the
same time can be commonplace or related to the tasks them
selves.
U nder these rather ordinary circumstances, the discounting
of awareness is optional; staff members may either ignore
awareness or take it into account. But in the third type of
situation, awareness m ust be ignored because other more im
portant matters are at stake. F o r instance, if a patients heart
suddenly stops beating, staff members must ordinarily take
heroic measures to keep him alive. These measures usually
convey to a patient, whether o r not he was previously aware
112
PART
t w o : ty pes o f aw areness c o n tex ts
of danger, that he is about to die, or that he is very likely to

die soon afterwards. The staff has no option but to bring him
into awareness; the gravity of the situation forces them to
ignore whether he is aware or unaware. Emergency measures
of any kind often are highly visible to a sentient patient, so
that staff members are forced to discount awareness. Personnel
on emergency and intensive care services engage rather con
stantly in this type of discounting.
Similarly, when a physician decides that certain procedures,
tests, or even diagnostic surgery is necessary to determine ac
curately the patients condition, then he risks permitting the
patient to become suspicious about, o r know of, his terminality,
but the stakes are so high that the risk is run. A m an who later
died of cancer in France makes this point vividly in his pub
lished diary:
He prescribed a fluoroscopic examination and X-rays
and I was examined by Dr. C artier himself and by the
radiologist. . . . The examination was incomplete; I
could not retain the barium, fed from below. . . . This
indicated to me that something was badly wrong . . .
as did the m uttered conversation of the two doctors, the
conversation of technical men over a technical problem.
. . . B ut when . . . I came out with the word cancer to
Dr. Cartier, he put his hand on my arm and assured me
that it was most unlikely; there were no signs right up
to the very edge of the suspicious spot, which barium
couldnt reach.
Later, Dr. C artier said, Youll have to have an operation,
heartily assuring him that many tumors werent cancerous but
that:
It was necessary to hurry, to take this while it was
young; he advised me to wait no more than three or
four days . . . I asked him point-blank: If I do nothing
about this, how long will I live? A look of shock
passed over his face . . . he took my arm. But that
would be
trying to
moment,
years if
113
foolish, he said. I told him that I was only

consider all the possibilities. After another
he answered: Perhaps a year, perhaps two
it is mauvaise. 2
In short, this physician had to risk alerting the patient, first

through the X-ray procedure and later by advising a diagnostic
operation. This particular patient forced the issue by direct
questioning, but not until after the phyician had aroused his
suspicion. Regardless of whether the patient questions him di
rectly, a doctors medical responsibility often forces him to
discount the patients awareness.
FA ILU R E TO DISCOUNT
Curiously enough, people sometimes fail to discount when
they should because the patient is dead. But if he is dead,
how can his awareness sensibly be taken into account? If we
seem to be stretching the limits of credibility here, it is only
to illuminate the ordinary instances of discounting while pa
tients are still alive.
When a patient has just died, people who see the corpse
rather soon thereafter tend, under certain conditions, to behave
almost as if the patient were alive and aware of his imminent
death. Temporarily the deceased is regarded virtually as a
sentient person. F or instance, nurses who have gotten deeply
involved with a patient will sometimes back away from post
mortem care of his body. Their avoidance can be interpreted
in many ways: our interpretation is that he is not quite a
corpse to them but still a person, under whose gaze they
would have to make funeral preparations. Relatives, immediately
after the patients death, often speak to him or caress him as
if he were alive. Although they know he is dead, they act as
if he knew what they were saying and doing, because his death
is so recent. Staff members rarely speak to or caress the corpse,
* Lael T, Wertenbaker, Death o f a Man (New York: Random
House, 1957), pp. 6-7.
114
PART
t w o : t y p e s o f a w a r e n e s s c o n t e x t *!
but in a lower key they sometimes act much like kinsmen. The
tendency to treat a recently deceased kinsman as if he were
still alive is reinforced by the nurses or physicians efforts to
protect family members against the shock of making symbolic;
farewells to a disfigured corpse, arranging appearances so that)
the deceased person resembles his former self. When they arc
successful, the effect is not only to fix a more normal imagq|
in the kinsmens memories, but permit them temporarily tcj!
regard and act toward the patient as not quite dead.
v
Persons who are present when a death actually occurs arc
often struck by the remarkably thin line that stretches betweeq
life and death. Momentarily, the person now dead seems hardly
dead because he was so recently alive. As long as this mystity
illusion persists, the onlooker tends to act as if he were not
discounting the awareness of a living person. Again, even hos
pital personnel are subject to this brief illusion.
Staff members who give post-mortem care are open te?
another kind of experience as well. General rules of decency;
and institutional instructions, prescribe that the corpse bi(
treated with respect. One should not handle the corpse roughly;
or make any disrespectful gestures toward it; in fact, nothing
should be done to the body that is not required by post-mortem
care. These rules may be neglected because personnel become
callous or because they are in too great a hurry to complete
the post-mortem tasks. Young nurses who witness such dis
respectful handling may become physically sick, or at least
irate, primarily because they feel that hum ans ought not to be
treated thus. They react not only to what they consider inap
propriate behavior, but also to their sense that this person
(the corpse) would have been horrified had he foreseen what
would happen to him after his death. The deceased persons
awareness is thus taken into account, in a strange way, by the
living, even though, realistically, that problem should disappear
after the patient has died.
O ur interpretation of post-mortem reactions may seem
farfetched, but it does highlight our main point in this chapter.
115
Under most conditions, peoples actions are very much affected

by the problems associated with the patients awareness, or
lack of it. Only under relatively special circumstances do people
ignore these problems, and the circumstances under which
they m ust ignore them are even more unusual. Thus the
natural impulse is, ordinarily, to continue acting as if awareness
were still a crucial issue when it no longer is.
Part Three
Problems of
Awareness
Direct Disclosure of
Terminality
One of the most difficult of doctors dilemmas is whether or
not to tell a patient he has a fatal illness. The ideal rule
offered by doctors is that in each individual case they should
decide whether the patient really wants to know and can
take it. However, 69 to 90 per cent of doctors (depending
on the study) favor not telling their patients about terminal
illness,1 rather than following this ideal individual decision.
So it appears that most doctors have a general standard from
which the same decision flows for most patients that they
should not be told. This finding also indicates that the standard
of do not tell receives very strong support from colleagues.
Many conditions reduce a doctors inclination to make a
separate decision for each case. Few doctors get to know each
terminal patient well enough to judge his desire for disclosure
or his capacity to withstand the shock of it. Getting to know
a patient well enough takes more time than doctors typically
have. Furtherm ore, with the current increase of patient loads
doctors will have less and less time for each patient, which
creates a paradox: with more patients dying in hospitals, more
will not be told they are dying. Even when a doctor has had
many contacts with a particular patient, class or educational
differences, or personality clashes, may prevent effective com
munication. Some doctors simply feel unable to handle them
selves well enough during disclosure to make a complicated
illness understandable. If a doctor makes a mistake, he may
'H e rm an Feifel, Death, in Norman L. Farberow (ed.), Taboo
Topics (New York: Atherton Press, 1963), p. 17.
119
120
PART
t h r e e : pro blem s
of
aw areness
Direct Disclosure of Terminality
121
be liable for malpractice. Some doctors will announce an im

pending death only when a clear-cut pathologists report is
available. Others do not tell because they fear the patient might
become despondent or mentally ill or commit suicide, because
they do not want him to lean on them for emotional support,
or because they simply wish to preserve peace on the ward
by preventing a scene.
A t the same time, a num ber of other conditions encourage
disclosure regardless of the individual patients capacity to
withstand it. Some doctors disclose to avoid losing the patients
confidence if he should find out through cues or accidentally.
Telling also justifies radical treatm ent or a clinical research offer;
it also reduces the doctors need to keep up a cheerful but
false front. Some tell so that the patient can put his affairs in
order and plan for his familys future, or reduce his pace of
living; others, because family members request it. Of course,
if the chances for recovery or successful treatment are relatively
good, a doctor is naturally more likely to disclose an illness
that is possibly terminal; disclosing a skin cancer is easier
than disclosing bone cancer.
The combined effect of these conditions some of which
may induce conflicting approaches to the same patient is to
make it much easier for doctors to apply to all patients a flat
no, he should not be told. F or when people are in doubt
about an action, especially when the doubt arises from inability
to calculate the possible effects of many factors about which
there is little information, it is almost always easier and safer
not to act.2
A doctor who does decide to say yes, tell the patient he
is probably going to die, has in effect decided to transform a
closed awareness context into an open one (o r possibly into
a suspicion context). Our intention in this chapter is to develop
a model for this transform ation; i.e., a model combining the
stages typically present in the response process stimulated by

direct disclosure of terminal illness, with the characteristic
forms of interaction between the patient and staff that attend
each stage of the response process (starting off with how the
disclosure was m ade). Thus our focus is just as much upon
how the staff initiates and attempts to guide and control the
response process through interaction with the patient as it is
upon the patients responses per se.
Many of the standard arguments given by doctors both
for and against disclosure anticipate a single, permanent impact
on the patient.3 He is expected to be brave, go to pieces,
commit suicide, lose all hope, or to plan for the future
and such. But the impact is not so simple. Disclosing the truth
sets off a generalized response process through which the
patient passes. To base the decision about disclosure on a
single probable impact is to focus on only one stage in the
response process; not only does it neglect the other stages, but
also it omits how each stage m ay be controlled by the staff
through appropriate forms of interaction. F or example, to
predict that the patient will become too despondent is to neglect
the possibility that he might overcome this despondency and,
with the aid of a chaplain o r social worker, prepare adequately
for his death and for his familys future. But to expect a patient
simply to settle his affairs is to fail to evaluate his capacity for
overcoming an initial depression, as well as the capacity of
the staff to help him at this stage.
The generalized response process is stimulated by a doctors
disclosure to the patient. The patients initial response is almost
invariably depression, but after a period of depression he either
accepts or denies the disclosure; and his ensuing behavior may
be regarded as an affirmation of his stand on whether he will,
in fact, die. Acceptance of the doctors disclosure may lead to
active preparation, to passive preparation, or to fighting the
2 For a conceptual analysis that applies to why there is less risk

for doctors in not disclosing terminality, see Thomas J. Sheff, Decision
Rules, Types of Error, and Their Consequences in Medical Diagnosis,
Behavioral Science, 8 (April 1963), pp. 97-107.
3 For an illustration of this kind of argument, see the discussion

between two doctors in How Should Incurably 111 Patients Be Dealt
With: Should They Be Told the Truth, Parents Magazine, 71 (January
1963), pp. 196-206.
122
PART
t h r e e : pr o b lem s
o f a w areness
illness. A particular patients response may stop at any stage,

take any direction, or change directions. The outcome depends
first on the m anner in which he is told, and then on his own
inclinations combined with staff management.
A doctor deciding whether to tell the patient, therefore,
cannot consider a single impact, but how, in what direction,
and with what consequences the patients response is likely to
go; as well as what types of staff are available and how they
will handle the patient at each stage. A doctor who says no
to disclosure because the patient will lose hope need not be
in conflict with one who says yes to give the patient a chance
to plan for his family. Each is merely referring to a different
stage of the same process. F or both, the concern should be with
judging whether the patient can achieve the acceptance-active
preparation stage.
O ur conception of the stages characteristic of a patients
response to direct disclosure is based largely on our interviews
with patients, doctors and nurses in the cancer wards of a
Veterans Administration Hospital. In this hospital the normal
procedure is to tell every patient the nature of his illness; as
a result, many patients are told of a fatal illness.4
By and large the patients in these wards are in their middle
or late years, of lower class status, and in destitute circum
stances. Since their care is free, they are captive patients
they have little or no control over their treatm ent, and if they
do not cooperate their care may be stopped. If a m an goes
AW OL the hospital is not obliged to readmit him, or if
it does readmit him, it can punish him by denying privileges.
Because the patients lack financial resources, they typically
have no alternative to their current free care, and lower
class status accustoms them to accepting or to being intimidated
into following orders from people of higher status. Since these
captive lower class patients cannot effectively threaten the
hospital or the doctors, the rule at this hospital is to disclose
4 This norm may be considered an aspect of batch treatment of
inmates of a total institution. See Erving Goffman, T h e Characteristics
of Total Institutions, in Amitai Etzioni (ed.), Complex Organizations
(New York: Holt, Rinehart and Winston, 1961), pp. 312-340.
123
terminality regardless of the patients expected reaction. These

patients seemed to exhibit the full range of responses to dis
closure of terminality which it is our purpose to set forth here.
(We cannot say how the distribution of these patients through
out the range of responses would differ, if at all, for higher
socioeconomic patients.)
So many cases of direct disclosure, and the consequent
variety of response patterns, made this hospital a highly
strategic research site for studying disclosure. In other hospitals
we found only a few cases of direct disclosure, and their general
aspects were the same as those of the VA cases.
DISCLOSING FA T A L ILLNESS
Disclosure of fatal illness to patients in this hospital has
two m ajor characteristics. First, the patient is told that he is
certain to die, but not when he will die. As we indicated
in chapter 2, expectations of death have two dimensions:
certainty and time of demise, with the first more readily
determined in advance. As one doctor put it: In my opinion,
however, no doctor should take it upon himself to say to a
patient, You have ten weeks to live or three months, or
two yearc or any time whatsoever. And another doctor said:
Doctors simply do not know when patients are going to die.
Stopping short of full disclosure tends to soften the blow to the
patient and reduces chances of error for the doctor.
Second, the doctors typically do not give details of the
illness, and the type of patient under consideration usually
does not ask for them. Primarily, this is a problem of com
munication: a doctor finds it hard to explain the illness to a
working class patient, while lack of familiarity with the technical
terms, as well as a more general deference to the doctor, in
hibits the patients impulse to question him. In addition, not
giving details is a tactic doctors use to avoid or cut down on
talk with the patient and to leave him quickly. Thus (as referred
to in Chapter 6) transforming the closed awareness context
to open does not make all im portant issues open to discussion.
124
PART
th r ee: problem s o f
a w areness
Issues such as time of death and mode of dying may be left

relatively closed, or on the level of suspicion over which the
contest for information and control will be waged.
In combination, these two characteristics of disclosure often
result in short, blunt announcements of fatal illness to the
patient. Even the nurses are often shocked by the doctors
bluntness. Nevertheless, they often feel that the patient is
better off for being told because, as one nurse put it, he be
comes philosophical in a day or two.
A short, blunt announcement may be softened, however,
in various ways. One way is to add a religious flavor: Youve
had a full life now, and God will be calling you soon. This
m anner is perhaps most appropriate for older patients. Another
is to muffle the language. To the patients question, Is it
cancer, Doc? the doctor responds, We dont call it th at . . .
and then gives it a technical name that the patient can under
stand only vaguely. The suspicion announcement also dulls
the blow: There is a high clinical suspicion that the tumor
removed was cancer. However, we wont have a pathological
report on it for ten days. The announcement that there is
nothing more to do (to
cure
the patient) can be muffled
with a hopeful lie, such as
but
then who knows,next week,
next month or maybe next
year there may be a drug that will
save you, or by suggesting to
the patient that he join an
experimental program that may help him, as well as mankind.5
Finally, there is the im portant statement that softens any form
of disclosure: We can control the pain.
In some forms, the blunt announcement sharpens the blow
of a disclosure by forcing a direct conjrontation of the truth
with little or no preamble. Some doctors are quite aware of
this directness and favor their colleagues use of this approach.
One doctor says, With average patients, we tell them what
theyve got. A nother says, I dont think the staff as a whole
5 If the patient takes the experimental drug, he continually checks
his condition by asking the doctor: What next, Doc? W hat now?
Am I better? Am I getting well? When the experiment is over and
the patient is still going to die, he must start through the response
process again with a depressing now what feeling.
125
goes along with the hard-boiled approach, but me, I try to

tell them the truth. The hard-boiled announcement is often
linked with a report to the patient of the results of his surgery.
In this hospital, patients are customarily told, two or three
days after surgery for cancer, whether they will die. For
example, one doctor walks into the patients room, faces him,
says, Its malignant, and walks out. To be sure, this tactic
also eliminates having to answer the patients questions. Another
rather direct confrontation is, We werent able to get it [the
malignant tumor] all out. A nother form of sharp announce
ment is the direct retort: when a patient asks, Doctor, do I
have cancer? the doctor replies, Yes, you do. (One doctor
commented, If they ask directly, we answer as directly as
possible. ) Lastly, the implied, but sharp, confrontation of
fatal illness is exemplified by the doctor who met a returning
patient with the order to sell her house and all her things, for
she would not leave the hospital again.
In this group of doctors who generally favor the short,
blunt disclosure there is one who does not. He refuses to disclose
in this fashion because he has had previous experience with
errors due to changing pathology reports, and because he tries,
through surgery, to make the patients last weeks more com
fortable. O ther doctors tend to disagree that his comfort
surgery is useful, but he continues because sometimes he
actually saves the patient for years. This doctor continually
maintains a cheerful and optimistic manner, never directly
disclosing to the patient that he will die, but actually giving
silent disclosure by his offer of comfort surgery or participation
in a clinical experiment. Silent disclosure initiates the mutual
pretense awareness context: both doctor and patient know of
the latters fatal illness, and both know the other knows, but
they do not talk to each other about it. The doctor reveals
the patients fatal illness by oblique reference to it in proposing
comfort surgery o r experimental participation, the meaning of
which the patient clearly understands. The patient thus begins
his process o f response-to-disclosure without the customary
stimulus of direct disclosure.
126
PART
th r ee: problem s
o f aw areness
DEPRESSIO N
The initial response of the patient to disclosure is depres
sion. The large majority of patients come to terms with their
depression sufficiently to go on to the next stage of the response
process. A few do not. They withdraw from contact with every
one and remain in a state of hopelessness. In this limiting sense
they become non-interacting, non-cooperative patients; the
nurses can not reach them. Depression is usually handled
by sedation until the patient starts relating to the staff. In one
case a nurse observed that a patient visibly shortened his life
because of his period of anxiety and withdrawal.
A C C EPTA N C E O R D EN IA L?
A fter an initial period of acute depression, the patient
responds to the announcement by choosing either to accept
or to deny the imminence of his death. In effect, he takes a
stand on whether and how he will die, and this stand profoundly
affects his relations with the staff from that time on.
In general, sharp, abrupt disclosure tends to produce
more denial, than dulled disclosure.6 W hen the disclosure is
sharp, the depression is more immediate and profound, and
denial begins immediately as a mechanism to cope with the
shock.7 T o predict an individuals response, however, one
needs the kind of intimate knowledge of the patient that doctors
would prefer to have. W ithout it, it is very difficult to say which
path to death a patient will take, or for how long. In some
cases, the patients response changes; he cannot hold out against
This hypothesis complements the discussion by Feifel ( op. cit.)
on the importance of how telling is done.
7 For another discussion of denial of illness upon disclosure, see
Henry D. Lederer, How the Sick View Their World, in E. Gartly
Jaco (ed.), Patients, Physicians and Illness (New York: Free Press of
Glencoe, 1958), pp. 247-250. Denial of dying is characteristic of our
society as shown by Robert Fultons data: see Death and the Self,
Journal of Religion and Health, 3, No. 4 (July 1964), pp. 359368. See the analysis of denial of death in American society by Talcott
Parsons, Death in American Society, American Behavioral Scientist, 6
(May 1963), pp. 61-65.
127
accumulating physical, social and tem poral cues. The usual

change is from denial to acceptance, though short-term reprieves
sometimes occur, when patients improve before growing worse,
or when a new drug helps for a few days; and in such cases
acceptance may change, for a time, to denial. The direction
an individual takes depends not only on how he is told, but
also on a variety of social and psychological considerations
impinging on the passage from life to death.
Acceptance: Patients may demonstrate acceptance of im
pending death by actively preparing for death, passively pre
paring, or fighting against it. A ctive preparation m ay take the
form of becoming philosophical about dying, death, and ones
previous life; with family, nurses, social workers and the
chaplain, patients review and discuss how full their life has
been. They may pose the destiny question: Why me? and
try to work through it with the philosophical help of others.
This approach leads the patient to draw the nurses into a dis
cussion which can be very difficult for them, since nurses are
trained by and large to help motivate patients to live, not to
die. But if a nurse is to help a patient prepare himself, she
too m ust accept his impending death and refrain from chastising
him for not fighting to live. Otherwise she is likely to consider
the patient m orbid and tend to avoid his invitations to help
him face death squarely. She will usually try to transfer the
burden to the social worker, nun, or chaplain when they are
available.
Some patients start immediately to prepare themselves for
death through religion. F o r others it is an easy transition to
slip from philosophical to religious terms, a transition often
aided by the chaplain, who then helps the patient prepare
himself.
Another form of active preparation for death is to settle
social and financial affairs, perhaps linking this effort with
philosophical or religious preparation. The typical helpers in
settling affairs are family members and social workers. For
example, upon learning th at he was going to die, the patient
128
PART
o f aw areness
turned to his wife and said, Well, weve got to get every
thing lined up; I promised [so-and-so] my. . .
This im
mediate getting down to the provisions of a will was considered
abnormal by one nurse who said, Ive never seen a reaction
like that, it was almost m orbid. A nother patient began dis
cussing with the social worker the various veterans benefits
they could obtain for his wife, and another tried to get his
wife, who was emotionally very dependent on him, married
to a hospital corpsman. One patient gave up his pain medica
tion long enough to put his financial affairs in order with the
aid of a social worker, for he knew that as soon as he was
too drugged to operate effectively, his family would try to take
over his estate.
To give the patient a chance to settle his affairs and to
plan for the future of his family, is, of course, an important
consideration when a doctor decides whether to disclose fatal
illness.8 He can seldom be sure, however, that the patients
response will take this direction or advance so far. Moreover,
some affairs to be settled are less im portant than others; still,
patients have a chance to pick up loose ends or accomplish
unfinished business. For example, before entering the hospital
for cancer surgery one woman said, I am going to do three
things before I enter the hospital things Ive been meaning to
do for a long time. Im going to make some grape jelly. Ive
always dreamed of having a shelf full of jelly jars with my own
label on them. Then Im going to get up enough nerve to saddle
and bridle my daughters horse and take a ride. Then Im going
to apologize to my mother-in-law for what I said to her in
1949. A nother patient with leukemia quit work and bought a
sailboat. He planned to explore the delta region of the Sacra
mento and San Joaquin Rivers until his last trip to the hospital.
Another form of active preparation is to attem pt a full
8 A study by Dr. Donald Oken showed that MDs are more likely
to disclose to businessmen because they needed to wind up business
matters; i.e., MDs perceived this matter as important: Donald Oken,
What to Tell Cancer Patients: A Study of Medical Attitudes,
The Journal of the American Medical Association, 175 (April 1961),
pp. 1120-1128.
129
life before death.9 This pattern is characteristic of younger

patients (in contrast to older patients who review the fullness
of the life they have h ad ), like the twenty-two-year-old man
who, when told he had between three months and three years
to live, m arried a nurse. If we have only two m onths, she
said, it will be worth it. The patient lived two years and had
a son. Faced by certain death he had achieved the most he
could from life.
Auto-euthanasia (suicide) is another way of actively pre
paring for death.10 It eliminates the sometimes very distressing
last weeks or days of dying. One patient, who had no friends
to visit him, felt that he was very alone and that no one in
the hospital cared, so he tried to hasten his death by suicide.
Some patients try to end their lives while they are physically
presentable, not wanting their families to see their degeneration.
Other stresses that encourage auto-euthanasia are unbearable
pain and the discipline imposed by a clinical experiment to
which one may be irrevocably committed. Others decide to
end their life when they can no longer w ork.11 Still others
prefer auto-euthanasia as a way of controlling their dying as
they controlled their living, thus wresting this control from
the hands of the staff and the rigors imposed by hospital
routine.12
Passive preparation for death, among patients who accept
9 This form of active preparation is appropriate in American society
which stresses the value that death is unacceptable until one has had
a full life; see Parsons, op. cit.
10 Shneidman feels a question deserving of research is why so many
cancer patients do not commit suicide, Edwin S. Shneidman,Suicide,
in Farberow, op. cit.
11 For an account of a cancer patient who planned to commit auto
euthanasia after he could no longer work, see Lael Wertenbaker, Death
of a Man (New York: Random House, 1957).
19 A growing problem that medical staff and hospitals must face
is that people wish to control their own way of dying; they do not want
it programmed for them by medical staff and hospital organization. To
achieve this end, many patients also wish to die at home; Fulton, op.
cit., pp. 363-364. See the analysis of this problem in Jeanne C. Quint,
Some Organizational Barriers to Effective Patient Care in Hospitals,
paper given at the American Medical Association convention, June 24,
1964.
J3 Q
pa rt t h r e e : p r o b lem s o f
a w areness
their terminality, also has some characteristic forms. One is

to take the news in a nonchalant manner, which nurses some
times find disturbing. One put it: But some take it quite
nonchalantly. Weve had several very good patients right to
the end. One that upset me was here when I came. H e was
the hardest for me to see die he was young and not only
that such a wonderful fellow. Even as sick as he was, he was
always kind and courteous. A part from the social loss factor
young and wonderful which usually upsets nurses,13
this nurse also found such a passive outlook on death rather
disquieting. As we noted in C hapter 6, the patients passivity
tends to be judged as an unacceptable style of dying.
Nurses, however, are grateful to patients who approach
death with calm resignation. This response not only relieves
them of the responsibility for cheering up the patient, but also
improves their own morale. Since it would not do for a nurse
to be less calm or resigned than her patient, a patient who
responds in this fashion raises and supports the nurses morale.
The non-verbal patient who simply accepts his fate and does
not talk about it also relieves the nurses of the stress of dis
cussing it, as she often must do with the more actively preparing
patient. The calm patient makes few or no demands on nurses
or social workers. A disquieting aspect of this response is the
loss of contact with the patient: It is very hard for us well
people to really grasp how they feel. One social worker
bridges this gap by sitting with the patient for a time each day.
She reports, Sometimes its a m atter of just touching their
hand whatever is natural to make them feel that you under
stand and care. Another disquieting version of the passive
response is expressed by the patient who emerges from his de
pression only to turn his face to the wall, the spirit drained out
of him , and passively wait to die.
Some patients accept their fatal illness but decide to fight
it. Unlike denial behavior, this fight indicates an initial accept13 Barney Glaser and Anselm Strauss, The Social Loss of pyj*
Patients, American Journal o f Nursing, 64 (June 1964), pp. 119-121.
131
ance of ones fate together with a positive desire to somehow

change it. Three forms of fighting behavior are intensive living,
going to marginal doctors or quacks, and participating in an
experiment. One patient, for example, started going out on
passes and living it up, asserting, Ill beat it, as if he could
hold death off by living life to the full. He kept getting thinner
and eventually died. This mode of resisting death can be readily
transformed into active preparation for death if it increases
the patients fullness of life before he dies.
Taking an outside chance with quacks or marginal doctors 14
gives some patients a feeling that they are actively combatting
the disease. A regular physician may perm it his patient to
go to a marginal doctor, since that visit keeps the patient hope
ful and busy while letting the physician see that the marginal
treatments do not injure his patient. Denied this permission,
a patient who wants to fight his disease in some way may
break off relations with his physician, so that the physician
loses control over both the patient and the marginal doctor.
A rupture like this makes it difficult for the patient to return
to the original doctor after the marginal treatm ent fails.15
The search for a way to fight the fatal illness can also lead
a patient into a clinical experiment. If he does not win his
own battle, he at least may help future patients with theirs. The
chance to contribute to medical science does not, however,
sustain the motivation of all research patients.16 Some, realizing
things are hopeless for themselves and finding the experimental
regime too rigorous to bear, try to extricate themselves from
the experiment. If the doctors decide to carry on anyway,
these patients sometimes interfere with the experiment by
14 In this connection see Beatrix Cobb, Why Do People Detour to
Quacks? in Jaco, op. cit.
15 For an illustration of how the doctor allowed, hence could con
trol a dying patients submission to the rigorous treatment of a mar
ginal doctor, see John Gunther, Death Be N o t Proud (New York:
Harper Bros., 1949).
13 Cf. Renee Fox, Experiment Perilous (New York: Free Press of
Glencoe, 1959). One gets the feeling that patients in Foxs sample were
all highly motivated to go on with the experiments to the bitter end.
^22
PART THREE: PROBLEMS OF AWARENESS
pulling tubes out, by not taking medicine, or by taking an extra

drink of water. Some attempt auto-euthanasia. (See Chapter
11 .)
Denial: Some patients deny they are approaching death

and proceed to establish this stand in their interaction with
staff members. Typical denying strategies are juggling time,
testing for denial, comparing oneself to others, blocking com
munication, becoming intensely active, emphasizing a future
orientation, and forcing reciprocal isolation. These strategies
result in a tug-of-war between staff and patient, with the staff
trying to institute an open awareness context, and the patient
trying to keep the context closed, or at least institute one of
m utual pretense.
In the cancer ward, it is relatively easy to deny impending
death by juggling time, for disclosure, as we have noted, implies
certainty that death will occur but no assurance as to when it
will occur. Patients can therefore invent a time, and this becomes
a way of denying that one is truly dying.17 Some literally give
themselves years. B ut even when a denying person is given a
time limit, he is still likely to start thinking, as one nurse said,
in term s of years, when it really is a m atter of a m onth or two.
A patient can test the staff in various ways to establish
his denial. A negative way is by not testing, or failing to ask
the expected questions. F or example, the doctor who tells the
patient he has a tum or adopts a grave m anner which indicates
that the tum or is malignant, and expects the patient to try to
verify it. B ut the denying patient never asks the doctor or any
one else. O ther patients do test the nurses indirectly, by asking,
Why arent I gaining weight? Why arent I feeling better?
Since these patients can be assumed to know why, the nurse
understands that they want these physical cues interpreted m
17 That in the case of dying, patients will tend to give themselves
more time than they actually have, contrasts with studies ofreco v ery
which show that patients are likely to give themselves few time than it
takes to recover. See Fred Davis, Passage Through Crisis (Indianapo
lis: Bobbs-Merrill, 1963) and Julius Roth. Timetables (Indianapolis.
Bobbs-Merrill, 1963).
^33
such a way as to deny that they indicate impending death.

Patients often ask nurses to manage temporal cues in the
same way, by interpreting their extended stay in the hospital
or slow recuperation in ways that point away from death.
A nother form of testing for denial is the polarity game. By
questioning a nurse or social worker about the m ost extreme
life-or-death implications of his illness, the patient forces her
to give a normalizing answer, which usually locates him a safe
distance from death. F or example, focusing on the dying
implications, the patient asks, Am I getting worse? The
medicine isnt helping. The forced answer is, Give yourself
a chance medicine takes a long time. Or, focusing on the
living implications, a patient replies to a social worker trying
to figure out his VA benefits for his family, Well, all right
but it wont be for long, will it? Since the social worker cannot
confront the patient with coming death if he really doesnt
know, she in effect denies it for him by classifying him with
the living but disabled: This is what welfare is for to help
the families of men who are disabled and cant work. Here,
the social worker understood the patients words as a request
for assurance that he wasnt going to die, and she responded
appropriately.
Patients may deny their fate by using other patients as
comparative references. Two common types of comparison are
the exception and the favorable comparison. A patient using
the first approach becomes very talkative about other patients
with the same disease. H e adopts a manner, or style, like that
of staff members; that is, people who do not have the illness.
In the end this borrowed objectivity and immunity lead him
to conclude that he is an exceptional case, that somehow the
illness that caused so many others to die will not kill him: he
will be cured. The favorable comparison is a distorted effort
to include ones illness in a nonfatal category: one patient
said, The doctor says I only have a [severe illness]. Another
literally dying patient said, Thank God, I am not as bad off
as [another patient near death].
134
PART
t h r e e : p r o b l e m s o f aw areness
Some patients try to prove they are not term inal by engaging
in strenuous activities. One patient, having been told he had
a bad heart, left the hospital and started spading up his garden
to prove the doctor wrong (and his own denial right). A nother
patient wouldnt stick to his diet. The death impending in the
present can also be denied through future-oriented talk with
the nurses. One patient began making plans to buy a chicken
farm when he left the hospital as soon as he learned he was
going to die. Communication blocks of various sorts aid denial.
Some patients simply dont hear the doctor, other refuse to
admit it, others cannot use the word cancer in any verbal
context, and still others avoid any discussion of the nature
of their illness or the inevitability of death.
As a result of the contest between staff and patient, with
one trying to open the awareness context and the other trying
to close it, a denying patient can start an accumulating process
of reciprocal isolation between himself and nurses, doctors,
family members and social workers. After disclosure, others
expect him to acknowledge his impending death, so they attempt
to relate to him on this basis. Doctors speak to him and nurses
give treatm ent on the understanding that his impending demise
can be mentioned or at least signaled. Family members and
social workers may refer to plans for his burial and his finances.
A patient who avoids the subject when he is not expected to
avoid it forces others to avoid it too, which makes them unable
to help or prepare him. One social worker said, helplessly,
about a denying patient, There was nothing I could do for
him. At this first stage of the isolation process, the patient
forces an implicit agreement between himself and others that
the topic of his terminality will not be discussed, thus instituting
a mutual pretense context. In the next stage, some of these
people may avoid all contact with the patient because he has
frustrated their efforts to help him. Nurses, doctors and social
workers tend to spend their time with patients they can help,
to prevent the feeling of helplessness that often overcomes them
while engaged in terminal care. As a result, the patient finds
himself alone, apart from receiving the necessary technical
135
care to insure painless comfort, and his isolation is complete.

Avoidance of the denying patient occurs because he refuses
to act as a dying patient, although it has been clearly pointed
out to him that this is exactly what he is. In contrast, an un
aware patient, who is not expected to act like a dying patient,
will not be avoided on these grounds. R ather he is likely to
attract others who will gather around him in silent sympathy
wishing they could help and perhaps tell him. In the end, both
the denying patient and the unaware patient may die without
preparation for death. But a denying patient still has a chance
to accept his impending death and prepare himself, usually with
the help of others, while the unaware patients chances for
preparation are mostly dependent on his doctor.
Once again, the benefits and liabilities of unawareness
(nondisclosure), as opposed to disclosure and its possibilities
for acceptance o r denial, depend on the nature of the individual
case. But on the whole, there is much to recommend giving
the patient an opportunity actively to manage his own dying.
As a strong controlling factor, staff members who interact with
the patient could consciously soften the disclosure, handle the
depression so as to encourage acceptance, and guide the patient
into active preparations for death. Perhaps they could even find
interaction strategies that would in the tug-of-war convert a
patients denial to acceptance. Yet staff members often hesitate
to tam per with a patients choice of passage to death. F or
example, one social worker said of a denying patient: Im
loath to play God on this. Unless it could serve a useful pur
pose would it really be helpful? Where a man shies away
from something maybe you should let him why make him
face this most terrible reality? T o help staff members overcome
this reluctance, readily ascertainable and unambiguous general
criteria are needed for deciding when to disclose terminality
and when to keep the patient unaware of it. Criteria that require
intimate knowledge of each patient offer no better a solution
to the doctor s dilemma than does a universally applied rule
of telling or not telling.
The Unaware Family
9
The Unaware
Family
A dying patients family creates a chronic problem for nurses.
Family members are relatively easy to handle one at a time,
but as a group they can join forces to put pressure on the
nurse to give them better visiting hours, m ore information,
special treatm ent, and so on. Nurses m ust control the family
so as to keep disturbing scenes at a minimum, and to maintain
family composure, so that family members can make crucial
decisions, visit the patient, and help with patient care, without
disrupting hospital routine.
The nurses problem is compounded, for she often has
to handle many family groups in the course of her day. In the
cancer ward, for example, a nurse may be in charge of a
num ber of long-term dying patients; she has to cope with all
their families through their long ordeal. A nd of course her
task is doubled or even tripled for any patient who has two or
three independent, sometimes hostile families (in-law groups,
two wives one divorced and their children, children s fami
lies, etc.).
A nother condition that increases this aspect of a nurses
work is the rotation of patients under her care. R otation con
fronts the nurse with new, unfamiliar families, while at the
same time the family of a previous patient may still be drifting
around the ward, asking her for information o r help whenever
they pass each other. In addition, on wards with long-term
patients, nurses become acquainted with the families of other
nurses patients, and these families come to make demands on
h er too. (A compensating feature in dealing with the families
136
137
of long-term patients is that the nurse can use her familiarity

with them as a source of control.)
Short-term dying patients generally require more constant
care, so a nurse is usually assigned fewer of these patients
and thus has fewer families to contend with. But the higher
turnover of short-term patients means that she does not get to
know their families as well and constantly has to deal with
new people. On some wards (operating room, intensive care
units, prem ature babies) nurses are well screened from the
families by doors, walls and rules and can avoid being distracted
from constant attention to their patients.
Thus, depending on the nature of the patients illness and
on the way the ward is organized, the family handling required
of nurses varies greatly throughout the hospital. M any of the
staff tactics in coping with family members are standardized;
others are of an ever-changing interactional nature, determined
in part by the awareness context surrounding the patient and
his family.
A family may be involved in a closed as well as an open
awareness context. A family involved in a closed awareness
context must collaborate with the staff to keep the patient
unaware, and staff members must handle the family members
accordingly, making sure they do not leak information to the
patient. When the family members are outside a closed aware
ness context; that is, when they are also unaware, the staff
must keep them out so as to keep the patient unaware. Of
course, family members (except young children) are seldom
outside an open awareness context; and those that are outside,
because they have not been in touch with typical medical
sources of information, are usually brought into awareness
as soon as they arrive on the scene.
In this chapter we shall consider, for the most part, the
unaware family of an unaware patient (th at is, a family outside
a closed awareness context), and the interaction tactics staff
members use to cope with problems generated by these condi
tions. In the next chapter we shall consider the aware family.
138
PART
t h r e e : pr o b l e m s o f a w areness
CO N DITIO NS F O R K EEPIN G T H E FA M ILY U N A W A R E

Several conditions encourage the staff to keep the family
unaware of the patients impending death. Although the doctor
may be fairly sure the patient is dying, he may not want to
go on record with this information in the early stages of a
disease. In long-term illness he will have plenty of time to
verify his prognosis before telling the family. Reversals, though
usually tem porary and seldom complete, do occur, and if the
doctors early prognosis seems to be contradicted, the family
may begin to question his judgment. Short-term reprieves of
even a few hours can cause an ever hopeful family to distrust
the doctor. A nd a doctor loses an element of control over the
family when it stops believing in his expertise. A distrusting
family often starts searching for new doctors often somewhat
marginal ones if not outright quacks and new cures.
Telling the family members at the beginning of a long-term
illness may also put them under stress for a longer time than
they need to prepare adequately for the patients death. Should
a reversal occur, the familys relief is balanced by a certain
amount of resentment at having been subjected to stress in
the first place. With too much advance warning, family mem
bers could prepare themselves so well that they give the
patient up before he dies, and may therefore not be sufficiently
available to help during the last stages of dying.
Putting family members under stress by telling them about
the patients true condition brings the ward staff new problems
of family control. As we shall see in the next chapter, once
family members learn the patient is dying they put pressure
on the staff for visiting privileges, more information, special
treatments, and help in preparing for their loved ones death.
The ecology of the dying situation is also disturbed. Feeling
that their circumstances justify their disdain for hospital regu
lations, the family members drift into unauthorized areas, not
content to remain in waiting rooms gleaning information as
The Unaware Family
139
best they can. En route they may disturb staff and patients by
breaking down and sobbing, asking questions and so on. Also
the staff may have to prevent family members from passing
the news on to the patient, usually unwittingly through cues,
say, of grief. And if the closed context changes to either a
suspicion or mutual pretense context, the family must be
supported and guided in the art of either not validating sus
picions or maintaining pretenses.
When there is enough time, then, these reasons for keeping
the family in the dark are likely to outweigh any pressure to
tell them the truth. Even when time is short, family members
may be kept unaware if the doctor suspects they are psycho
logically incapable of bearing up under the news. A nd though
time is usually short on an emergency ward, staff doctors will
prefer not to tell the family that a patient is dying, if they
expect a private doctor to take over, because it is considered
his prerogative to disclose or not.
In many cases, of course, especially in veterans hospitals
and state hospitals, there is no family to tell. The patient may
be without kin, or perhaps no one in his family cares enough
to visit the hospital o r m aintain sufficient contact to be reached
by the seriously ill, come at once telegram. A nurse, a fellow
patient, or a friend may stand in for family, and indeed many
people in this position feel strongly that they are the only
family the patient had before death.
CLOSED AW ARENESS AN D T H E SUSPICIOUS FA M ILY

Keeping the whole family outside a closed awareness
context creates, for nurses, the problem of managing family
members suspicions to forestall their growing realization that
the patient is dying. O f course, if the family is not suspicious,
or does not press for validation of any suspicions it might have,
then the nurse has only to cope with a relatively self-composed
group that is likely to obey visiting-hour rules, not to ask for
exceptions in treatm ent, to wait patiently in prescribed areas,
140
PART
th r ee: problem s
o f aw areness
and to be reasonably satisfied with standard statements about

the patients condition. In short, the nurse whose patients
family is both unaware and unsuspicious can rely on the
ordinary hospital procedures used for families of patients who
are not dying.
If a suspicious family is to be kept unaware, however, staff
members must allay suspicions based on either physical or
tem poral cues given by the patients condition. Since a basic
property of suspicion is doubt as to ones accuracy in judging
a cue, family members will ask the nurse or the doctor to
verify their readings of the patients condition. The staff mem
ber, in his role as an expert cue reader, thus has considerable
control over the credence that family members will give their
own suspicions. And since families would rather hope than
know the worst, staff efforts to keep the family unaware often
succeed, unless the patients condition is clearly bad.
Nurses have many available interaction tactics to forestall
growing awareness. When several family members begin to
suspect the worst, each suspicious member may hide his sus
picions from the others, because each believes he is the only
one who thinks the patient may die; not yet certain of his
belief, he wishes to spare the others. In hiding their suspicions,
they mutually sustain their faith in the recovery of the patient;
thus all of them remain outside the closed awareness context.1
No one wants to lose his own faith (as one nurse said, I think
she was afraid we might tell her. ), or jeopardize the faith of
other family members by asking nurses or doctors the questions
1 This may be termed a misperceived pretense awareness context
within the family unit. Unlike the mutual pretense context in which
all know and share a truth but pretend not to, in the misperceived
context each thinks he knows the truth but that the other does not,
and so he pretends he does not know the truth either. The misperception
on each persons part is that he does not share the truth with others^
whereas if he dared to check he would find out he does. Then, they
might find out either that it is not necessary to pretend they do not
know the truth, or that it is necessary to engage in mutual pretense:
For a good example of a misperceived pretense awareness contend
whose force generates a subculture, see David Matza, Delinquency and
D rift (New York: John Wiley and Sons, 1964), Chapter 2.
The Unaware Family
141
really necessary to verify suspicion. Staff members can take

advantage of this situation simply by staying out of family
interaction, or, if called upon, by lending their weight to the
familys pretenses which support their faith in recovery.
A variation of this procedure is to m ake one family mem
ber aware of the patients true condition and let him help keep
the other members unaware. Usually the strongest member
of the family is told even if he is not the closest relative or
must even be brought in from out of town. A relatively de
tached cousin would be preferred over a m other or a wife who
is on the verge of complete breakdown. A grown son may be
told rather than the wife of a dying man. A father of an unex
pectedly dying prem ature infant is usually told first, and he tells
his wife when, in his judgment, she can take it. When the
strong m em ber is told, it becomes his responsibility to decide
whether others m the family and which others should also be
made aware. If he decides not to tell anyone else, his burden is
doubled, for not only is he unable to share his grief with other
family members but also he must collaborate with the staff
to allay their suspicions. A t a time of great sadness he is forced
to appear relatively cheerful.
In larger families, a group aware of the patients impending
death may form, its members supporting each other to keep
the others unaware. Again, the family takes care of itself;
under these circumstances, the staff can rely on the family
members who are in on the secret to control those left out.
These are but two varieties of the general tactic letting fam ily
members control each other which is used in many other
situations, as we shall have occasions to note.
Other tactics, involving more staff intervention, are not as
easy to apply. An event that typically raises the suspicions of
even the most hopeful family members is a sudden change in
the patients location to a private room, nearer the nursing
station, to a pre-operation ward, or (worse still) to an intensive
care unit. The staff is immediately asked why? And if the family
is still to be kept unaware, the nurse must counter a suspicion
142
PART
that m ay be very close to certainty. She immediately ushers

the family into a room and shuts the door to insure a private
discussion. Invariably, her first explanation is that there is no
cause for concern, the move to the new location is only to
provide the patient with better care. But this may not reduce
suspicions, because patients are put in single rooms to die,
are taken to the intensive care units for access to special
equipment with which to forestall a possibly impending demise,
are put near the nursing station so as to be under constant
surveillance by all nurses on the floor, and are taken to the
operating room for a last effort to prevent death. So additional
tactics are needed.
The next one is to give the family all available favorable
information on the patients condition and to suggest that his
changed location indicates nothing so drastic as dying, or that,
with the information available, nobody could possibly be sure
that he is dying. Then, both to discourage a recurrence of any
suspicious, m orbid thoughts and to get the family out of the
way and avoid a scene, the nurse suggests that they go home
and rest or go out for food, promising to notify them if the
patients condition should change. Sometimes she assures them
that anyone who becomes anxious again may call her at any
time for more information. The nurses strategy, in other
words, is to reduce suspicion with expert information and then
try to prevent its recurrence by making additional expert infor
mation readily accessible to the family. In this way the family
continues to depend on the nurse as a readily trusted sourcfe
of expert news, and the nurse in turn maintains control over
their awareness.
Now and then a patient who is expected to recover will
turn bad while relatives are present! This requires fast action
on the nurses part, to keep the family unaware at the same
time they are responding to the striking change in the patients
condition. H er best strategy is to show concern in her words
and facial expressions without expressing undue alarm. If
she fails, thereby revealing the gravity of the situation, the
The Unaware Family
143
relatives may become aware so suddenly that panic is highly

probable. In this situation, then, the nurses main concern is
to prevent an emotional upheaval and all the repercussions
that could follow. The family can be eased into awareness
later, if the patient is surely dying.
A standard tactic in norm al situations, role switching, has
implications for the whole staff. To handle questions and probes
in a way that will not force her to bring the family into aware
ness, a nurse may first counter a request for information with
I dont know or Im not allowed to say. The family mem
ber is then switched to a person at the next higher level of
authority, who may be empowered to state the patients true
condition. This staff members role vis-a-vis the family member
may be more appropriate for dispensing information. In ex
treme cases the family mem ber is referred from an aide up
through various levels of the nursing staff, with perhaps a few
side trips to an orderly, social worker, nun, chaplain, or ward
clerk, and ends by asking the doctor, the very person who still
feels it is best to keep the family out.
T o be sure, role-switching is stimulated in part by the
fact that a nurse has no right to disclose a patients terminality
to a relative unless the doctor explicitly permits it. Professional
ethics thus become involved in keeping the family unaware, and
for nurses this sometimes creates a dilemma, for they may have
no other reason to keep the family out and many good reasons
to bring it in. Only when the family is aware, for example,
can its members help with the intense, constant physical and
emotional care a dying patient requires, or understand new
or complicated treatments, or begin to prepare for the patients
death.
In one case, neither a father nor his son knew that the
father would die soon. The attending nurse felt strongly that
both should know, so the father could give the son some last
words of encouragement and support and bequeath the many
activities that would soon be the sons responsibility. Yet, be
cause doctors orders forbade disclosure, she was forced against
144
PART
t h r e e : pr o b l e m s o f aw areness
her own feelings seemingly to approve the meaningless talk

between the pair. Although she had the power to keep the
family members out, she could not bring them in even to help
the patient. H er helplessness had to be muffled in her inter
action with the patient and family member. In this situation,
as in many others, nurses would often find it much easier to
be honest with the family. It would help them maintain their
own composure and better achieve the com fort goals of
nursing care.
A nother common situation that makes it possible to control
the familys awareness occurs when family members demand
information on the patients condition. Perhaps one of the
crudest techniques, certainly one th at does not work well if
family suspicion is running high, is to hem and haw with
vague or very general statements. The very worried relative will
then probe for the nurses meaning of what she is trying to say;
and at this point she must use one of the many escape strate
gies in her repertoire. She can invoke some highly legitimate
and compelling reason for being elsewhere, for example: An
other patient needs m e, o r The doctor is expecting me right
back, and so on. In many situations the nurses can simply
evade contact with questioning relatives, without even rejecting
them too blatantly. Nurses have back rooms to hide in, and a
legitimate running gait with which to breeze past family mem
bers. W ards like the intensive care unit or prem ature baby ward
have doors to be kept shut, and glass and screens to hide
behind; requirements for sterility can be used to prevent
contacts; and a ward clerk can answer the phone and give a
pat statement on the patients condition no change today,
ate well or doing fine, and so on.
In desperation for information, family members may even
turn to other patients who seem to have been around long
enough to know what is going on. Of course, the doctor cannot
hold these other patients responsible for keeping the awareness
context closed. But often, especially if they have been helping
in the care of the dying, they can be depended upon to collabo
The Unaware Family
145
rate in keeping the secret from the patient and will be reluctant
to reveal it to his family. (In their search for information on
the patients status, relatives may even accost the sociologist
doing field work, since he seems nice enough to approach, looks
intelligent, and is hanging around seemingly not doing much of
anything. If he does respond, the staff may use him to absorb
the relatives anxiety, knowing that he will not jeopardize his
research position on the ward by leaking closed inform ation.)
CONDITIONS FO R M AKING T H E FA M IL Y AW ARE

The previous discussion illustrates that, at best, staff success
in keeping the suspicious family unaware is very tenuous. In
the ever-developing interaction between staff and family, it
is only a m atter of time before suspicions are confirmed by
the patients obvious condition, and by the familys increasing
skill in piecing together information and trapping or testing
the staff (this is a contest, just as in C hapter 4 ). The tactics
we have described, then, lose their meaning and fade into other
patterns of interaction based on conditions for making the
family aware.
Even before the patients m oribund status is obvious, other
conditions may initiate the transition to confirmed awareness
among family members. W hen staff members are certain that
the patient will soon die, they feel that someone must be
told, that at least one family mem ber has the right to know.
If none seems strong enough, then a close friend is selected,
who discusses with the doctor when various family members
should be told. A t the other extreme, when disclosure is direct,
the patient is told first, and he tells the family. The general
procedure, however, is to disclose first to the closest or most
intimate family member deemed strong enough to take the
blow. This person sooner or later tells others, and the news
spreads through the family.
A nother factor encouraging disclosure is that family mem
146
PART
bers can help with terminal care, and thus facilitate the ordeal
associated with dying. Linked with this factor is an im portant
one we have already mentioned disclosing to family members
gives them time to prepare themselves for the patients death.
Hopefully, this preparation also renders family members com
posed enough to help during the final ordeal.
Immediate contingencies, especially for short-term and
emergency dying patients, often make it essential to tell family
members quickly, so that they can authorize a treatm ent or an
operation, take leave of the patient, and make social, financial,
and emotional preparations. Sometimes a relative m ust m ake
a lengthy journey if he is to see the patient before he dies. Under
these emergency conditions, relatives may feel cheated if they
are not told promptly enough to give them a chance to partici
pate in the patients last days o r hours. Participation may not
only be a personal need of a family member, but also may be
expected of him by friends and other relatives. A general norm
in American culture is that no one should die alone; preferably,
death should be attended by a close relative.2 Telling the family
too late may m ean that the patients life may have to be unduly
prolonged while awaiting the arrival of a relative. On the other
hand, too sudden a disclosure may leave the family ill prepared
to cope with these last-minute activities, and results can be
traum atic indeed. O ther factors that affect disclosure to the
family are its ethnic status, language, religion, and education,
as well as the degree of rapport with the doctor. These are
among the criteria that a doctor uses to judge whether the
members want, and can take the news, and whether he will
be able to make them understand the nature of the patients
illness.3
See Robert Fulton, Death and Self, Journal o f Religion and
Health. 4 (July 1964), p. 364, for properties of this norm.
T he relationship of social factors such as ethnic status, social class,
language, religion, education to properties of disclosure to the family
(particularly if, when, and how) is an important research problem. Th$
research should also develop the intervening interaction process that
links the relationship of a social factor to a kind of disclosure.
The Unaware Family
147
TRANSITIONS TO AW ARENESS
Staff members use a great many special tactics to ease the
familys transition to awareness. M ost of these tactics are rela
tively purposeful variations on the general procedure for
disclosing terminality to the family.
The customary assumption th at someone m ust be told
has resulted, in some hospitals, in formal procedures for gradual
disclosures to families. A t some point in the patients progress
toward death this point varies, but typically it occurs when
uncertainty becomes firm certainty he is put on a critically or
seriously ill list. This action, usually ordered by the doctor,
sets in motion many activities designed to bring at least one
family mem ber into awareness. If none are on the scene, a wire
is automatically sent to the closest kin: Y our (wife) has been
put on the seriously ill list, please come at once. O r a phone
call is made: Y oud better come over here, your little boy
is injured. These communications, while short of full disclosure
of terminality, create suspicion and prepare a family member
for full disclosure as soon as he reaches the hospital. Once on
the ward the family m em ber is ushered into the privacy of a
room or office and is gently told by the doctor, Unless he
takes a turn for the better, Im afraid he wont pull through.
The doctor then answers questions, while other staff members
come to complete the process of making the relative aware of
the impending death. A nurse, social worker, nun, clerk or chap
lain may take over the last stages of the transition in an effort
to avert a possible breakdown or a scene.
The same process of disclosure is also followed when the
expectations of an aware family mem ber must be altered in a
direction closer to death; for example, from uncertain and
time unknown to certain and time imminent. When
the hospital has no formal procedures, the doctor may initiate
the process on his own, or a nurse may initiate it by reminding
the doctor that the patient has reached a stage that warrants
148
making the family aware or changing their death expectations.

For family members who are continually asking for information,
this disclosure process need only be made part of answering
their latest questions by telling them the new diagnosis. F or their
part, some family members simply refuse to believe the patient
will die, no m atter what any of those doctors say.
Gentle disclosure. When the patients condition, especially
the rate at which he is dying, permits it, gentle disclosure to
the family is preferred. Gentleness may be accomplished by
delegating disclosure to the nurse; although the impact of the
doctors authoritative presence is then absent, disclosure is
softened by the womans touch. One nurse reported: Some
times they do not realize I gently try to say H es not im
proving and then H es failing a bit. She was trying to coax
the family into awareness. A nother nurse waits until the family
seems ready before disclosing: I let the lead come from them
[parents] I have some personal feelings that some people
arent ready to talk about this, that people have to work some
thing through before they can talk about it or are even ready
to talk.
The nurse may be a bit more brusque if she has been
trapped into bearing sad tidings by a doctor who wants to
evade the task by delegating it to her. The doctor in th il
situation tells the family that they are not sure yet and to
check back with the nurse who will be authorized to give the
diagnosis. The nurse, in turn, may express her resentment of
this strategy by dealing rather harshly with the patients family;
In this case the doctors higher status permits him to saddle
the nurse with the dirty work, even though ethics call for
disclosure to come from the higher status person.4
Doctors too often make disclosure gentle by saying, W ere
4 For properties of dirty work, see Everett C. Hughes, Men and
Their Work (New York: Free Press of Glencoe, 1958), pp. 49-52,
70-72, 122, 137. Hughes sees dirty work as a task of a professional
that is physically, socially and morally beneath his dignity. White
disclosing to a family is not beneath the doctors dignity, it is simply
a difficult task that his superior status may allow him to delegate
nurses.
The Unaware Family
149
doing all we can, or Its out of our hands now, and then
outlining what will come. This briefing tactic can be usefully
interwoven with the known tendency of family members to
take care of each other. In one case a doctor faced seven family
members, including the patients wife, who was on the verge
of collapse. The doctor first briefed the whole group together,
then took the wife to the patients bedside. When she was ready
to leave the patient, obviously close to breaking under the
strain, the doctor returned her to the waiting family, who im
mediately gathered around her in close collective support. At
this point the doctor disappeared, signaling the nurses to stay
away. Thus the weakest family mem ber was brought into
awareness, without imposing on the staff the burden of picking
up the pieces. In addition, the doctor could phrase his briefing
in fairly technical language because the family provided its own
emotional support and control for the wife. H ad the wife
been alone this would have been too harsh an approach. The
presence of stronger family members protected the doctor by
permitting him to keep the wife at a safe distance, emotionally.
The social distance imposed by the use of technical language
made it doubly appropriate for the family to gather round and
help the wife; it also released the doctor from the need to
express much sympathy, so that he could better maintain his
own composure during the strain of instigating the transition
to awareness.
A prime means of softening disclosure to the family is to
leave out the tem poral dimension of the impending death, which
makes the event itself appear less certain than it is. By saying,
Its out of our hands now, the doctor implies that he is
helpless, unable to stop the illness or to ascertain when death
will come. As death approaches, and it becomes appropriate
to reveal the time it is expected to occur, a doctor may either
be very vague It will be a short time or abandon his
effort to soften the blow, making his announcement precise
enough to enable the family to pace its preparations more cor
rectly, if not comfortably.
Gentle disclosure can have negative effects on both family
150
part t h r e e : pr o b lem s o f
AWARENESS
and staff; these must always be weighed against the possible

positive effects of a harsher disclosure. A gentle disclosure m ay
be too weak to stimulate adequate family preparation. Leaving
out the tem poral dimension in disclosure also risks lack of
preparation by the family, for family members invariably give
their loved one m ore time than he actually has.5 If they are
not completely ready for the death they may miss the oppor
tunity to participate in the patients last hours. In addition,
unprepared family members are still prone to surprises and
shock. Thus they are more likely to trouble the staff at the
time of death, with breakdowns that might have been forestalled;
They are also less able to decide upon and attend to such
m atters as autopsy, disposal of the body, dispersing the pa*
tients personal effects and meeting his obligations to the
hospital.
Harsher disclosure. Accident, suicide attem pts and suddefll
onset of critical illness may require a relatively abrupt, sur
prising disclosure to the family, if death is very likely to occur
immediately. The harshness with which a surprise disclosure
is made depends both on how near the patient is to death, and
on the familys physical proximity to the patient. W hen the
family is not on the scene, the phone call or telegram to
come quickly will create an expectation strong enough to pre
pare them to accept confirmation of their suspicions on arrival;;
The suspicion preceding full disclosure ameliorates the shock
of suddenly learning that a mem ber of the family is dyings
Sometimes suspicion is raised by an event such as the infamous,
church bombing in Birmingham for one Negro m other whose
daughter was killed in the explosion, the announcem ent of
death merely confirmed a realization that occurred as soon as
she heard the bom b explode. The family on the scene whose
relative has yet a day to live may also be eased quickly into
5 In contrast, the family tends to give a relative in the recovery
status less time than it takes to get well. See Fred Davis, Passage
Through Crisis (Indianapolis: Bobbs-Merrill, 1963), Chapter 4.
The Unaware Family
151
awareness this way, with a suspicion announcem ent to be con

firmed later in the same day.
Often, however, the patients condition changes so fast
that there is simply no time to prepare family members in
advance or to indulge in the comparative luxury of a gentle
disclosure. A non-terminal patient may suddenly turn bad
while a family m em ber is visiting, or a patients heart may stop
beating while his family is waiting for him to come out of the
operating room. In these cases family members are practically
present at a death for which they are totally unprepared. Some
one m ust tell them, and if a doctor is not present as is often
the case when the patient was not expected to become terminal
so suddenly a nurse may disclose without doctors orders.
One nurse told a son and his wife during a visit, Y our m other
is barely breathing. I dont expect her to last the hour a
direct disclosure with both dimensions of the death expectation
put clearly and succinctly. The nurse realized that she had no
time for a preliminary suspicion announcement; she had to give
the family a chance to prepare and participate as soon as they
could collect themselves. Aware of her breach of norm al ethics,
she said, I had to tell them. You couldnt say please step
outside. I had to tell them .
Because they prefer to announce an impending death, how
ever rapidly it may be approaching, rather than an unexpected
death, which may occur while the family steps outside, nurses
may feel forced to disclose dying on the spot in an abrupt
manner. If a nurse must announce a death which she must
if the family faces an obviously dead patient even before the
patient is officially pronounced dead by a doctor, the strain on
both nurse and family is great, and a scene is almost sure to
follow. An announcem ent of this kind is also a more serious
breach of professional announcem ent ethics than an unauthor
ized statem ent that the patient is dying. Quick action is often
essential, too, for if the nurse, in asking the family to step
outside, permits the family members to suspect that death will
be immediate, she m ay lose control over their behavior. In
152
PART
t h r e e: pro blem s
of
aw areness
one case, a relative, asked to step outside, became suspicious

and ran back into the room to find the patient dead; she grabbed
his hand and wouldnt let go. In another case the relative tried
to revive the patient by rubbing his cheek. In both cases the
relatives had to be removed from the room by the staff. In
short, it is generally much less disruptive for all concerned if
the nurse can announce unexpected, sudden dying instead of
unexpected, sudden death.
The shock for the family is most harsh, of course, when
death is so sudden that no disclosure of immediately impending
death is possible. In this limiting case the notion of a non
awareness context applies to staff, patient, and family; all normal
sentiments surrounding the patient are abruptly shattered.
Everyone involved is surprised, but of course the family is
hardest hit. (Victims of sudden heart attacks typically die with
a look of complete surprise on their faces. A doctor called in for
such an emergency reported that this surprised look was mixed
with hatred for the regular physician who had, only a few days
before, pronounced him in perfect condition.)
Staff members too can be so shaken by the surprise that
they make untoward gestures and indelicate announcements!
In one case friends had surrounded the wife of a stricken man
to protect her from unduly blunt disclosure, but the doctor in
desperation and helplessness said loudly, There is nothing I
can do. The m an will die in a few m oments. The friends judged
that the impact on the wife was harsher than it would have;
been had the disclosure been less abrupt.
In an emergency ward case, just after a surprise death,!
staff members discovered the patients m other in the waiting!
room. The doctor was so involved in the technical aspects at!
the death, since its details were somewhat threatening to hil
self-confidence and medical competence, that he simply went
over to the m other, told her the sons fate, and walked off in a
state of preoccupation. The public nature of this unexpected
announcement was devastating to the mother. Other people
in the room, including staff members, were highly indignant atj
the absolute cruelty of the doctor. Finally realizing what he
The Unaware Family
j 53
had done, the doctor took the woman into a private room to
smooth over the situation.
Disclosure space. As our description of the disclosure pro
cedure suggests, management of space is a strategic factor in
bringing families into awareness. Ideally it occurs in the privacy
of a closed room. A t the opposite extreme, space is disregarded,
as in the case just described, and disclosure occurs through a
blunt surprise announcement made in a public place filled with
people. The less urgent the situation for instance, uncertainty,
with time unknown the more likely a doctor is to disclose in
the hallway instead of his office. Even in the hallway, privacy
varies some places are virtually private and others are full of
heavy traffic. In a closed context, the patients room is generally
taboo for disclosure to the family, though the ends of larger
ward rooms or rooms of comatose patients may not be.
Variations in the effort to manage space depend on the
amount of time the doctor has available (as determined by the
rate at which the patient is dying as well as the doctors work
load), and his relations with both the patient and his family.
The spatial provisions the hospital has made for family mem
bers are also im portant. Some wards have their own waiting
rooms. Some hospitals provide only a central waiting room on
the ground floor, and others provide no facilities at all except
perhaps a chair in the hallway. These various places may be
used for less urgent disclosures that seem to require less than
complete privacy. Invariably, however, all hospitals have near
at hand a room or office where disclosure can be made privately.
N o disclosure. Under certain conditions, the preferred
strategy, o r non-strategy, is simply not to make an explicit
disclosure. When the doctor thinks that terminality is obvious,
in view of the patients visibly moribund condition, he feels
that he need not initiate the disclosure process, expecting family
members to realize what is happening, and to seek help from
a chaplain, nurse o r social worker for their preparations, or
make an appointm ent with him if they need information. While
154
PART
t h r e e : pr o b lem s o f
a w areness
in m any cases his observation is correct, in others it is an

avoidance strategy, when the doctor is not up to announcing
an impending death. Doctors who do not like the deathsentence responsibility tend to force everyone to operate on
a suspicion level until the patients condition becomes obvious.
A n im portant aspect of the physical arrangement of wards
is the degree of visibility the family, not actually at his bedside,
has of the patient. The physical arrangement of a ward which
allows high visibility may encourage a doctor to conclude that
the patients status is obvious to family members. F or example,
in m any emergency wards, family members need only look
and listen through a door usually open beyond which staff
members are frantically working on the patient, and they can
confirm any suspicion they might have that the patient is dying.
Some emergency wards have bedrooms, treatm ent rooms, and
operating rooms arranged around a nursing station, with the
waiting area for families nearby. Family members cannot help
observing their relative as he is hustled from room to room,
and usually this permits them to get a good idea of his chances
or even to see his death. A nd w hat is not seen is easily
imaginable, or can be filled in by observing other visible
patients who are in a similar condition.
O n most types of hospital wards, however, the waiting
family is completely segregated from the patient and from the
areas in which treatm ent occurs. Under these circumstances,
family members know only what they are told until they are
perm itted to see the patient. The patients condition may by
then be glaringly obvious. However, in this latter instance the
staff has a chance to prepare the family and thus to avoid
possible disruptive consequences of not disclosing.6
* In this chapter we did not consider what happens when the patient
is aware and the family is not. If this occurs in the open context then
the family members will be readily brought in except for those mem
bers who could obviously not take it. A more interesting situation,
which bears research, is when the patient is aware and the context is
closedthe staff do not know he is aware and the family is unaware.
The burden is on the patient to pretend to family members that all is
going all right. This happens in heart cases when a patient knows he
could go any time, but does not tell his family that he even is a
heart case.
10
The Aware
Family
No clear line divides the family in transition to awareness from
the fully aware family, but staff members can see the difference,
and they change their tactics accordingly. Once a family is
aware that the patient is dying, managing that family takes on
new dimensions. Perhaps the most critical change is in its
visiting patterns. The usual relaxation of visiting restrictions for
an aware family creates for the staff new problems associated
with maintaining family privacy, helping family members main
tain their composure and prepare for the patients death, and
also coping with their intensified need for information.
VISITING
Relaxation of Visiting Rules. When they learn that the
patient is dying, relatives usually increase their visits, coming
more often and in larger numbers, and more of them visit on
private schedules. (Sometimes, of course, particularly for the
old, neglected patients in state hospitals, family members stop
coming altogether after they become aware of the impending
death; this may also happen sometimes to young patients who
are estranged from their families.) If the awareness context
is kept closed to the patient, these increased visits can become
an obvious cue that must be m anaged by both staff and family
to reduce the patients suspicion of his true condition. O ther
wise the increased visits may inadvertently transform the aware
ness context to an open one or to one of mutual pretense.
155
156
PART
The Aware Family
157
To accommodate these increased visits, hospital and ward

rules are relaxed. Family members are allowed to visit patients
at almost any time, and one member may be permitted to stay
around the clock. In the case of one woman, as soon as
the physician informed her family, the patient became the
center of attention of several sisters and brothers, her own
children and many grandchildren, nieces and nephews, all of
whom crowded the small TV room every day, staying long
beyond visiting hours, and hovering about her with affectionate
attentions. 1 The family expects the rules to be relaxed or
waived, under the circumstances, and the nurse who does
not relax visiting rules for at least some of the relatives risks
a family protest that may be carried to top administrative levels
for a decision. In contrast, the obstetrics ward of one hospital,
which of course is not used to having dying patients, does not
alter its visiting rules for the family of a woman who is posted
as terminal. The work of the ward is geared only to the routine
of taking care of healthy women and their babies.
Obviously this influx of family members creates many
management problems for the staff. One problem is to see
that the patient has enough rest. Sheer numbers of people
can sap the patients energy, and even when the nurse tries to
limit unrestrained visiting to the intimate kin, there may still
be too many people and visits. Another difficulty that inter
feres with rest is that some families bother the patient or
pick, pick, pick just as they are used to doing at home. But
since the nurse relaxes visiting rules at her own discretion, she
can control a troublesome visitor by reinstituting the regula
tions for a good cause, and send him back to the waiting room
or even out of the hospital.
The patients domestic problems may follow him to his
deathbed in another form: especially when visiting rules are
relaxed, independent sets of family members may visit simul
taneously. Hostile interaction at his bedside, between his first
and second wives, is hardly restful to a dying patient.2 In one

case an embarrassed nurse had to ask one wife and her children
to leave the room and wait their turn. But in other cases,
independent family groups, who might be unwilling to tolerate
each others presence under ordinary circumstances, join forces
in this grave situation to make the patients last weeks or days
as comfortable as possible.
1 Rose Laub Coser, Life in the Ward (East Lansing: Michigan State
University Press, 1962), p. 59.
This is a problem on the increase for nurses with the increase

both of dying in hospitals and of the divorce rate.
W eekend Leave-Taking. A nother problem for the nurse,

made more acute by the relaxed regulations, is that of the
repeated weekend farewell. Aware family members who live out
of town sometimes come every weekend during the last weeks,
thinking each time that they may be taking leave of the patient
forever. If the patient is not aware that he is dying, the nurse
must try to allay the suspicion generated by all these tearful
farewells. Ultimately this weekly scene may create a context
of mutual pretense awareness between patient and staff, which
the staff initiates, and which is potentially harmful to the pa
tients faith in the staff. And if the family lives a hundred miles
away or more, the nurse cannot assure them that they will
be notified in time if the patient takes a turn for the worse.
They must come each weekend, not knowing which will be
the last. In some hospitals these problems are maximized be
cause a high ratio of terminal patients live out of town.
Attention Patterns. Stable family relations are built on many
standard patterns which are broken only for special reasons.
One such pattern attention given and expected is very often
broken by aware families. Relatives not only visit more often
when they learn that the patient is dying, but they bring flowers
and candy; they kiss and reminisce, and so on. To the patient,
these are unexpected, perhaps unprecedented, attentions. If
he is to be kept unaware (often at the familys request), the
nurse m ust control these revealing gestures. She can sometimes
158
PART
invoke rules to prevent some attentions the patient must not

be touched, relatives are not allowed on the patients bed
unless a staff mem ber is present, flowers are not allowed except
in private rooms, etc. She can also try to explain away the
unprecedented attentions when the patient becomes suspicious.
Family members also must explain away unexpected attentions,
even so simple a one as a visit from a relative who ordinarily
wouldnt cross the street to see the patient unless something
was drastically amiss.
A n aware family in the closed awareness context has many
opportunities to disclose unwittingly to the patient. As the
interaction continues, staff and family efforts to keep him at
least only suspicious m ust eventually fail, altering the awareness
context from closed or suspicion to m utual pretense, or causing
it to vacillate between these states. F o r example: The sister
told me that on Thursday night before Alice died, all her family
was sitting around her bedside and she was sleeping or pretend
ing to be asleep, when suddenly she opened her eyes and in
clear and angry tones said, W hat is this, a death watch? *
Since no one made an explicit disclosure in this situation, no
one ever knew whether the patient died in a state of suspicion,
or pretense. Consequently, all interaction between family and
patient had been designed to reduce suspicion and m aintain
the pretense.4
Status-Forcing Scenes. W hen nurses relax the visiting rules
they raise the risk of having to face status-forcing scenes.5
Giving the family relatively free access to the patient means
that family members are always around to observe and judge
3 Florence Joseph, Transference and Counter Transference in the
Case of a Dying Patient, Psychoanalysis, and the Psychoanalytic
Interview (Winter, 1962).
This is a good example of how an open awareness context would
have saved family and patient much wasted effort, and would have
allowed them to participate as a total family in preparing themselves
for the death.
5 On status forcing see Anselm Strauss, Mirrors and Masks (New
York: Free Press of Glencoe, 1959), pp. 76-84.
The Aware Family
159
the patients care. All too often this untutored concern is

focused on trivia, or it takes the form of demands for undue
heroics. Relatives try to force staff members to m ake these
unnecessary or extra efforts by reminding them of their obliga
tions as medical-care experts; they force them to enact their
official roles, perhaps with threats or reprisals if they do not.
In the status-forcing scene which ensues, the nurse must
adopt tactics which vary by how much she can satisfy the family
mem ber without compromising her own position or harming
the patient. If, for example, the treatm ent begged or demanded
is not feasible, the nurse can give the family m em ber an
expert or technical reason, using a slightly rough m anner to
remind the relative that knowledgeable discipline is required
in such a serious situation. Hopefully, the family m em ber will
submit to these technical exigencies; apart from being cowed
by the nurses brusque manner, believing that the restrictions
are technical helps him to limit and control his own expecta
tions. But if the family mem ber does not subside, he can be
insulated from staff by being put in a waiting room filled
with people (in essence he is drowned out of interaction with
staff) and told to await information on changes in the patients
condition.
When relatives beg for impossibly heroic measures, nurses
make every effort to accompany their reasonable refusal
with comforting measures such as sedation or a shoulder to
cry on. One nurse told us that a woman who begged for
heroics to save her husband, rushed into my arms, crumpled
on my chest, buried her face between my two little breasts,
crying and wailing. I held her and let her cry. Meanwhile
another nurse said, Do we have any liquid amytol to make
her drink, to calm her down? On the other hand, when the
patients case is hopeless, the family may ask the staff to stop
treatments that result in the useless prolonging of life so
painful to all concerned.
Family members use status-forcing scenes also to motivate
the staff to do everything possible, to the very end. As patients
160
PART
of
aw areness
near death it often becomes evident that certain treatments are

superfluous, and staff members may slack off a little. A family
member who witnesses this may panic, accusing the staff of
negligence and promising that if a stone is left unturned,
someone will pay. This forces the staff to continue the super
fluous treatments, which can be very frustrating because they
accomplish nothing. To overcome this frustration, nurses often
switch their motivated involvement to the family, thinking of
the treatment for the patient as if it were actually for the
distraught relative. We have term ed this tactic object switch
ing. It is a general tactic that nurses used to cope with their
helplessness in giving superfluous or meaningless care to a
dying (o r any other kind) patient. In switching their involve
ment to another party to the situation, they feel they are
actually working for this person and not for the patient. Switch
ing involvement to another person both motivates the nurse
to go on working and starts a new relationship with the person
switched to, which has interactional consequences. We shall
have occasion to mention this tactic again.6
If staff members are not careful, status-forcing scenes may
produce conflict within the staff, as when a relative tries to
enlist one staff member on his side in an argument with another.
A relative may express various kinds of dissatisfaction with a
nurse, for instance, to get the doctor to order superfluous
treatment, and the doctor may very well comply, perhaps
because he is concerned about his relations with the family
(a problem common to private physicians), or perhaps
because he believes the family mem ber is right. In one case,
a doctor who was interested in little items of care saw to
it that the nurses turned the patient regularly, at least while
he and the nurses were under pressure from the family. The
clinical purpose of this requirem ent was unknown to the nurses,
but they did it usually just before the relatives arrived.
6 Object switching is a general strategy in all work that people
use to keep up motivation for a task, the object of which they either
have no involvement in (e.g., assembly line work) or loo much (e.g.,
dying).
The Aware Family
jg j
Normally, of course, nurses and doctors cooperate to keep

the family happy, providing as much superfluous care as is
harmless and not too time consuming. This collusion is to a
certain extent a response to the familys power over the staff: 7
an irate relative can compromise the whole staff by seeking
a doctor who will do something! sometimes the best doc
tor at a university medical center, but also sometimes a
marginal doctor, or an outright quack. Thus the power over
the patients care that staff members derive from their superior
medical knowledge is partially offset by the familys power to
search for another doctor or hospital. If present care does not
meet expectations, this search becomes more urgent and less
discriminating. This balance of forces between family and staff
is implicit in every status-forcing scene.
Privacy versus Care. As long as visiting regulations are
enforced, it is not difficult for nurses to respect the general
social norm that family privacy should not be violated.8 When
the family is allowed only certain hours with the patient, staff
members can care for the patient freely during the remainder
of the day without worrying about whether they will interrupt
a family gathering. But when family members are constantly
present at the dying patients bedside, the nurse must resort
to interaction tactics. H er daily round takes her in and out
of the patients room periodically. When visiting rules have
been waived, she must consider each time whether a family
member is there, whether he requires privacy and whether her
task overrules the demands of privacy. If her task cannot wait,
then the relevant question is whether she can carry it out in
front of a family member without grossly violating the normal
7 The collusion plays of one team as a coping response to the power
of another is intimated in Erving Goffman, Presentation of Self in
Everyday Life (New York: Doubleday Anchor Books, 1959).
8 On the societal structure of privacy see: Alan P. Bates, Privacy
A Useful Concept?, Social Forces, Vol. 42 (May 1964). See also
Arthur Stinchcombe, Institutions of Privacy in the Determination of
Police Administration, American Journal of Sociology, 69 (September
1963), pp. 150-160.
162
PART
t h r e e : p r o b l e m s o f a w areness
privacy expectations. Such tasks as feeding, taking blood pres

sure, m arking charts, and giving arm shots can be accomplished
in the presence of others, since these tasks at most merely
suspend private conversation for a short time.
Certain nursing tasks cannot ordinarily be perform ed in
front of family members however: those that require body ex
posure and manipulation for medication, bathing, changing
bandages and the like. In Am erican society a stranger, no
m atter how legitimate his purpose, cannot usually expose or
touch certain areas of an adults body in the presence of an
intimate of the person, for this is considered an intrusion into
their private relationship.9
Family privacy is often disturbed by the nurses need to
care for other patients in the same room. A way to handle
this problem so that the family will not be disturbed, nor the
nurses routine disrupted, is to provide the family a room or
section thereof, an office, an isolated part of the hall, or simply
a curtain around the patients bed, where family members
can feel secure from interruptions not directly involved in
their relatives care.10 Nurses assure the visiting family that
this region is entirely given over to their private contacts with
the patient; by the same token, they assure themselves freedom
from having to defer to ever-present relatives.
M anagement o f Space. The private-region solution to the
dilemma of privacy versus care depends, of course, on the
availability of sufficient space. W hen family members are
allowed to visit constantly, the general problem of distributing
An American wife returning from France with her husband, who
had to have an operation commented on this norm: As was expected
of me in the United States, I had gone out during the examination.
(This examination was on the lower right side of the husbands body.)
Lael T. Wertenbaker, Death o f a Man (New York: Random House,
1951), p. 37. This norm of privacy is common knowledge to all hus
bands who have been with their wife in labor. Husbands are asked to
leave the labor room when the doctor performs a rectal examination
of the dilation of the cervix.
10 In Asian hospital wards where there are up to 40-60 or more
beds, a screen is usual w ith.other patients therefore cut out from the
scene.
The Aware Family
163
space, always present in hospitals, becomes critical. Increased

visits from an entire family can inundate a waiting room or the
patients room. Status-forcing scenes absorb halls and waiting
rooms, since these scenes tend to drive other people away, and
a patients room becomes too small when two or more inde
pendent family units are present. A t one hospital, a gypsy
family with its twenty-plus members overflowing everywhere
caused confusion throughout the ward. To maintain order
in the ward, and especially to protect the space around the
patient, family members m ust be carefully distributed throughout
the ward, or even the hospital.
Certain criteria emerge for distributing family members.
The members who obviously cannot be helpful, who are on the
verge of collapse, or who are likely to create a scene, may be
insulated from the staff and patient by being directed to offices,
to waiting rooms or treatm ent room s on the ward or to the
hospital lobby. In cases like this, visiting rules may be rein
voked with the implication that they will be relaxed again later.
A nother criterion for distributing the family throughout the
hospital is the visitors relation to the patient. To prevent in
undation of the patients room by many family members, the
nurse may restrict close contact with the patient for any length
of time to the closest relatives (usually children, parents, or
spouse). Relatives in the immediate family (cousins, aunts and
uncles) are allowed to come less often; they are put in waiting
rooms on the ward and told when they can approach the
patient for a brief visit. Distant relatives may be asked to wait
in the main lobby of the hospital; they are permitted to come
to the ward waiting room only during regular visiting hours,
and then they must wait their turn for a few minutes with
the patient.
This criterion for distribution dovetails neatly with the
tactic of letting family members take care of each other. Often
the strongest or most dependable member is in the immediate
family and is therefore likely to be readily available in the ward
waiting room, to help console an intimate family mem ber who
breaks down at the patients bedside. The strong family mem
PART
ber can then, if needed, take the sobbing one down to the lobby
to receive the support of the whole family. Keeping the intimate
family members close at hand also allows them to spell each
other at the patients bedside, thus reducing the strain on any
one member.
U nder the stress of waiting, however, family members may
develop drift routes, the m ore distant relatives trying to get
closer by slipping back up the scale of distance the staff has
put between them and the patient. They drift up elevators, out
of waiting rooms, and past the nursing station to the patient s
door and in to his bedside. W hen they are noticed and chal
lenged, they obligingly return to their assigned waiting space
but soon they start drifting back again. Nurses do have some
control over this tendency, especially in the intensive care wards
where space is at a premium. Once at the door the family mem
ber usually stops and peeks through a window or otherwise
attracts a nurses attention. She may either invite him in or
give him a time to return for a legitimate visit. As long as
relatives acknowledge this door norm , nurses have a means
of control that they can use to redistribute the family members
to their appropriate waiting places.
The Relative as Worker or Patient. A hospital visitor can
not remain at his dying kinsmans bedside for very long without
taking on a role other than that of relative or visitor sug
gested by the situation, one that integrates him with the staffs
daily work. Since the kinsman or relative role is external
to and often interferes with hospital routine, it is useful to
classify the relative in such a way that he becomes a dependable
participant in the work context. The two m ajor roles adopted
by family members are patient and worker, both of which
link them directly to daily activity.
A less frequent alternative is to turn the constant visitor
into a non-person. Completely ignored and discounted, non
person relatives are least disturbing to staff work. In state and
county hospitals for the aged, and in VA hospitals and on
emergency wards, where family visits are less common, few
lh e Aware Family
165
provisions are made for them, and the non-person alternative

is frequently used. Often under these conditions, however,
staff members have so little time and interest for individual
patients that they prefer to convert relatives to workers instead.
Of the two major roles, the preferred role is usually that
of worker. If the relatives are going to be around all the time
they might as well help, so the rationale goes, and the simple
fact that many dying patients require more than the normal
amount of attention prom pts the nurses to ease their burden
by taking advantage of family members constant presence.
A relative may help without being clearly cognizant of it, merely
by watching the patient. A nurse can leave a patient who must
be watched with a family member who does not know that
he is actually doing a job, because she can depend on him to
notify her if something happens. Or, she can ask the family
member to watch for a specific change or even for death and
call her immediately; in return, the relative is left in privacy
with the patient. Sometimes the family mem ber is clearly aware
of being part of the medical team and is grateful for the chance
to help. Help can vary from such tasks as keeping a cool cloth
on a patients forehead to taking over most nursing care, as
when the relative is also a nurse. Of course, as we have indi
cated, the family member must usually be aware that the
patient is dying before he can be an effective helper. An
unaware relative may very well become confused and panicky
if the patients condition changes, and hence he is not depend
able enough to be used as a worker in an emergency.
In the hospitals we studied, family participation is not
institutionalized, and the nurse who initiates and permits it
is held responsible for what the family worker does. But in two
hospitals described in the literature, the family is worked into
the daily round and therefore bears responsibility.11 Not only
does this give the nurse more time for other patients, but it
" See K. M. Carpenter and ]. M. Stewart, Parents Take Heart
at City of Hope, American Journal o f Nursing, 62 (October 1962), pp.
82-85; and Leonard Reissman and John H. Rohrer, Change and Di
lemma in the Nursing Profession (New York: G. P. Putnams Sons
1957), pp. 28-48.
166
PART
t h r e e: pro blem s o f
aw areness
means that the family shares any imputation of negligence in

the care of a dying patient. Family helpers were used, in the
hospitals we studied, m ost extensively by nurses who could
accept sharing their work, but not their
responsibility.In the
City of Hope Hospital, nurses in the parent participation
program on the pediatrics ward were chosen for their ability
to share both work and responsibility; but in Charity Hospital,
sta ffin g was so difficult that nurses or aides had no choice
but to work with visiting attendant family members.
As we shall see in m ore detail in the next chapter, aware
families are also screened to see whether they can help or even
take over certain decision-making responsibilities. They are
asked about alternative treatments, operation strategies, differ
ent forms of hospitalization, disclosure to the patient and
the m ost crucial issue whether the patients life should be
prolonged, given the specific contingencies of the case. These
grave responsibilities can be delegated
to the aware family
only if it is prepared for the patients demise.
Under some conditions, nurses treat a family member as
if he were another patient. The m other of a dying prem ature
infant is formally a patient. But family members who are not
formally patients may break down or become paralyzed with
the news of their kinsm ans impending death and, for the
mom ent anyway, they become patients, given sedation, comfort,
and consolation and put to rest in the hospital or at home.
Nurses with a psychotherapeutic orientation turn family mem
bers into patients in helping them prepare for the death of
their kinsman. And a nurse sometimes relieves the strain of
a status-forcing scene, as we have noted, by shifting the focus
of her involvement from the patient, whose case is hopeless
anyway, to the family mem ber who is pressing for more care.
In essence, she turns from a patient who is lost to a patient
who will live and can gain substantially from her efforts, thus
sparing herself the frustrating helplessness of being able to
do nothing for the m oribund patient. In object switching she
directs her efforts to where rewards can be anticipated.
The Aware Family
167
In sum, the relaxation of visiting rules alters the relations

between nurses and family members. W hen the rules are en
forced, family members are allotted a specified time alone
with the patient, and the nurse schedules her work accordingly.
H er interaction with the family consists mostly of greeting
relatives and ushering them in and out. But when the visiting
rules are relaxed or waived, the nurse may be in m ore or less
constant contact with the family, and she m ust find a way to
integrate these exchanges with her work routine. She can do
this most advantageously by assigning the visitor a role
preferably as a worker, but also as a patient who fits into the
hospital routine. This tactic also helps resolve the privacyversus-work dilemma. The family helper has a useful period
of privacy with the patient, and the family patient can legiti
mately be separated from the dying patient for rest and quiet
preparation.
PR E PA R A T IO N F O R D E A T H
Degrees of Preparation. Staff and family members both
benefit from disclosure to the family, for disclosure makes it
possible for the family members to compose themselves with
preparations for their relatives impending death. Depending
on the way the staff handles the disclosure, and the type of
expectation the family is given at the time, disclosure may
result in various degrees of preparation. Perhaps the most
im portant condition is the type of death expectation given
the family when it is brought into awareness. On the basis of
this expectation the family establishes the delicate balance
between accepting and not accepting the patients fate. This
balance results in a degree of preparation which, in turn, de
termines the familys relative ability to remain composed during
dying and at death.
Maximum forewarning occurs when the family is told that
the patient is certain to die at a specific time. In this situation,
the family members can time the settling of affairs and manage
168
PART
their feelings of grief so as to be most fully prepared; and

at death staff members will probably say, They are taking it
very well, She was wonderful, or She was trembling, but
werent we all.
Family members who are told that the patient will die, but
at an unknown time, will be relatively well prepared at the
end, but less so than if they had known when death would
occur. Not having had a time limit, they will still have some
arrangements to make. The death may still come unexpectedly,
and therefore be more disturbing than it might otherwise have
been. Families usually give the patient more time than he has,
if the doctor does not specify time of death; and since doctors
tend to prefer a gentle disclosure in which time is often not
specified, most families find themselves in this situation.
W hen the family is told that death is uncertain, they are in
effect alerted to begin preparation without giving up hope. If
the time that they can expect a resolution of the uncertainty
is also unknown, their preparation is likely to be very slight.
Some families simply refuse to begin preparing, and just con
tinue to hope. Staff members find that families in this situation
are hardest to handle, since death comes to them as a surprise,
and they have made virtually no preparations to help maintain
their composure. A family with this degree of awareness may
therefore act similarly to an unaware family.
Staff members (employing the transition strategies mentioned
in the last chapter) therefore try to help the relatives change
their expectations as the patients condition worsens, so that
they will begin to settle the patients affairs and consider what
life will be like after his death. The general process of revising
the familys expectations to fit the patients true condition is
contingent on other factors that both help and hinder the
revision. Altering the family expectation only with cues is apt
to leave it on a suspicion level of awareness, still in doubt.
The family that still has doubts is likely to continue hoping,
which in turn reduces preparation. On the other hand, staff
members are reluctant to announce a definite time; although
The Unaware Family
169
they are sure of the time they do not like to be held accountable
for it. A nurse put it so: I do not like to play G od. In one
case, a m other spent her days at her sons bedside, expecting
him to die but, since time was unknown, not expecting death
for a while. Suddenly she noticed that the boys breathing
had become more difficult, and thinking he might soon die, she
asked the nurse about it. The nurse avoided the issue, leaving
her suspicious but uncertain, so that when the boy did die
the blow was harder than it would have been had the nurse
verified her suspicions.
We reiterate, families tend to give patients more time than
they have it is a way of sustaining hope. W ithout a roughly
specific idea of the time of death, family members may tend
to delay their preparations on the basis of a short-run expecta
tion that the patient will continue to live for a while. They
expect him to be alive the following day, even when they are
certain he will die sometime. They expect, therefore, a daily
o r even a weekly living relationship with him, forgetting the
end m ust come. But doctors and nurses can encourage family
preparation, and thereby enhance composure, by keeping fam ily
members up to date with realistic expectations. To this end, it
is advisable for staff to give accurate responses to fam ily mem
bers who are testing suspicions, even though this may require
them to make the statement of probable time of death that
they are generally reluctant to give.
The disclosing of time of expected death is also paced
according to the period of illness, both of which in turn affect
the amount of time the family has to complete preparations.
The patient who is certain to die, but at an unknown time, and
whose illness is a long, lingering one, provides the best oppor
tunity for family preparation. Cases like this afford both
stimulus and time for grieving and for arranging affairs. The
problem here is to see that family members are told the ex
pected time of death early enough to allow them to complete
their preparations. Of course, if the doctor gives no certain
expectation that death will occur in the foreseeable future, as is
170
PART
most often the case, then the long illness may be of little bene
fit in allowing the family to prepare itself. W hen a transition
in expectations finally does occur, the family may have no
more time for preparation than the family that learns of the
impending death, only a few weeks in advance. Adequate
preparations take time even when the family members are
told that the patient is certain to die in a few weeks so that
they are stimulated to make full preparations, these few weeks
may not be enough. Short-term illness, or a fatal accident,
often leaves the family only a few days or hours, but even
then an adequate expectation (certainty, known tim e) is neces
sary to stimulate even the minimal preparation that is possible.
Often, however, even on emergency wards where the patient
may be on the verge of dying, the family is only given an
uncertainty expectation. Thus even deaths expected by the
staff can be very surprising to a family that has not had ade
quate death expectations to stimulate its preparations.
Illnesses associated with reprieve patterns that is, the
patient is on the verge of death many times but recovers some
what hinder full preparation. Reprieves spark hope, so that
the family is never completely prepared for the death: They
think the next reprieve may be complete, not tem porary. Thus,
even when the doctor announces a time of death, usually a few
days ahead of time, the family members may feel that the
miraculous return to life they have watched before might
happen again. As one uncle said, It was hard to fully reconcile
ourselves since weve seen her revive before and she could
do it again. W hen a pattern like this is established, the family
may refuse to give up all hope even if the doctor tells them
they must, which may put the family through many tortuous
episodes after each, wishing the patient had died. Further,
the short notice he is able to give them is usually not enough
to allow the family members time to reconcile themselves and
begin preparations.
Linked with lingering and reprieve-pattern illnesses is the
ever constant hope that medical researchers will find a cure
The Aware Family
171
in time: They are working on it constantly. Some families

travel from medical center to medical center and even make
telephone calls to research centers all over the world to see
how close they are to the cure, adjusting their preparations
accordingly.
Helpers of the Family. Family preparations are considerably
affected by the people who step in to help them arrange affairs,
listen to their grieving and reinforce their belief that death is
certain. These helpers may form a well organized team, as in
a Catholic hospital where a nurse reported: If Im with the
patient and their relatives break down, I flash the signal light,
and a head nurse comes to take them to sister or the priest.
In this hospital there were three floating sisters whose only
job was to help families and patients through critical periods.
Whenever a patients condition became dangerous, the nun
was called right away to help family members prepare them
selves. She would give them a rationale for their current death
expectation: Its a blessing that he goes now rather than linger
on for m onths. She may take them to the hospital chapel to
pray and show them religious pictures. Grieving family members
characteristically want to talk to someone about their impending
loss: for the most part this talk is not focused, and it requires
only a good listener. The sisters job is to listen patiently and
sympathetically.
In other hospitals, chaplains, social workers and ward
clerks help the family, thus relieving the nurses of a burden
that might otherwise interfere with their normal obligations
to other patients. W hen the hospital has no facilities for aiding
families, and does not require the nurses to help the family
(that is, their patient loads are not allocated with this considera
tion in m ind), then family members are likely to be treated
as non-persons. In one such hospital for lower class people
only the nurses whose hearts were touched these people
are so neglected would bother to stop and talk with the
relatives, sometimes offering them coffee from their own urns.
172
PART
To be sure, the pattern typical of most hospitals is that both

nurse and doctor spend a little time with the family. They listen
for a few minutes and reinforce the familys death expectations
with a supporting rationale. Other hospitals try to find social
workers and chaplains who will help the family while the
patient is dying, not just at death.
Helpers of the family can furnish support even when they
are not aware that an individual is grieving. They need only be
sensitive to the fact that the person is going through some
thing and treat her with great care so as not to make her
spill over. One woman, sitting with a dying patient, became
restless; tears trickled down her face, she tried to read, she
went in the bathroom to freshen up, she went to the waiting
room to be alone. The nurses did not think of her as grieving,
but they realized that she was upset and did not push her
by asking questions. They left her alone, ready to help her if
she approached them.
This kind of support from nurses who are not precisely
aware that someone is grieving helps to keep the relative
from breaking down under the strain of waiting for the patient
to die. If the nurse does not treat the relation as a grieving
person, he is less likely to collapse just from receiving sympathy,
and this in turn causes her to respect him for his emotional
stam ina.12 Not only does her respect furnish additional support,
but his apparent emotional control during waiting makes her
more tolerant of minor breakdowns as the patients end ap
proaches. H ad she been aware that the relative was grieving
all along, she might have expected increasing restraint, and
have been annoyed by even a minor breakdown at the end
because by now he should have been well prepared. Thus para
doxically, giving explicit training in helping people prepare
themselves for a death in the family is perhaps less important
than using sensitivity as a criterion in selecting staff members
to help the family.
12 In another context it has been noted that undue sympathy breaks
down the composure with which a person may be coping with a dis
agreeable physical condition: Fred Davis, Deviance Disavowal, Social
Problems, 9 (Fall 1961), pp. 122-125.
The Aware Family
173
Stages of Grieving. Within a single family, individuals may

be at different stages of grieving because each m em ber may
form a different kind of death expectation, or may be brought
into awareness at a different time, or may require more or less
time for grieving. Some relatives may spend more time with
the patient, which stimulates more grieving. Usually a very
close relative is ahead of the others; he has known for a
longer period that death is certain, and he has probably spent
more time with the patient. Having had more opportunity to
complete his preparations for the patients death, he is more
likely to have finished grieving as well. But when a distant
relative who arrives late on the scene is overcome with grief,
he forces the more intimate one to act very sad, if not equally
overcome, so as not to appear unduly callous about the loss
of someone so close. (This status forcing as in being forced
to play the wailing wife often occurs also at funerals.)
W hen lingering is very prolonged, family members may
finish grieving before the patient is dead, which may be a
source of strain to the family members. They may begin almost
to resent the constant visiting. Men want to get back to work,
women to their households. M oney is being drained away.
Beginning to wish the patient was dead, famliy members may
feel intensely ambivalent or guilty. In situations of this kind,
nurses and other helpers of the family can be most helpful
if they have been aware of grieving and can sense the point
at which it is ended. Then, as in cases we observed, nurses can
assure the relatives that their feelings are normal and nothing
to feel guilty about. They offer to spell the family at the patients
bedside Go home and wash up, Ill keep watch for you,
Go get your rest, or Go get some coffee. Thus the family
m ember relinquishes his position on the medical team little
by little, as the nurse resumes the tasks she allocated to the
family when the increased visiting began. Even if the patient is
still alive, the family members gradually cease to help the staff,
and their visiting diminishes, sometimes to the point where
they do not visit at all. If money becomes short or the family
does not want to pay hospital bills indefinitely, the lingering
174
PART
(often aged) patient may be sent to the geriatric ward (referred

to by some personnel as the vegetable patch ) of a state
hospital. There he is fed and given routine care, but he
receives only infrequent visits from his family. One patient
became comatose but lived on for years with scarcely a visit.
In contrast, nurses who are less sensitive, or are unaware that
family members can finish grieving before a patient dies, may
feel disgusted with rejecting relatives, and of course this will
affect their relations with the patients family.
Nurses also grieve for their patients, though they may
reserve emotional outbursts for later, private occasions. A
pediatric nurse told us that she did not break down until a
week later, and then when she was at home only then did
she realize that she too had to grieve for her patients. Because
a nurse is usually less involved than family members, and
because she is apt to be better informed about the patients
condition, she is normally at a more advanced stage of grieving
at any given time. This gives her the stability and sympathy
necessary if she is to help grieving relatives. Being at roughly
the same stage of grieving can cause the nurse to avoid family
members for fear she and they both will break down! If a
nurse falls behind family members, perhaps because of denial
or too brief a period of awareness or too short a time on the
case, she risks treating an intimate relative in term s of her
own grief, assuming that he too is still very sad. This may
revive the more painful stages of the relatives feelings of im
pending loss, which he has already gone through once. She
also, like the family mem ber who is behind the close relative
in grieving, may force the relative to act overcome.
IN FO R M A TIO N
When a dying patients family is brought into awareness,
the members need to keep abreast of the patients condition
is intensified; hence their demands for information increase.
Staff members necessarily develop many interaction tactics to
The Aware Family
175
control both family questioning and the information they will

give. Since we have already presented a good many of these
tactics and their consequences, we shall devote this section
to some special problems in managing the aware familys con
stant search for information.
Staff members in wards with closed doors can control their
accessibility to visitors with various door norm s norms for
behavior which are associated with the side of the door each
party is on and thus limit demands for information. Once a
family mem ber has been allowed to pass through the door, he
has the right to request information on the patient as he
walks to the bedside, and the nurse m ust answer the query
as best she can. A t the same time, however, she can also give
rules as to how long he may visit and how often, again limiting
the familys access to her for more information. Nurses may
find questions especially bothersome because the family member
is sometimes in a mild panic and does not listen to the answer
anyway: as one said, You try and explain to them, but they
cant hear. Thus nurses learn to answer with as little inform a
tion as possible to avoid a sense of futility and wasted effort.
If family members are not allowed through the door, then
of course, the staff is inaccessible to queries. If the door is left
ajar, visitors may be able to glean inform ation from voices or
from glimpses of the activity within. Sometimes a relative
is told he cannot enter, but m ust return to the waiting room
and wait for the. nurse to bring news to him. The nurse may,
however, intend to keep this promise only when she is going
somewhere else and the waiting room is en route, or when a
drastic change occurs in the patients condition. Partly because
it is so hard for them to get information, family members tend
to regard the first nurse who steps out the door as fair game
for questions. As they do with nurses in an open nurses station
facing outward, they ask about their patient regardless of
her familiarity with the case.
Bringing the family into awareness sometimes stimulates
constant phone calls to the ward for information about the
176
PART
th r ee: problem s of
aw areness
patient. To avoid tying up the lines, staff members may repeat

a routine report over and over No change, Rested and
ate well, and so on or they simply tell the switchboard
operator not to put such calls through or to repeat the routine
report. A family member who does reach the ward nurse is
likely to crowd the call, asking as many questions as possible
in rapid succession before the brief span of an appropriate call
is ended and the nurse cuts off the conversation. Of course,
this makes it hard for either party to get or keep anything
straight.
Family awareness eases communication for both staff and
family, as we have indicated. Staff members do not have to
avoid certain sensitive topics, and the relatives can ask more
intelligent questions. But even in this open awareness situation,
dimensions of the patients condition such as mode of dying
may still be closed; or staff and family may have different
expectations as to certainty or time of death, so that ambiguities
of communication exist (as discussed in C hapter 6 ).
A nother kind of communication difficulty is associated with
staff announcements to the family. In this section we have
focused on the familys problems in obtaining information, but
as we have suggested previously, staff members must announce
information at various stages in the patients dying, and then
they encounter additional obstacles. Family members may be
difficult to locate in time for a quick bulletin on the patients
changed condition, for example, and there are always questions
about how the news is to be broken abruptly or gently
and whether the staff member should display emotion. Not the
least im portant are techniques for getting away from the kins
men after making the announcement.
11
Nothing More to Do:

The Problem of
No Recovery
Perhaps the most crucial phase in dying begins when there is
nothing more to do for the patient. Nothing more to do
the designation used by the staff in every hospital we studied
means that the patient no longer has a chance for recovery,
either through natural processes or through medical help. As
long as there is still a question as to whether the patient will
recover, then all parties concerned are uncertain in their death
expectations. The nothing more to do phase begins with the
certainty that death will occur, at either a known or an unknown
time. Certainty may occur at any time during the patients
dying. It may begin with the original diagnosis, though the
patient may linger for years. In cases of this nature, however,
periods of uncertainty about death are bound to occur because
of partial reprieves or new drugs or surgical techniques. Or,
the nothing more to do phase may occur only moments before
death, when the staff, working intensively to save the patient,
realizes just before death that the case is hopeless. In sudden
death, of course, this phase is missing.
The nothing more to do phase of dying is crucial because
at its inception the fundamental goal for the patient changes
from recovery to comfort, and as a consequence, the activities
of the staff change radically. Before this phase begins the staff
focuses its efforts on saving the patient with all available means.
After this phase begins, any effort to save the patient becomes
177
178
PART
superfluous, and staff activities are focused on ways to make

the patient comfortable until death. With this change of goals
and activities occurs a change in the rewards the staff derives
from patient care. F or the personnel, the highest professional
reward is in the patients recovery and return to his normal
personal and social life.1 M aking the patient comfortable is
less rewarding, partly because it requires less medical and
nursing skill, and especially because merely providing comfort
to a patient who is dead-in-anticipation is still a kind of medical
failure: there is no longer any chance to achieve the primary
success, recovery.
Since the staff mem bers strongest m otivation to become
involved in patient care is the chance to cure the patient,
nothing m ore to do patients are excluded from some wards,
particularly the intensive care units which are geared to inten
sive, split-second heroics. On these wards all patients should
have the possibility of recovery to avoid any dilution of high
staff morale and motivational involvement. To this end, screen
ing -measures are used to insure that any nothing more to do
cases sent to the ward are refused entrance. W ithout such
measures, the ward might become a dumping ground for
the dying, and this might damage staff motivation to continue
all-out intense efforts to save other patients.2 On most wards
1This paragraph delineates the two principal goal-reward patterns
of medical personnel and the institutionalized priority between them,
and how the motivated involvement in patient care of personnel varies
accordingly. The goal-reward patterns of positions provide a useful
basis for analyzing work behavior. See for clear example: Howard S.
Becker, Blanche Geer, Everett C. Hughes and Anselm L. Strauss, Boys
In White (Chicago: University of Chicago Press, 1961); James Coleman,
The Adolescent Society (New York: Free Press of Glencoe, 1961);
Barney G. Glaser, Organizational Scientists: Their Professional Careers
(Indianapolis: Bobbs-Merrill, 1964); and W arren O. Hagstrom, Anomy
in Scientific Communities, Social Problems, 12 (Fall, 1964), pp.
186-195.
1 Ruth Laub Coser found in one ward that was a dumping ground
for nothing-more-to-do elderly patients that the staff developed a
vicious circle of rejection of these patients. We might suggest that
this rejection process was in the service of maintaining the staffs moti
vation to achieve recoveries. See Life on the Ward (East Lansing:
Michigan State University Press, 1962), p. 123.
Nothing More to D o T h e Problem of No Recovery
179
that contain dying patients there is some typical stable ratio

of both certain and uncertain death. In most w ard situations,
then, the lesser rewards of providing only comfort are usually
balanced by a sufficiently large num ber of terminal patients
who do provide the challenge of potential recovery.
Linked closely with the change in goals of patient care,
from recovery to comfort, is the moral dilemma of whether
to prolong life, and if so, for how long. We shall present later
in this chapter some of the standard resolutions to this complex
and continually debated issue. Suffice it to say here that before
the patient reaches the nothing more to do phase, the recov
ery goal requires prolonging his life at all costs, since the patient
still has a chance. But after the nothing more to do phase
begins, after those responsible for his care are certain that the
patient will die, prolonging life can become useless. It may
require meaningless effort from the staff; and if it means
excessive pain for the patient, it conflicts with the goal of
providing comfort, which is also disturbing to staff. Excessive
prolonging may induce intense emotional pain in the family,
another problem for staff. All this pain is senseless and unneces
sary, from the perspective of many staff members. Thus, the
problem is to bring the nothing more to do phase to a close
in an appropriate manner.
Since this m ajor change of goals has such a profound
impact on patient care, the particular awareness context relative
to the nothing more to do phase is an im portant consideration.
Much of the interaction among staff, family and dying patient
hinges on whether each party is aware that the patient is
certain to die and whether the time of death can be predicted.
In this situation, there is much room for differential expectation
among all parties. Legitimately, doctors are the first to decide
that there is nothing more to do for the patient; as we have
seen, it is the doctors prerogative to decide whether or not to
tell nurses, family and patient, who may already be suspicious,
if not aware. Typically, the doctor is aware before the nurse
is. She, in turn, is typically ahead of the family, and all are
180
PART
usually aware before the patient is told. Even when disclosure

proceeds directly toward an open awareness context, there
will be moments when individual death expectations are differ
ent o r when the awareness context is closed or suspicious, which
will create problems of interaction and patient care. To take
an unusual example, suppose that a doctor on an emergency
ward suddenly stops trying to save a patient. O ther doctors and
nurses are dumbfounded, and family members promise them
selves to bring a malpractice suit. The burden is clearly on
the doctor to disclose, as quickly as possible, that there was
really nothing more to do, giving the reasons for his own
sudden transition in awareness. This clear-cut situation seldom
occurs, because it is complicated by numerous contingencies,
as we shall indicate presently. A far more likely outcome is
that the doctor is so intensely involved in striving to save the
patient that other doctors and nurses become aware first and
must tell him that it is hopeless and time to give up. O r
family members, suspicious of the truth, may beg him to stop
and let the patient die gracefully and in comfort. In short,
the problem is more often undue prolonging and how to stop it.
In the nothing more to do phase, doctors activities in
response to the change of goals can be placed on a continuum,
ranging from ceasing all recovery efforts to still trying valiantly
to save the patient. We shall consider these activities as they
bear on the interaction between the doctor and nurses, family
members and dying patient, in different types of awareness
contexts.*
PR IO R IT IE S IN PA T IE N T C A R E
Once the doctor is convinced that there is nothing more
to do for the patient, his problem becomes one of spending
his time legitimately, according to professional values. Because
3 Confining our discussion of interaction within awareness contexts
limits for this chapter an otherwise very extensive topic. In another
book, Passage Toward Death, we shall consider staff activities in terms
of timing the patients passage through the nothing more to do phase.
See also, Tem poral Aspects of Dying as a Non-scheduled Status
Passage, American Journal o f Sociology (June 1965).
"Nothing More to Do": The Problem of No Recovery
181
his scarce skills must also be applied to a large num ber of

patients who can recover, his nothing more to do patient
has less valid a claim on his time; the simple priority assigned
to patients who can be helped tends to take him elsewhere.4
As one doctor put it, If you know it is necessarily ultimately
fatal, then you may not work too hard. 5 The increasing
absence of a doctor who has previously been more attentive
can make an unaware patient suspect that he has been given
up for lost, since he obviously isnt getting better. Thus, this
basic structural feature of medical practice the high priority
assigned to patients who can recover linked with the nothing
more to do phase may endanger a closed awareness context,
for most people know that doctors disappear only when either
recovery or death is certain. The patients suspicion can easily
change to open awareness, if the doctor cannot adequately
counter the accusations of neglect and incompetence (one
patient yelled: Youre no longer a doctor ) leveled by a
suspicious patient when the doctor makes one of his brief,
infrequent visits.
SENDING T H E PA TIEN T H O M E
The standard mechanism for handling this problem of
priority in care is to send the patient home, if this is feasible.
The doctor who tries to get rid of the patient justifies this
action by referring to another fundamental value in medical
4 The private doctor, compared to the hospital doctor, is likely to
withdraw less time from the patient because of his location in the lay
referral system compared to the hospital doctors location in the pro
fessional referral system. See Eliot Freidson, Patients View o f Medical
Practice (New York: Russell Sage Foundation, 1961), Part 2, for the
contingencies of both the lay and professional referral systems that
control the doctors behavior.
5 All of us have seen the patient who is slowly dying of a chronic,
debilitating illness and has been placed in the room farthest from the
center of the ward. The doctors drop in briefly during rounds, glance
at his chart and leave almost immediately. The general attitude of the
wards is: There is really nothing more we can do for him after all,
hes dying anyway. [underlining is ours] Time, November 20, 1964,
p. 42. We see and hear this kind of statement quite often, which con
tinually validates our point.
182
pa r t t h r e e : p r o b l e m s o f aw areness
practice hospital space is reserved for the people who can

benefit from hospital care. In some cases, a term inal patient
must remain in the hospital because he needs its facilities for
sedation or its special equipment. In other cases the doctor may
send the patient home, though he expects him to return for
his last few days, when he will require the attention of a larger,
well-equipped staff. Sometimes the doctor can send equipment
home with the patient. In some of these cases, the family
simply sends the patient back to the hospital when it can no
longer tolerate his mode of dying. If his pain and condition can
be handled at home, however, and if his family can take the
ordeal, the doctor is often willing to send a patient home to
die. Implicit in this action, he knows, is a decision not to pro
long the patients life, since the home will normally not have
the equipment or supplies to sustain a dying patients life for
yery long if he is close to death. Also at home there are fewer
people to observe or hear the patient die, and the family will
usually agree to no prolonging after having lived with this last
stressful stage of dying.
To a sentient patient, the implications of being sent home
are ambiguous. Especially if the doctor does not visit the home
very often, the patient may infer that he has been given up
for lost. But most patients are eager to go home, if they have
a home; and if they have no cause for suspicions they may even
be pleased, since going home may also indicate progress toward
recovery, or at least the end of the crisis. Thus, the doctors
decision to send a dying patient home will not necessarily
jeopardize a closed awareness context.
The familys awareness is another matter. If the family
has been unaware that he is dying, the patients return can
cause a shocking transition into awareness, for now it is quite
obvious that he may not be going to recover.6 Otherwise, the
* In contrast to the pessimistic outlook for the dying patient when
he returns home, the outlook is optimistic for the recovery patient:
For the child and family the homecoming was a great event. Regard
less of the childs condition or handicap, the recovery outlook of
family members took a more optimistic turn when he returned home.
Fred Davis, Passage Through Crisis (Indianapolis: Bobbs-Merrill,
1964), p. 83.
183
probable result is a suspicion awareness context, muffled by

mutual pretense, during further interaction with doctors and
nurses. Before a patient is sent home, someone, often a social
worker, or a private doctor, usually tries to find out whether
the family can care for the dying patient, either unaided or
by hiring a practical nurse. Sometimes, in such a situation,
the social worker joins forces with aware family members to
ascertain what is best for both patient and family, regardless of
the doctors orders. One patient had a sister who felt that
she couldnt take it and said so directly: No, I dont want
[the patient] hom e. The social worker recommended no to
the physician. In other cases, a family mem ber and a social
worker, both aware that the patients death is very near, may
try to persuade the doctor to let the patient stay in the hospital
during his few remaining days. A relative in one of these cases
said he felt that it was useless to send the patient hom e be
cause he had such a short time to live.
The expected time of death after there is nothing more
to do is im portant in determining whether the patient will
be sent home, thus saving hospital space and medical skills.
If he is slated to linger, the family, aware or not, is under very
strong pressure to accept him at home. If they cannot or will
not do so, the patient can be sent to a nursing home or to a
terminal ward in a state hospital. These latter alternatives
permit the doctor to exert pressure on the family to accept the
patient, for the idea of the state hospital is ofen repugnant to
relatives, and m ost nursing homes may be too expensive. Some
families, however, may welcome these alternatives. But when
the family knows that the patient has but a few days to live,
they can use this expectation to force the doctor, and the hos
pital, to keep the patient at the hospital.
Sending the patient home is sometimes of course the most
satisfactory solution for patient, staff, and family.7 The patient
dies in a familiar environment, surrounded by his kinsmen,
7 Preference for dying at home has been shown by two other studies;
see Robert Fulton, Death and the Self, Journal o f Religion and
Health, 3 (July 1964), p. 364.
184
PART
and the staff is spared a period of strain, if not an ordeal. The

ideal conditions for dying at home are difficult to achieve,
though when all or most of them are present the family is
usually relatively eager to have the patient at home. The patient
should be fairly free of pain and not physically deteriorating
too rapidly, so that he is neither comatose nor completely
blurred by sedation and he is presentable. Fam ily members
should have already had sufficient time to grieve, so that they
do not break down. Ideally, too, the awareness context should
be open enough for the family to openly take leave of the
patient.
E X T R A O R D IN A R Y PR IV ILEG ES
As an alternative to discharging the nothing more to do
patient from the hospital, doctors may balance their reduced
attention by allowing or ordering extraordinary privileges for
him.8 These privileges are never included in routine treatment
of a curable physical condition, and they are strongly taboo
for patients who will recover, but they are both compensation
for lack of medical attention and a source of comfort during
a patients last phase of dying. F or example, the patient may
be given a pass to leave the hospital for a day or a weekend
to go home or to town, though a patient in comparable physical
condition, who was recovering, presumably would be bedridden
and watched constantly. A nother privilege is arranging for the
patient to go out for a ride. The patient may also be allowed
to eat what he wants his last meals whereas if he were
recovering he would be on a stringent diet. O ther special excep8 See the related discussion by Renee Fox on the red carpet treat
ment of patients in which the special privileges are used to induce
patients into an experiment and to keep them happy while in it. Some
of the privileges accorded the nothing-more-to-do patient for his com
fort and to compensate loss of attention would preclude being a re
search patient (e.g., steak dinner or a pass). Thus, the content of the
privileges and their purpose vary, but the common factor between our
two studies (and for that matter in all studies of total institutions) is
that the people in charge have the power to select from a broad range
of privileges, some to give to their inmates to achieve their purposes.
Renee Fox, Experiment Perilous (New York: Free Press of Glencoe,
1959), pp. 88-89.
"Nothing More to Do: The Problem of No Recovery
185
tions have to do with ward routines dying patients may be

allowed to remain in bed, for instance, when others must walk
a bit (o r vice versa), or to skip taking baths, or take extra
ones.
F or the previously unaw are patient, the relief that comes
with deviance from stressful routine may fail to compensate for
the growing realization that he has these privileges only because
the doctors can no longer help him. Thus, these privileges may
become transition conditions that are very likely to transform
the closed awareness context into a suspicion, pretense, or
open context. But to compensate somewhat for disclosure, the
doctors reduced attentions are no longer misunderstood, and
if the patient accepts his fate, he can appreciate the increased
comfort afforded by the privileges.
Staff and family members who are not yet aware of the
patients certain death may become aware rather quickly if
they see the patient eating an unusually choice or large meal,
for example. Implicit in these privileges, in fact, is a decision
not to prolong the patients life unduly and sometimes even
to shorten it somewhat for comforts sake. Thus, a student
nurse was shocked when she saw a patient with a serious liver
ailment eating a steak dinner. Not realizing the patients true
status, she remarked, It will kill him! In actuality, it probably
did shorten his life by a few days, com pared with w hat routine
treatm ent might have accomplished at the price of more pain.
Staff and family often feel relieved that the dying patient is
spared the stress of a rigorous medical regime during his final
hours.
T H E IN T E R E ST IN G CASE
Although it contradicts the usual priorities in patient care,
a patient dying in a medically interesting way, or suffering
from an interesting condition, may receive special attention
as an object of study and as a teaching case. 9 This increased
attention does not necessarily m ake the unaware patient sus9 See related discussion on medical stardom, in Ren6e Fox, op.
cit., pp. 148-153.
186
picious. H e may feel that his chances of recovery are improved

by the attention of so many doctors. The unaware family, too,
may become very hopeful when they find num erous experts
concerned with the patients condition.
Hope for the patient wanes, however, and even turns to
high suspicion if his case is so interesting that the doctors
decide to keep him alive for the rest of the semester and start
applying various kinds of equipment to prolong life. Medical
equipment can be one of the surest indicators to patients and
families that death is certain, but is being delayed. Once aware
of dying, the patient may then have to ask for his own death,
to put a stop to undue prolonging. A nurse told of a patient
who was kept alive for over three weeks on a pacem aker for
teaching purposes. This was so hard on both the patient and
her family that the patient, knowing she would die, told the
doctor to stop the pacemaker. She was dead within thirty-six
hours.
DO SO M ETH IN G C A R E
Doctors who are fearful that reduced attention will bring
an unaware family or patient into awareness possibly with
unpleasant scenes, hopeless depression, withdrawal, suicide,
or retreat to marginal doctors or quacks often do something
in order to preserve the closed awareness context for the patient
or his family.10 T hat is, they act as though they were still
trying to save the patients life. The easiest tactic is to pass
off an effort to make the patient comfortable by sedation,
for example as a way to cure him. The extreme example is
surgery, actually intended to make the patient comfortable, but
presented to the family as a means of helping him recover. O r
the doctor can make meaningless trips (th at is, for no medical
reason) to the patients bedside, to m aintain the illusion that
he is still trying to save his life. The doctor makes these actions
as visible as possible, examining the patient briefly and then
10 See related do something activities on Fox, op. cit., throughout
Chapter 3.
187
spending his time comforting the patient and his family. In

this way the doctor gets credit for doing something, and even
if the family still insists on seeing a professionally marginal
doctor, with his highly promising treatm ent, the original doctor
retains a measure of control over the patients dying. In his
periodic examinations, he can make sure the marginal treatm ent
is doing no harm while it keeps the patient occupied and
hopeful.
The doctor usually buttresses do something care by
advising patience ( Give the new treatm ent time to work ) and
by referring to research promise ( They are working on it at
the M ayo Clinic ). Sometimes the doctor who uses these
rationales must eventually admit that his hospital cannot provide
the equipment or the unique specialized care and knowledge
to handle the case. He may then recommend sending the patient
to a university medical center where it can be handled more
adequately. Often this decision is quite realistic: a case that
is classified as nothing more to do in one hospital is not
necessarily so classified in another. In an urban university
medical center, for instance, superior resources and facilities
may improve a dying patients changes for survival immeasur
ably, com pared with what they would be in less advantageous
an environment.
The extent of do something care varies among hospitals
and types of medical practice. Private doctors, pressured by
the client referral system, practice do something care con
stantly, sometimes making several trips to the hospital each
week.11 State, county, and veterans hospitals, where many
patients are a non-paying, captive clientele, induce less involve
ment of physicians with the patients, and do something care
is applied less often. When dying patients have several doctors,
it is likely that no one doctor will feel greatly responsible for
inattention to them in the final stages, and so these patients
will receive less do something care. Lingering patients who
remain in the nothing more to do phase for months or years
11 See Freidson, op. cit.
188
PART
are likely to receive a good deal of this kind of attention, espe

cially from private doctors. It keeps them from becoming
aware, if they are not already aware, and it helps the doctor
convince the family that we did everything possible, since it
keeps him visibly busy with the patient until the end. Do
something care for lingering patients may constitute a test for
a new treatm ent, and it may even save the patient, for the longer
a patient in this condition hangs on, the more probable it is
that research will find a way to save him.
C LIN IC A L R E SE A R C H
The goal of recovery, with its high priority for the doctors
attention, can be reinstated, at least provisionally, by enrolling
the dying patient in a clinical research project. In general, the
basic legitimate condition for this proposal is the doctors
absolute certainty that, given present knowledge, there is
nothing more to do for the patient in any available hospital.
(F o r patients with sufficient personal financial resources, avail
able hospitals may include those in Britain, France and Ger
m any.) A t this time, the family and patient must be presented
with the facts of nothing more to do, even if they were
hitherto unaware. Then they are given the alternative: to risk
the research with its promise of recovery, however slight. The
negative side effects of the treatment or drug are not always
presented. If the patient is adult and sentient, he m ust consent
in writing; otherwise a family member must consent. Sometimes,
however, captive patients are put on a study drug unawares.
Parents who donate their child often do so without letting
the child know that he is dying, so that the closed awareness
context is maintained, at least in the beginning.
Recruitm ent into research is timed according to the current
type of death expectation and the availability of a testable
treatment. Among patients in the nothing more to do phase
there are more potential research cases than research experi
ments. A cancer patient, for instance, certain to die, may have
189
been lingering for months when a new chemical drug to be

tested is sent from the National Institutes of H ealth to the
hospital; the patient is then asked to be a research case, with,
if necessary, an attendant change in his awareness. Unaware
patients and families may be converted to awareness for naught,
if they do not agree to the experiment. The longer a patient
lives in the nothing more to do phase, the m ore likely it is
that an effort to recruit him for research purposes will occur
which may alter his own and his familys awareness.
O n the other hand, sometimes an experiment requires
waiting until a particular dying patient has reached the nothing
more to do phase. Having watched the patient closely, until
they are virtually certain of his impending death, doctors nego
tiate with patient or family immediately to try a new drug or
a transplant, for example. In one instance:
The death watch went on while the surgeons did every
thing possible to pull Callahan through; they even
dropped the tem perature of his entire body to decrease
his brains need for oxygen. But he showed no sign of
regaining consciousness. O n M onday, when all hope had
faded, Dr. Russell tackled his most difficult task. He
told the woman who was about to be widowed what his
colleagues wanted. [The wife] replied without hesita
tion: G o ahead if it will help someone else. 12
Once the nothing more to do phase had started, the doctor
in this case could legitimately ask the wife for the patients
liver, which was to be transplanted to another patient in the
nothing more to do phase who had volunteered for the test.
Again, negotiating for a research project requires an open
awareness context, which sometimes must be instituted quickly
say within an hour and sometimes must be readied weeks
or months in advance.
F or patients who are not to be recruited for research, how
ever, especially when experiments with a high probability of
1! Time, October 4, 1963, p. 87.
190
PART
saving the patient are going on at the hospital, doctors have

a stake in maintaining a closed awareness context for both
patient and family. For, if the family became aware that the
patient was certain to die, they might dem and that the patient
be given the experimental drug or that the new machine be
tried. Since for most experiments, resources are limited, there
are not enough chances to go around. This situation becomes
even more difficult for the doctors when the news media make
known that a new machine or drug with great promise is being
tested. In a university medical center, of course, both the
facilities for research and the pressure to give a patient a chance
to participate are much greater than they are elsewhere.
A patient can also volunteer for basic rather than clinical
research. In basic research the objective is to find possible cures,
not to test potential ones, so there is no implicit change back
to the goal of recovery. Yet for a lingering patient the hope
is always present that his project will succeed in finding a
cure, and he will be the first subject to recover.13 Thus, some
people volunteer to continue their dying at the National In
stitutes of Health, for example, where basic research is done
along with clinical research.
In clinical research, the chief drawback for all concerned
is that it takes time, and this means that for some patients life
will be unduly prolonged. Once committed to the research, the
patient and his family are expected to see it through to the
end. Certainly most research doctors are highly committed;
generally, their attitude was, If it is a study, we go all out.
One doctor did say, If a patient is in the study and near
death, we provide ordinary treatment. Dont use extra measures
just because they are in the study. If he is in agony, I wouldnt
keep a man alive just because of a study.
Nurses, on the other hand, tend to go along with research
in the beginning, but their collective mood in response to undue
prolonging is Why not let him die? W hen the doctors
continue to prolong the patients life, the nurses often feel
Fox, o^. cit., p. 150.
Nothing More to Do": The Problem of No Recovery
191
frustrated by their helplessness in being unable to let him die.

Occasionally they refuse to acknowledge the legitimacy of the
research effort the well-being of future patients suspecting
that the research patient is really being exploited for personal
career purposes, and making statements like: They keep them
alive until the research report is written. Yet, the nurses
remain helpless since, as one put it, Usually you hate to say
anything like, Why not let him die?
Resolution of this problem of unduly extending research
patients lives often depends on the awareness context. If the
patient is unaware, as in the case of a child committed by his
parents, or a comatose adult, then family members must decide.
This is not an easy decision to make, because often they can
see the possible benefit to future patients and so feel morally
obligated to allow an indefinite prolonging. As one nurse said,
The m other felt obligated, if it teaches them something.
Aware patients react to prospective prolonging according
to their death expectation whether they think of themselves
as lingering (tim e of death unknow n) or believe that without
excessive staff effort they would die quickly. A patient who
thinks he will linger anyway m ay expect the research to make
a possibly uncomfortable fate less so; hence he will probably
stay with the study. But if he feels that the research is increasing
his discomfort, he may ask to be released from it so that he
can live out his last weeks in comfort. A patient who is aware
that without research m edication or equipment he will die
quickly, must weigh this against the pain of having his life
prolonged, and decide accordingly. Again, this is a difficult
decision to make, and a patients intention to remain in the
study is always subject to reversal.
Because the equipment necessary to prolong life may be
so elaborate, many research patients readily become aware
that death would come within a few hours or days if they were
not in the project. Thus, when prolonging becomes very painful,
the patient is apt to lose any moral commitment to the research
he might have had, abandon any prospect of recovery, and want
192
PART
simply to be left alone to die. A basic principle in Clinical

research is that a subject should be allowed to withdraw when
he feels that continuation is mentally and physically impossible.14
But this principle is most appropriately applied to clinical re
search patients who are either not certain to die, or who are
expected to linger on fairly comfortably for weeks or months.
They have a moral right to save themselves discomfort; immi
nent death is not the issue.
A patient expected to die quickly if released from the ex
periment, on the other hand, may find it difficult to get out of
the research if he is not in undue pain. The doctors point of
view m ay be that his comfort, to be attained through his immi
nent death, is less im portant than the success of the experiment
if it can be achieved in a few m ore days or a week. A patient in
this situation can withdraw most easily by having his family
request it, but family members are not always willing to do this.
M uch as they might like to help the patient, they may be
influenced by the potential benefits to medicine in prolonging
his life, or they may still harbor the hope that recovery is a
real possibility. If the patient negotiates for dismissal, doctors,
who tend to be highly committed to the research, usually try
to persuade him to go on with the research, using the idea
that it is better than immediate death.
A patient who succumbs to a doctors argument, however,
may have second thoughts after the doctor leaves and try to
take measures into his own hands, in spite of his personal
obligation to the doctor and the potential benefits to mankind.
One research patient, having recovered from the side effects
of chemotherapy, was sent back for more. He asked the nurse
to remove his restraints, and she agreed, saying they would be
14 Fox, ibid., p. 47. It will be shown, however, in following pages
that this rule is often not the operating criteria for deciding on a
withdrawal. It is easy enough to say that the patient in his distraught
condition is not mentally capable of deciding whether he is mentally
and physically capable of continuing. It must, therefore, be decided
for him (!) by family or the doctor, and thus the patient loses this
apparent control.
Nothing More to Do": The Problem of N o Recovery
193
ordered back on in the morning. The patient nodded that he

understood. After the nurse left the room, he wrote on a
slate that he didnt want to go back for more treatm ent and
pulled his catheter out. When the nurse returned, she found
him in a pool of blood. She did, however, keep him from killing
himself.
Nurses who adhere to the let him go belief, based on the
comfort goal, try to help the research patient. They may ask
the doctor indirectly why he does not let the patient die, by
asking why he is prescribing a treatm ent for example, trache
otomy or more blood that will prolong life. The doctors reply
is likely to be on the order of because he needs it! So the
nurse either must ask directly, Why not let him die? or must
carry out her tasks of helping to prolong the patients life, with
no satisfactory answer. O ther nurses only a few simply walk
out, indicating they will not give the treatm ent or prepare the
patient. In reply one doctor said to a departing nurse: I
think we ought to give him the blood right now! One nurse
felt morally obligated to help a patient in his struggle if he
said he wanted out. She said, When he tells me that I want
to die, it is at this point that I would go right up to the
doctors and say, L et him die.
As we have noted, the research proposal itself may disclose
to the patient that he will die, and the change of awareness
sometimes induces depression (see Chapter 8 ). In part, how
ever, the proposal reinstitutes the recovery goal, and this carries
a lift for patients whose awareness has just been transformed,
as well as for those who are already aware. Indeed, an apparent
reprieve often does occur in the first days or weeks of the
experiment: tum ors go down, transplants work, the progress
of the cancer is halted, and so on. But often the new treatment
eventually fails, the patients awareness is transformed back to
the nothing more to do phase, and his depression is all the
worse because he has lost a renewed hope. Again, he must go
through the process of responding to the disclosure of fatal
illness outlined in C hapter 8.
194
PART
PR O LO N G IN G L IF E
A part from the needs of teaching and research, several
conditions encourage unduly extending the life of patients in
the nothing more to do phase. Some of these conditions
preclude changing the goal from recovery to comfort, so that
the doctor continues to strive for recovery even though he
knows that death is very certain. O ther conditions perm it a
change in goal from recovery to comfort, bu t still generate
efforts to prolong the patients life.
Change o f goal and time of death. Knowing when the
patient will die, after he has reached the nothing m ore to
do phase, is crucial for the doctors decision as to whether he
should begin a treatm ent that may prolong life indefinitely and
therefore excessively. If he is sure when the patient will die,
he is less likely to prescribe a treatm ent that will prolong
life even for a short time, preferring to let the patient die in
the natural course of the illness. One doctor told us that
if a condition, symptom, o r disease is secondary to the m ajor
problem, then it will not be treated if the effect might be to
prolong death, which is certain and close by. Thus in a
case of metastic cancer, a patient developed pneum onia in the
two days preceding her death. The doctor did not treat the
pneumonia, suspecting that to do so might prolong her life
and assuming that the new illness might make her go a little
quicker.
If the doctor is not sure when the patient will die, however,
he is likely to order treatments blood transfusions, antibiotics,
an operation that may extend the patients life. W ithout a firm
expectation as to time of death, he is usually not willing simply
to wait and let the patient die, since unless something is done,
the patient may live on indefinitely in much greater discomfort.
Com fort surgery may be performed to allow the patient to
eat, or to breathe more easily; antibiotics or blood transfusions
Nothing More to Do: The Problem of No Recovery
195
m ay be given to improve the patients over-all physical condi

tion; and equipment may be used to help keep him alive.
Striving for comfort may thus prolong the patients life, some
times with much more comfort, but sometimes, too, with in
creased suffering near the end. The difficult problem is to
decide whether, in giving the patient m ore time, the likely out
come is more comfort or more suffering. In practice, the
criterion is clearly that buying time is worth it if you can
m ake them comfortable. A surgeon in one hospital became
famous for perform ing the comfort miracle. H e operated,
however, mainly on patients who were expected to linger on
painfully, not those who were expected to die shortly.
A m ajor social source of uncertainty about time of death,
which in turn encourages prolonging, is dissension among the
doctors involved in the. case. All may be certain that the patient
will die, but the attending, private and resident physicians may
differ on the anticipated time. Aware of these differential ex
pectations, the doctor on the spot when bleeding starts, for i
example, may feel obligated to prolong the patients life even
though he believes that she could go anytime. If his colleagues
expect the patient to live for some rather considerable length
of time, he may doubt his own expectation or fear an accusation
of negligence. Disagreement with experienced nurses can also
create this doubt. The immediate social influence of relevant
experts can force adherence to the ideal medical goal of
sustaining life, though this goal is conventionally ignored in
most such cases.
Change of goal and closed awareness. In a closed awareness
context, relaxing efforts to sustain the life of a nothing more
to do patient very nearly constitutes an abrupt disclosure that
death is imminent. As we suggested in our discussion of do
something care, the doctor may prefer to keep the patient
of family members unaware for the time being, to avoid the
problems that such a sharp blow a few hours or days before
death might create. Although he has altered his goal from
196
PART
recovery to comfort, if the patient takes a turn for the worse

the doctor will continue to act as if he could still save him,
and thus prolong his life. The doctor will treat the pneumonia,
give blood, perform the tracheotomy, and so on. This gives
him time to bring about a more orderly transition to awareness,
after which he can let the patient die according to the normal
course of his condition.
In gaining this time, a doctor not only avoids unpleasant
scenes associated with a sharp break into awareness, but he
also protects himself from accusations of negligence by demon
strating that he did all he could. This demonstration is, as
we have said, one of the principal functions of do something
care, which is often necessary to soothe unaware family mem
bers who might otherwise contemplate a malpractice suit after
an abrupt change in their awareness followed by the patients
fairly quick death. To them it may have seemed that the doctor
just let the patient die. And as we have emphasized, all
parties are better off if both family and patient are made aware
of the impending death early enough to give them time to
prepare for it. Therefore, prolonging the patients life extends
somewhat the time they have to prepare and softens the blow
when the doctor breaks the news to them. A nd the doctor is
also allowing himself time, perhaps to reappraise the patients
condition, before he drastically revises the awareness context
by announcing that death is imminent.
An orderly transform ation of family awareness involves
more than one stage of interaction, taking all the extra time
th at the doctor gains by prolonging the patients life. Before
approaching a family member, he may talk with the nurse to
see whether she agrees that they should let the patient die.
She usually assents implicitly, without making an explicit
statement: she suggests that the doctor speak to the family
about it. This is a way of saying that she will cooperate by
helping the family handle the disclosure of imminent death.
It also indicates to the doctor whether the family is, in the
nurses judgment, ready to receive the news, and even whether
Nothing More to Do": The Problem of No Recovery
297
a family member can be approached with a request to permit

the patient to die an action which may further reduce the
doctors burden of responsibility. From his discussion with
the nurse, the doctor appraises the support he can count on
from nurses and family, and what he must do on his own with
some risk of having people turn against him. This information
enables him to manage a more orderly transition to awareness
than would have occurred had he simply let the patient die,
with a consequent shocking transition to awareness for the
family.
Change of goal and open awareness. When everyone is
aware that there is nothing more to do, the goal of recovery
has, in effect, been changed to a goal of comfort. This is the
ideal situation for ceasing to prolong a dying patients life,
but several conditions still tend to prevent letting him die.
The doctor m ust decide on the means by which a patient is
to be eased out of life, and whether it should be done covertly
or with the explicit collaboration of others.
The means by which a patient is allowed to die depends on
whether his life is being prolonged naturally or by equipment.
If the patient is lingering on simply as a result of his natural
condition, then the basic m ethod for letting him die is not to
resort to heroics, such as giving blood or using equipment if
he should take a turn for the worse. The patient is simply
eased out of life comfortably with pain-killing drugs. Withholding
heroic treatm ent is by and large the only control the doctor
has over this form of prolongation. The patient must be given
basic sustenance You cant let him starve to death but
not emergency treatment.
The medical ideal, of course, is to prolong at all costs, but
according to one doctor, The family will never say Keep him
alive as long as possible. In cases of extreme pain or deteri
oration, this condition may influence the doctor to help the
patient die sooner than he would have naturally. Under very
loose orders from the attending doctor, the resident doctor or
198
pa rt t h r e e : p r o b l e m s o f aw areness
the nurse performs an invisible act. The standard one is

to m anipulate the patients narcotics, giving either an overdose
or a prem ature dose. By this time, negotiations with both nurse
and family have been carried out, to insure that letting or
helping the patient die will have no adverse consequences for
staff members. Normally, these people feel relieved that the
patient is out of his misery; often all parties, even including the
patient, are involved in the decision one reason that the
invisible act is seldom jeopardized by publicity.
In a few cases family members may insist that the patient
be kept alive by any means available. This insistence usually
arises from two conditions: painless dying and a reasonable
expectation of a cure from research. Provided the physical
deterioration is not excessive, painless dying allows the patient
and his family to await a cure with hope not trampled by
misery. But if the research hope becomes illusory, the doctor
or a nurse may still perform the invisible act, concealing it
with attem pted heroics or a belated acknowledgment that the
patients condition had deteriorated.
Allowing a patient to die who is being kept alive by equip
ment poses different problems. In one sense, equipment gives
the doctor much greater control over the patients life than
he has with natural prolonging; he also has more time to plan,
to calculate risks and negotiate with the family and nurses. A t
this stage nurses invariably favor letting the patient die, real
izing, of course, that they do not actually have to pull the
cord. Only the doctor has this power, preferably with the
familys permission. If the family is unwilling, and can afford
the cost, the patients life can be prolonged for some time;
but when the money runs out, there is great pressure on the
family to allow the equipment to be stopped. A t this point
even fhe hospital administration may becothe involved. In one
case a daughter had paid everything she had; when she was
out of money, the hospital wanted to send her father to another
hospital where the equipm ent was available free of charge,
although it was obvious that the patient would die if moved.
199
In the face of financial ruin, if for no other reason, a family

mem ber usually consents to stopping the equipment. Otherwise
the staff may carry out the invisible act. According to a
nurse in one denominational hospital, this is bolstered by the
rationale: The church says we are under no moral obligation
to use extraordinary means to keep the patient alive. (E quip
ment is considered extraordinary.) The invisible act is also
rationalized by the staff m em bers desire to avoid letting an
aware family member, unable to make a judgment under stress,
use up all his money for nothing, that is, for no good medical
purpose.
Family members in this stressful situation often cannot
decide whether the patient should be allowed to die at any
given time, even when they are aware that death m ust occur
in the near future. But when a relative is aware he usually must
be consulted. The doctor may have to wait until the relative
has reached an advanced stage of grieving and is prepared
sufficiently to discuss the matter. This may take so long that
again the doctor is forced either to prolong the patients life
unduly or simply let the patient die on his own responsibility.
This is still another situation in which it is im portant to the
staff to have available a strong family mem ber who can make
decisions on a rational basis during a time of emotional crisis.
A nother condition that tends to encourage extending the
dying patients life is the rule that next-of-kin must be notified
when the doctors are sure that the patient will die soon and
there is no hope. If relatives live m any hundreds of miles
away, the doctor is under pressure to prolong the patients life
until they arrive. In this instance, adherence to the standard
rule of notification, thus creating open awareness for the family
member, may have adverse consequences for the patient and
the staff members who m ust watch him. If the doctor knows
of this possibility beforehand, he may evade these consequences
by delaying notification and leave the closed context intact
until after the patient is dead.
200
PART
N o change o f goal and discounting awareness. U nder cer

tain conditions the doctor does not change his goal from
recovery to com fort even when he realizes that there is nothing
more to do for the patient. Instead, he discounts his own
awareness in favor of other purposes or regards it as an illegiti
mate basis for action.
The inertia of intense heroics sometimes leads a doctor to
continue an all-out effort to save a lost patient even though he
is dimly aware that there is no hope. This kind of discounting
can easily be reversed by a nearby colleague or nurse who is
not too intently involved to observe that the only possible
outcome is death. In one emergency case a patient who was
not very sick, and certainly was not expected to die, suddenly
had a seizure and stopped breathing. The doctor in charge
made a heroic effort to revive him with chest massage, but
finally another doctor asked gently, Do you want to continue
that if his pupils are fixed and dilated? This query brought
the intense doctor to his senses; he said, No, I guess not,
and stopped the chest massage. In this case the inertia was
particularly strong because death was unexpected. Although
the doctor stopped, he still could hardly believe what he was
quite aware of now H e didnt seem that sick; he wasnt
that sick; what was wrong? H e went on to ask the family
for a post mortem, and in addition engaged in a lengthy verbal
post mortem with his colleagues to figure out why he had
not been able to recover the patient.
Only half believing that the patient is lost, a doctor may
feel that, having stopped attempting recovery, he must wait
at the patients bedside until either he dies or there are indi
cations that he will recover after all. One doctor turned off
a respiratory machine because the patient was no longer
breathing. H e realized there was nothing more to do and
stopped recovery efforts but he was not yet ready to give
up all hope of saving the patient. So he waited, alone, by her
bedside for two hours to validate his awareness by seeing her
201
die. M ore typically, however, the doctor trusts his judgment

that a case is hopeless and returns to the patients who can
recover, leaving word with the nurses to notify him should the
unlikely reversal occur.
Although the medical ideal of prolonging life at all costs
and with all possible facilities conflicts with the awareness
that prolonging is useless and unduly painful, the ideal often
wins out. As one doctor said, There are too many instances,
in my opinion, in which patients in such a situation are kept
alive indefinitely by means of tubes inserted into their stomachs,
or into their veins, or into their bladders, or into their rectums
and the whole sad scene thus created is encompassed within
a cocoon of oxygen which is the next thing to a shroud.
Prolonging based on this medical ideal most often occurs
in particular hospital contexts and with particular doctors. The
atmosphere in research medical centers, whether foundation,
university or institute, is liable to stimulate more than the
usual zeal to prolong life. Jewish community hospitals also
seem to have an intensely saving atmosphere that often
produces undue prolonging. One doctor said of such hospitals,
Intense hospitals make for intense doctors and intensive care.
Intensive care units in any hospital, with their atmosphere of
heroic recovery, are geared to prolonging at almost all costs.
Equipm ent is clustered near every bed, so that prolonging
may be just a m atter of leaning over and hooking the patient
up. The atmosphere in these wards counters staff attempts
merely to make the patient comfortable and let him die. In
contrast, other hospitals (e.g., C atholic) de-emphasize the ideal
of prolonging at all costs by discouraging extraordinary efforts
unless they are deemed clearly worthwhile.
Well aware that circumstances like these may force dis
counting the verdict of nothing more to do, a group of doc
tors in one research hospital have solemnly shaken hands
and agreed that under no circumstances will we keep treatment
going indefinitely if one of us has an overwhelming stroke and
is being kept alive by tubes. The doctor reporting this added,
202
PART
This is not only true of us; it holds true for many groups of
doctors. These doctors are trying to ensure that awareness
of nothing more to do will prevail over the strongly em pha
sized and organizationally supported medical ideal that life
must be sustained at any cost. They have established an
invisible system to assure that the invisible act will be
performed.
Doctors, particularly the younger ones, often cannot bear
the thought of losing a patient, which comes with recognizing
that there is nothing more to do. Fired with the ideal of saving,
they put patients on life-prolonging equipment, in the vain
hope that they can achieve recovery after all. They search the
literature for clues to new treatments. (A young doctor in a
hospital in San Francisco found such a treatment, saved his
patient and became famous for it.) F o r the m ost part, however,
the effort is of no avail, and eventually the doctor realizes he
is responding to a deeply internalized ideal that can no longer
be fulfilled. Accordingly, he changes his goal and ceases pro
longing. One doctor rescued a patient who suddenly hem or
rhaged but la te r. recognized his lost opportunity to let the
patient die: If I had given it a second thought, I wouldnt
have saved him . This also occurred to a nurse who had
automatically inserted a falling tube but wondered, immediately
afterwards, why she had done it.
Several other conditions, linked with the medical ideal
of prolonging, result in discounting awareness of nothing more
to do. It is often very hard just to watch a patient die. The
sense of helplessness also a consequence of the medical ideal
may cause a doctor to reinvoke the recovery goal and try
to prolong the patients life, hoping to save him later. F or the
moment, at least, this reduces his frustration and helplessness
at being unable to do anything effective. Some doctors feel
that a patients death puts their professional competence in
question, rather than the state of medical knowledge generally;
and as a consequence they will not relinquish the recovery goal
until they feel assured (often after reminders from colleagues
203
and nurses) that their own skill is not at issue. A favorite

rationale of the doctor who persists in the ideal of indefinite
prolonging is an assertion that he is simply an instrum ent of
society. H e feels that society has vested in him the duty of
sustaining life and that he is therefore obliged to abdicate per
sonal responsibility for judging the advisability of whether a
particular life should or should not be prolonged.
Nothing More to Do: The Problem of Comfort
12
Nothing More to Do:

The Problem of Comfort
In the previous chapter we spoke of nurses primarily in relation
to doctors efforts and activities. Now we shall focus on what
the nurse does on her own when she becomes aware that
there is nothing more to do for the patient, and changes
her prim ary goal from recovery to comfort.
W hen the patient is expected to recover, his prim ary source
of care is the doctor who specifies and supervises treatments.
But when the patient is in the nothing more to do phase and
comfort is the only goal, the nurse takes over as the main
custodian of care. It is her responsibility to insure patient com
fort within the limits of the doctors orders for sedation, meals,
privileges, treatments and such. Of the many problems this
responsibility poses for the nurse, we shall consider only those
linked with the awareness of the parties involved in the dying
situation. We shall discuss the nurses transitions into awareness
of nothing more to do ; how she then insures comfort; and
when painless comfort is no longer possible, what she does
about the collective mood that develops among nurses to let
him die.
T H E N U R SES TR A N SITIO N TO AW ARENESS
The doctor may tell the nurses that a patient is dying, but
usually not that he is in the nothing more to do phase of
dying. The doctor withholds this information for several
reasons. To the nurse involved in the case, the change in the
patients condition may be so obvious, as when he suddenly
204
205
loses consciousness, that the doctor does not need to say any
thing. Even if the nurse can see no change, she can tell from
the doctors slackening of effort that there is nothing more to
do. This meaning is clear when, for example, the doctor sud
denly stops blood transfusions. Or, the doctor may post the
patients name on a critical or term inal list, so that this inform a
tion is routinely distributed to the nurses involved. A nd the
nurse spends so much more time with the patient that she
is very likely to witness this strategic change in his condition
before the doctor does, and rush to tell him.
As the nurse comes to realize that there is nothing more
to do, she adjusts her goal from recovery to comfort care.
One nurse expressed the goal change this way, You can still
make the patient comfortable and you can still entertain him
and you can still distract him . Thus, in relinquishing the
prospect of recovery she was planning comfort activities. As we
noted in the previous chapter, however, losing the chance to
achieve the prim ary medical goal of recovery, and with it the
highest nursing reward, often results in less involvement and
effort in patient care.
Because this reduced effort is such a common reaction
among nurses, a doctor may again be reluctant to tell a nurse
that there is nothing more to do, if she is not already aware.
His purpose is to keep the nurse alert to recovery as long as
possible, until she becomes aware on her own and starts
thinking in terms of routine comfort. If the nurses awareness
is on the level of suspicion because the doctor has not validated
her suspicions, she is likely to remain doubtful and alert for
counter-cues, still hoping for recovery. H er tendency is to wait
and see; as one nurse put it, If he comes out of it, well work
on him. He only has to give us the slightest cue. Should the
doctor stop working himself, a nurse who still thinks there is
a chance may be horrified and blurt out, Do something,
doctor! A nurse who . is sure the patient is in the nothing
more to d o phase might hesitate to continue working, for
her efforts might result in painful prolonging. F o r nurses who
206
PART
are committed to and responsible for painless comfort, pain is

legitimate only in the service of recovery.
Nurses are judged by both superiors and doctors according
to the effect the change in goals has on their work. Those
nurses who tend to work as hard or almost as hard after the
news is out may be assigned to patients who need com fort care.
As one doctor said, I always try to get her for patients who
cant be saved. She still works so hard to make them com
fortable. Shes wonderful. So far as this appraisal of the nursing
staff is the basis for a clear-cut division of labor, it tends to
balance the drawbacks of reduced challenge, involvement, and
effort. Some nurses volunteer for comfort care, while others
avowedly find they cannot do it because they give up.
In many instances, as mentioned, the nurse becomes aware
there is nothing more to do before the doctor does, which, as
we shall see presently, creates a num ber of additional problems.
It may be a nurse who m ust get the doctor to post the patient
on a seriously ill list, to notify the family, to order more
sedation, or to stop prolonging the patients life; and it is
her job to reach him outside the hospital to find out what to
do when a patient suddenly turns worse. In sum, the differ
ential death expectations between doctor and nurse guide the
transition into awareness of nothing more to do, and the
subsequent interaction and purposes of each party.
GOO D W ILL C O M FO RT
In the previous chapter we reviewed some of the extraordi
nary privileges granted to the nothing more to do patient
to make him more comfortable. To the privileges ordered by
the doctor, the nurse can add others within her power to
bestow. She, too, may excuse the patient from having baths,
being turned in bed, or taking walks; she may release his
constraints or give him passes and better meals. Com fort can
also be achieved through exceptional attentions, or good will
care. The nurse puts herself out to chat with the patient, helps
207
female patients with hair combing and makeup so necessary

for feeling good and for family visits keeps the room cheerful,
spends time with the patient for companionship, and lets the
patient set the pace for much of their interaction, whether
routine or not.1 F or non-routine interaction she is ready to
comply With the patients requests. F o r routine care, giving
the patient a measure of control over the activities linked with
hospital and medical routine is in itself an act of good will. Of
course, in situations such as those in wards for aged in state
hospitals, the small num ber of nurses and aides in attendance
over many patients makes it virtually impossible to give the
patient much control of this kind.
In closed awareness contexts, the unsuspecting patient
enjoys this apparently beneficent attention and the leeway in
his hospital life, but good will care, like extraordinary
privileges, can also be a reliable cue to an unmending patient
that something is awry. Thus, a suspicious patient is likely
to respond by badgering the nurse with questions about the
implications of good will attention, and she must control his
assessment of these implications (as discussed in C hapter 3 ).
In the open awareness context, responses to good will
care depend on the way the patient is taking his dying. The
patient who is preparing himself for death may feel that good
will care is his due, since his status is exceptional. He may
simply enjoy the attention, or he may interpret it as a hypo
critical compensation for not really being able to help him.
1 Letting the patient set the pace of interaction is a gift of control
over his situation to the patient. If it is true, as Fox says, that the
sick person is expected to cooperate with his physician, and in this way
do everything he can to facilitate his recovery, then the pace of
interaction is, to begin with, mostly under the control of the staff, and
the patient must cooperate with the pace offered him in treatments,
comforts, baths and so forth. Thus, to offer the patient some control
over this pace of interaction is indeed a gift from the nurse, for it is
a relinquishment of a power the nurse has to begin with and need not
give up, but may, since she no longer expects the nothing more to
do patient to facilitate his recovery. Pace of interaction is an im
portant category for the study of how interaction is controlled, a prob
lem of traditional concern in social interaction theories.
208
PART
A patient who denies his own dying may refuse any attentions
that segregate him from typical or recovering patients, while a
withdrawn patient ignores them just as he ignores all efforts to
pull him out of his withdrawal. Nurses who go out of their
way to engage in good will care are distressed by refusal
or disregard, for the loss of contact reduces them to helpless
ness. They can then give only rudim entary forms of comfort
(perking up a pillow ). There is almost nothing to do until
the patient allows himself to be reached again. Withdrawal
is the patients most extreme control over the pace of inter
action with nurses.
PAINLESS C O M FO R T
M ost of the nurses energy is concentrated on maintaining
painless comfort for the patient. While painlessness is a goal
that is desirable for all patients, it takes on added importance
for a patient who has nothing left to hope for but painlessness.
F or such a patient, lethal risks to insure painlessness through
narcotics are legitimate, though they would never be taken
for patients who can be expected to bear pain in the service
of recovery.(The possibility of addiction is now irrelevant be
cause the dying patient can never be taken off the narcotics.)
Painless comfort is a goal strongly emphasized by doctors,
family, nurses and other patients (as well as the patient him
self), for few people can stand anothers pain. If a patient
is in too much pain, everyone is upset by the scream at the
end of the hall, which threatens to disrupt the routine work
as well as the staffs collective moods and feelings the senti
m ental order of the ward. Thus, the nurse is under considerable
social pressure to control the patients pain successfully.
Differential Death Expectations. To accomplish this task,
a nurse m ust have adequate sedation orders from the doctor.
How adequate the doctors instructions appear to the nurse
depends on their differential appraisal of the patients stage of
209
dying. If both doctor and nurse are aware that the patient is
in the nothing more to do phase, then the doctors instruc
tions are very likely to allow the nurse to use her own discretion
in the dosage necessary to keep the patient free of pain. She
may even risk lethal dosages. None of the (nurses) believe
in euthanasia, but its just that as you give these heavy doses
of narcotics you think that this may be the last one he can
take. One nurse reported that her instructions gave her a
leeway that would allow the patient to die in the line of her
duty. The implicit instruction was: Give [this degree of medi
cation] to the patient e v e ry
hours until dead.
If the doctor is aware that there is nothing more to do
for the patient, and gives very loose instructions to a nurse
who is unaware of the patients condition, she may be fright
ened by so much discretionary power. She will cover herself
by phoning the doctor to check dosages or to request lower
dosage orders, checking with the head nurse, sticking as close to
the order as its looseness permits, and so on. She fears that if
anything happens to the patient she could be accused of negli
gence in her interpreting of the loose orders. She also fears
loss of self-control when faced with pain that she knows she
can within the limits of her orders blot out with a heavy, but
potentially lethal dose.
A nurse who realizes somewhat before the doctor does
that the patient is either in, or soon will be in, the nothing
more to do phase, faces other interaction problems. If the
doctor, in preparing to leave the ward, has not given her
sufficiently loose instructions to ensure painless comfort, she
must act fast. She will have to corner the doctor and either
remind him that he has left no instructions or suggest to him
possible changes in old or current instructions. Saying she
might need them, and asking what she should do, she prods the
doctor into giving leeway. If this tactic fails, or if she is too
shy to try it, two alternatives remain. One is to engage in
invisible acts of increased dosage: some nurses, particularly
private nurses, do take m atters into their own hands. The
210
other alternative is to try to reach the doctor by phone when

the patients pain becomes uncontrollable under present orders.
This is one condition under which the nurse and her supervisor
will m ake every effort to reach the doctor, no m atter where
he is, what h e is doing, or what time it is. Having reached him,
she tries to bring the doctors awareness into line with hers
by pointing out changes in physical condition, indicating that
there is nothing left to do but insure painlessness. Then she
requests a change in sedation orders.
Patients Awareness. The patients part in the means the
nurse uses to insure his comfort depends on the awareness
context. A nurse may ease the transition to an open awareness
context by promising painless comfort until death. A typical
sequence of interaction is the following: The doctor tells the
dying patient that there is basically not much more to do for
him medically except to control his pain. This announcement
immediately leads the patient to develop tactics for securing
sedation. Normally, a patient in this stage does not worry about
becoming an addict,2 or about becoming immune to any one
painkiller. He is concerned only with minimizing his pain, and
he starts by trying, by various interaction tactics, to control
the nurses management of it. The pressure of this attem pt often
makes a nurse wish she were maintaining comfort under the
cover of a closed awareness context; for then the patient, still
motivated to recover, will put up with some pain and leave
it up to her to manage it.
Furtherm ore, the announcement by the doctor usually leaves
the patient very depressed. To cope with this initial response,
the nurses stand by with medication to relieve emotional as
well as physical pain. As one nurse put it, If no one had
helped him that day I think he might have committed suicide;
he was in an awful state. We gave him m edication. And
thus he was immediately introduced to the looser regulations
* Unaware cancer patients do worry about addiction, but patients
who are aware they are in the nothing more to do phase tend to ignore
it since the pain will not cease, death is inevitable, and therefore nar
cotics must be taken to the end.
Nothing More to Do": The Problem of Comfort
211
surrounding sedation. He would no longer have to put up

with as much pain; he could depend fairly well on the nurses to
insure comfort.
Most often, the patient wants the nurse to increase dosage;
and a common tactic for this is using verbal indicators of pain
moans, complaints, and so on. The nurse m ust judge whether
the patient is in physical pain and, if so, to what degree. This
task is difficult since the indicators of pain are often poor or
unreliable: a patient who says he is in pain may be expressing,
to one degree or another, only emotional pain. Although a
disturbed patient may have to be tranquilized, a narcotic should
not be wasted on emotional pain, for its effectiveness, which
declines with tolerance, must be reserved for physical pain often
over a long period of time. Thus, when a patient uses moaning
and groaning tactics to obtain m ore sedation than ordered, the
nurse must use countertactics to test him if she has reason to
doubt the physical origin of his pain. She may give him a
placebo or tranquilizer; she may walk past his room a few times
to see whether the moaning and groaning occurs only when
her footsteps are audible, and so on.
M aneuvering of this kind tends to result in irregular dosages
and fluctuating levels of pain for the patient, which in turn
may make him wish to control the scheduling of injections
and dosages. Some patients dispense with manipulation of
verbal indicators and proceed directly to try to establish control
over schedules. The patient on a periodic (rather than dem and)
schedule may concentrate on trying to shorten the time between
injections. H e may act confused in his time perspective, ringing
for the nurse at the three-and-a-half hour point and asking for
another shot on the grounds that his four-hour interval is up.
O r he rings for the nurse after two hours to say: Youll
be sure to be on time for my next injection, wont you? thus
trying to control the schedule by enforcing it. He may also
get family members to keep close watch for transgressions.
Patients also try to negotiate for a different schedule on the
grounds that the regular one is not adequate.
Patients on dem and schedules depend primarily on their
212
PART
handling of indicators of pain, since dem and means when

the nurse judges the patient needs it. Such a patient may also
stick his hand out for medication every time a nurse passes.
If these tactics do not work, he may threaten suicide or try
to negotiate for a regular schedule, or he may go through
other staff members to put pressure on the nurse. H e can ask
the doctor to order the nurse to give him regular or more
frequent sedation, or he can enlist the aid of family members.
A family mem ber can create a status-forcing scene, saying it
is the nurses job to relieve the patients pain, but she is, it
appears, negligent in this duty. (Needless to say, anxious family
members often do this without prom pting.) Lastly, a chaplain
may be asked, as a neutral, to institute a conference of patient,
nurse, doctor and family to arrange a sedation schedule that
is satisfactory to all.
The dem and schedule can be very demoralizing to the
aware patient, since it necessarily means more pain for him.
Because the nurse is watching for symptoms, the patient must
withstand some pain before it is reduced by her action. This
pain may m ake the patient think I am surely going this
time. On the other hand, a patient who finds himself support
ing longer and longer periods without sedation may think he is
improving and feel hopeful, though he is experiencing only
a tem porary reprieve. His pressuring the nurse to increase or
reduce sedation will be guided by these conditions linked with
the demand schedule.
We have described the typical situation but not the in
frequent patient who is aware that death is certain but refuses
narcotics, perhaps to keep clear-headed enough to arrange his
affairs before the family takes over, or perhaps merely to
demonstrate his prowess. (These patients too, however, are
often given narcotics in spite of their wishes, if their pain
disturbs others, for the nurses job is to protect the senti
mental order of the ward as well as to preserve the hospital
routine. This is one instance in which patients are not allowed
to manage their own dying when it interferes with the social
organization of the hospital.)
213
To reduce dosage, patients may use a no complaining

tactic. Then the nurse m ust watch for other indicators of pain
since, in the interests of others on the ward, she cannot let
the patient support so much pain that he breaks down or sets
a bad example of nursing care for other more fearful patients.
The goal of painless comfort m ust be pursued, but under these
conditions it is very difficult to achieve, for when a patient says
very little it is hard to ascertain how much a given dosage
reduces his pain.
Cant Talk. A patient on a dem and schedule increases
the pressure on the nurse if his physical condition deteriorates
to the point where he can no longer ask for sedation because
he is snowed by sedation. This problem applies particularly
to research patients who are kept on a dem and schedule for
painkillers, to minimize narcotics interference with the study
drug. As one nurse said, They are kept under sedation, but there
is this carelessness. They dont give medication unless its asked
for and these patients often arent in a condition to ask, yet
I am sure they are in pain. Nurses use several methods to
establish degree of pain: turning the patient to watch his
response or asking him to blink or move his hand if he is in
pain. Sometimes the nurse talks with the patient about this
stage of dying before it arrives, either on her own initiative or
because the patient negotiates with her. In one nurses words,
the response is: We always reassure them that we will give
them whatever they need to keep the pain under control. This
promise, however, ignores the fact that when verbal contact
with the patient is lost, indicators of pain are harder to read
and verify. Thus, We sedate heavily in the end, a nurse told
us; this is one way of playing it safe.
U nder this cant talk condition, before death, the nurse
is relieved of several pressures emanating from the patient. She
no longer needs to explain, or parry the patients request
for an explanation, every time she changes a periodic schedule
or after each demand-schedule injection. As he approaches
death, a patient on a periodic schedule puts less and less
214
PART
pressure on the nurse, when a regular increase in dosage is

built into the periodic schedule. This causes the patient at
this stage to be either too snowed by narcotics to talk effec
tively, or not to be able to talk at all. Thus, the nurse no longer
has to worry about avoiding talk of death, nor does she have
to listen to Why dont you let me die? or F o r Gods sake,
let me die, or D ont let me die, nor m ust she handle his
negotiation for more narcotics. In the last phase of dying,
nurses are particularly vulnerable to the patient who begs for
help, especially for painkillers. They want to provide more
sedation as a way of doing something within their power
in order to relieve their own feelings of medical helplessness.
It is also very difficult for them to take a patients indeterminate
state of pain; and one way to control the am ount of pain is to
sedate heavily.
Some of the benefits otherwise inherent in the open awarenes context are lost when the patient can no longer talk. From
the patients point of view, he loses whatever control he had
established over insuring painless comfort, and over manage
ment of his dying, when he loses the ability to negotiate
for more narcotics. One advantage of open awareness is that
the patient can take leave of his family, but when he is heavily
sedated this advantage is lost too. (In the words of one nurse:
Theyre snowed and cant talk to their relatives. ) Even if
the patient can communicate, heavy sedation makes him men
tally vague, which tends to disturb relatives who wish to make
final farewells. Patients in this condition are also unable to
sustain useful relations with a chaplain or social worker. In
short, the aware patient who cannot talk cannot interact very
well even on the basis of open awareness.
L E T T H E PA T IE N T D IE
When painless comfort is no longer possible for the dying
patient, nurses are reduced to a state of helplessness, which
one nurse expressed as follows: You cant do anything, even
"Nothing More to Do: The Problem of Comfort
215
to relieve the pain in the end, sometimes, and the patients

keep wanting things done for them . 3 They can no longer
achieve the only goal of patient care that remained to work
for. As they watch a patient lingering or having his life pain
fully prolonged, a collective m ood develops among the nurses,
an urgent wish to let him die. This is their m ost frequent
response to their inability to provide comfort. (A less pre
ferred alternative is to put the patient into living sleep with
drugs a social death.) One nurse stated this let die alterna
tive to painless comfort explicitly: We cant relieve her pain,
everybody wants her to die for her sake, meaning her suffering,
we cant even help her. Once we cant help her, let her die for
G ods sake, for our sakes as well as theirs.
O ther nurses are less articulate but as direct: The quicker
she dies, the better without suffering. Why should he be
saved to go on living like a vegetable? Why should he be
saved for a life of continual suffering? M aybe it is for the
best, maybe it is a blessing if he goes. A nd in reference to
research patients, I think a m an should be left to die in peace
with plenty of narcotics. W hen the patient does die, the col
lective mood of the nurses changes to one of great relief. No
longer must they be frustrated by their inability to provide
comfort and by the waste and senselessness of undue linger
ing or prolonging.
To be sure, this collective m ood runs counter to the medical
ideal of prolonging life. A student nurse displayed clear recog
nition of this ambivalence: We had a patient, a cancer patient,
who wanted to come to the hospital. Everything was done
for him, even if he didnt want it. They gave him everything
3 Sheldon Messinger, in a private communication (September 13,
1963), gave us a comparative view of helplessness.
. . you mention
medical helplessness in passing. A most impressive phenomenon. Death
isnt the only place where it is found, although it may be among the
most salient. I remember an MD at N apa telling me that she used EST
(electric shock therapy) in a particular case not because she thought
it likely to do any good in the long run, but because the patient was
in pain and Im trained to help. I had to do something. What else was
there to do?
216
PART
to prolong, to save this person, if that is what you want to

call it. This isnt actually saving, but once you come to the
hospital everyone is concerned to preserve your life, the person
has no choice; well, it is just a question of ethics. She realized
that in the nothing more to do phase, prolonging is not
saving and can only be justified in term s of an ethic, which is
weak support indeed in these cases. Thus, m ost nurses feel
that let him die is the best alternative to painful prolonging,
though some remain ambivalent. One nurse, evidently wanting
to support the ideal, questioned her judgment that the patient
was really in the nothing more to do phase-: How do I
really know he is going to die; how do I really know; therefore,
how can I possibly let him die? A nother nurse resolved her
ambivalence by means of unquestioning compliance with the
ideal.4 She registered annoyance, indignation and irritation at
the thought of not prolonging at all costs as well as at the
thought of actually helping the patient die. O ther nurses regard
the ideal as a public shield over the invisible acts, or system
of acts, by which they let the patient die.5
Patients and Open Awareness. In the closed awareness
context the collective m ood of let him die develops spon
taneously among the nurses, regardless of the patients overt
demands. In the open awareness context, however, this m ood
may be supported and even initiated by the patients requests
to be allowed to die. A patients longing to die Please God
let me go is reciprocated by the nurses Its a blessing if he
goes. Just as a patients struggle to live encourages nurses
to help him, his struggle to die may arouse their sympathy and
4 The resolution of problems of work by using formal rules termed
rule tropism by Alvin W. Gouldneris an important aspect of nurs
ing care that is worthy of future research. See Alvin W. Gouldner,
Comopolitans and Locals: Toward an Analysis of Latent Social Roles
I, Administrative Science Quarterly 2 (December 1957), p. 299.
5 See Anselm L. Strauss, Barney G. Glaser and Jeanne C. Quint,
The Nonaccountability of Terminal Care An Aspect of Hospital
Organization, Hospitals, 36 (January 1964), pp. 73-87.
217
prom pt them to help him attain his objective. One nurse,

mentioned in the last chapter, stood by for a patients request
to die, so that she could step in and ask the doctor to let
him die.
Request is too mild a way to describe this behavior:
many patients actually beg and plead to be allowed to die.
A doctor has lucidly described this pattern, covering the two
essential conditions of nothing more to do and too much
pain: A cancer patient for which every conceivable avenue
of treatm ent has been explored with total failure, and this
patient m oreover is suffering from excruciating pain and is
pleading for release. 6 Patients who say, Please leave me
alone, I want to die, I dont want any tests like this, or Why
dont you let me die? confirm the nurses conviction that the
best way out is death. Statements like Im dying right now,
get me out of my misery, may express a transition from a
closed awareness context to an open one; the patient registers
a let me die m ood as he registers realization of his change
of status. Such a quick change in awareness context can bring
into the open the nurses collective hidden desire to let him
die, and prom pt their active assistance in achieving the mutual
goal. Too quick a transition, however, may cause an unaware
nurse to doubt the patients reading of his own condition and
consider him m orbid.
Family and Open Awareness. The nurses collective mood
may also be shared by aware relatives of the pain-ridden,
nothing more to do patient. They are often permitted to make
the final decision, but sometimes they may be forced to plead
(" Please turn that drip off ) if the doctor still considers pro
longing the best course. O f course, unaware family members
would be horrified by the nurses collective mood, and some
times even aware family members who encounter it create a
status-forcing scene to remotivate the nurses.
The ordeal suffered by family members at this time also
San Francisco Chronicle, June, 1963.
218
PART
th r e e : pro blem s o f
aw areness
supports the nurses collective desire to let the patient die. It

was torture, we felt m ore sorry for her [mother] than for him
[patient] at times. Nurses may shift the object of their involve
ment at this time (as we have suggested in C hapter 1 0 ): the
patient cannot be comforted now, but the family mem ber can.
As the new object of the comfort goal, the very much alive
family mem ber takes on the patient status, and the nurses, as
a result of this object switching, now have a goal to work
for and a reward to expect. This shift of involvement may, at
the same time, paradoxically reverse the nurses intentions to
let the patient die, when the only way to keep the family
member comfortable seems to be to comply with his desire
to postpone the patients death.
L ETTIN G T H E PA T IE N T D IE
In the last chapter we indicated how a doctor, with the
help of the nurse, may let a patient die; here we shall discuss
some means available to a nurse acting on her own for letting
the patient die. The use of these means depends on differential
death expectations between nurses and the doctor and on
awareness contexts.
Non-rescue Tactics. The tactics available to the nurse
depend on whether she agrees with the doctors evaluation of
the patients status. If she and the doctor agree that there is
nothing more to do, she can exercise non-rescue tactics to
let the patient die, without fear of reprisal from the doctor
and with confidence that he will protect her from family
accusations of negligence. Since she spends m ore time with
the patient, the nurse is more likely than the doctor to be in
a position to save or not save the patient from immediate
death with split-second operations should he turn worse.
She can replace a fallen tube, ring for help, give an injection,
massage his chest; or she can, if she chooses, let the patient
die. Lack of action is in itself consistent with the reduced effort
characteristic of this phase of dying, but it is hard to resist
Nothing More" to Do": The Problem of Comfort
219
rescuing the patient during the critical moments when action

must be taken instantaneously if it is to be effective, and there,
is little time to ponder more alternatives. Consequently, a nurse
who performs a split-second rescue maneuver may wonder
later why she went to the trouble to preserve or prolong a
patient in vain.
Several conditions lead a nurse to rescue even when she
would prefer to let the patient die. First, the medical ideal of
prolonging, however senseless it may seem to her, is too deeply
internalized to be overcome readily in the final act, especially
where there is no time to think. A nother condition depends
on the physical setup of the ward. In an intensive care unit,
for example, it is easy to follow the natural response and simply
to hook the patient up to the necessary equipment, which is
right next to his bed. On a medical ward, though, some difficulty
in requisitioning supplies or equipment quickly enough might
encourage a nurse to use non-rescue tactics. A third reason
is adm iration for a doctor who has made heroic efforts to save
the patient against low odds for his survival. In this situation
the doctor is more im portant to the nurse than the patient;
again her involvement shifts from a dying patient to a very
much alive person. In the words of a nurse who experienced
such object switching : I was so concerned in trying to suc
tion in everything. I guess why I did this is because he was
Dr. X s patient and he had worked so hard to save him, it
really upset m e. A nother contributing factor here was the
high degree of social loss; the patient was a young m an with
a good future.7
A nother condition pressuring the nurse to rescue, when she
can count on even the doctors support for non-rescue, is the
probable visibility of her failure to act.8 Non-rescue tactics
7 See Glaser and Strauss, The Social Loss of Dying Patients,
American Journal of Nursing, 64 (June 1964), pp. 119-121.

8 An aspect of nursing care needing future research is how the
hospital structures the visibility of a nurses work from specific parties
to it. On visibility see Robert K. Merton, Social Theory and Social
Structure (New York: Free Press of Glencoe, 1957), referenced on
page 645; and Erving Goffman, The Presentation of Self in Everyday
Life (New York: Doubleday Anchor Books, 1959), Chapter III.
220
PART
are easiest on night shifts, in single rooms, in restricted areas

such as premie wards (except during baby viewing tim es), or
on private duty. When her behavior is likely to be witnessed
by staff members not connected with the case (whose death
expectations presumably are more optim istic), or by family
members, then the nurse will rescue. When family members
are present much of the time, as they so often are during
the final stages of dying, opportunities for applying non-rescue
tactics are particularly scarce. Even if the relative is aware that
there is nothing more to do and wants to end the painful
prolonging, a nurse may be reluctant to let the patient die
in his presence as a result of her inaction. Linked with visibility
of her inaction is the time it takes the patient to die. If the
patient takes too long to die in front of a relative or unaware
staff member, the nurse is forced to rescue. She can do nothing
for only a few minutes, unless her behavior is unusually well
shielded from observation. If the doctor disagrees with the
nurse, maintaining that the patient is not in the nothing more
to do stage, she can risk a non-rescue tactic only if she
can avoid being seen for a long enough time, since as a nurse
she is expected to save any patient whose death is not con
sidered inevitable.9
In addition, several structural conditions facilitate the
nurses use of non-rescue tactics, apart from her agreement
with the doctor and the visibility of her actions. Hospital policy
may specify that extraordinary efforts especially the use of
equipment are not to be taken for patients clearly in the
nothing more to do phase of dying. General, legal and
medical rules as well as hospital rules bar nurses from using
such rescue devices as certain kinds of equipment, certain
9 Under what general conditions of work are professionals forced,
allowed, and helped to act unprofessionally in order to get a profes
sional job done is a problem in the study of professions worthy of
future research. Present study tends to focus on how professionals are
trained to act professionally, not how they are forced, allowed, or
helped to act unprofessionally for professional goals. This is an im
portant problem in nursing care. See Merton, op. cit., pp. 343-345, for
his notions on institutional evasions, which may help such research.
221
types of injection, or certain procedures such as open heart

massage. Thus, in some cases a nurse can withhold her own
actions with impunity, though she still m ust try to reach a
doctor.
Reaching a doctor can itself be used as a non-rescue
tactic. The doctor may have left no instructions to follow in
case of emergency; no resident may be in attendance to take
over (this depends on the hospital or the daily work shift).
Or, a resident on duty may be unwilling to go too far with a
private doctors patient, as may other doctors who happen to
be on the ward. Reaching the doctor responsible for the case
may be difficult or take too long for quick action, so that
inaction for a time is unavoidable. These conditions, of course,
might lead a nurse to use rescue procedures forbidden by the
rules, but she is forestalled since illegal efforts to rescue a
patient from death could jeopardize her position as a nurse and
even result in a manslaughter charge if she fails. A nurse may
risk an illegal rescue effort when she disagrees with the doctor,
and is less certain than he is about the inevitability of the
patients death. But when doctor and nurse agree that there
is nothing more to do, a nurse who cannot reach the doctor
in charge of the case has a perfect opportunity to ease the
patient gently to his death.
Sedation Tactics. W hen doctor and nurse agree th at the
best course is to let the patient die, the doctor may order cessa
tion of most or all equipment in use and give the nurse loose
sedation orders to insure painless comfort until the end. These
sedation orders allow the nurse much discretion and therefore
encourage her to use active tactics to help the patient die (as
opposed to the more passive non-rescue tactics). U nder these
conditions she can m anipulate sedation with little or no fear
that she will be accused of negligence. By m anipulating seda
tion, or its counteractants, she can ease the patient to death
when adverse conditions multiply and become too much for
all concerned: the patient is begging for death, he is deteriorating
222
PART
physically, pain cannot be controlled, the family are going

through an ordeal that is pushing them to the breaking point,
and so on.
Loose sedation orders give the nurse enough latitude to
administer potentially lethal doses in the hope of achieving
painless comfort. In this way she can help the patient die. One
nurse said, It may kill her, but why not let her go free of
pain? A nother said, None of the (nurses) believe in euthana
sia, but its just that as you give these heavy doses of narcotics
you think that this may be the last one she can take. If the
dose does turn out to be fatal, then the nurse can withhold the
counteracting stimulant. If it is not fatal, then she can increase
the dosage next time. And since she is following or stretching
the orders of a doctor who agrees with her assessment of
the patients fate, she need not fear accusations of negligence
from him, and she can be confident that he will protect her
should others raise questions about her interpretation of
orders. A nother factor encouraging this strategy is the patients
attem pt to control or increase sedation. The nurse can purposively yield to a patients moaning and groaning or dis
torted time perspectives so that she eventually is giving seda
tion oftener.
Non-rescue tactics depend on whether the nurses inactivity
during a crisis can be observed by others, but sedation tactics
are not nearly so contingent on visibility. A nurse can apply
the needle or give the pill in the presence of others, since it
is part of her job, and the patient need not be in an obviously
critical condition. Any adverse effects are coo subtle for most
observers to distinguish from those of routine medication. When
charting what has happened, however, the nurse generally must
be backed up by a doctors orders. Thus, a nurse who disagrees
with a doctor about the patients fate takes quite a risk in
manipulating sedation. She m ust falsify or avoid recording her
invisible act, and this is too drastic a deviation from pro
fessional standards for her to contemplate usually. Normally,
as we have indicated, a nurse in this situation will first try to
"Nothing More to Do: The Problem of Comfort
223
convince the doctor that the patients case is hopeless and

that letting him die is the sensible thing to do. They may also
talk with a relative to obtain the familys consent and support.
W hen a nurse disagrees with a doctor who is certain there
is nothing more to do and accordingly gives flexible sedation
orders, she may be alarmed, as we noted before, by the re
sponsibility implicitly given her for the patients survival. H er
tendency then is to seek more control in her orders, to prevent
her from going too far in striving for the patients comfort. Of
course, this is a problem only for a nurse who still believes
that the patient can survive.
Several structural conditions inhibit use of sedation tactics
to let the patient die. If the patient is a research subject, even
a doctor who agrees that he is in the nothing more to do
stage m ay not give sufficiently flexible orders to allow the nurse
much latitude. She is supposed to minimize sedation, so that
it does not interfere with a clinical experiment, and if she gives
overdoses on her own, the doctor is not likely to back her up.
She is expected to keep the patient going until the end of the
study.
In a closed awareness context, the patient may beg not
for death but for a reduction in narcotics, because he is afraid
of addiction. Sometimes unaware family members have the
same concern. If social and family affairs are still unsettled,
use of sedation tactics is often precluded by patient or family
request. And if a family mem ber is traveling from afar for a
last visit with the patient, sedation can be manipulated only
to prolong life while keeping the patient conscious.
Open Awareness and Auto-euthanasia. A nother tactic for
letting the patient die, which is possible in the open awareness
context, is to perm it auto-euthanasia. The prelude is a distress
ing scene in which a pain-ridden patient begs to die, while
the nurse has no way to help or cannot bring herself to help
him. But she may tacitly agree to let the patient take his own
life. She may even provide him with a setup for his suicide
224
PART
th r ee: problem s of
a w areness
by leaving pills at his bedside, removing his constraints, or

leaving him alone in his room or in a bathroom . The next
thing she knows is that the patient has pulled out all his
tubes, or is sitting in the bathtub with his throat slit, o r is
lying in the street below. Such attempts, though, usually take
so much time that another nurse will often discover and be
forced to rescue the dying patient.
Auto-euthanasia is, of course, also attem pted by patients
without a nurses surreptitious assistance. In one case a patient
refused to give the doctors or nurses any information on what
she had done to herself. Since her condition could not be
counteracted, she thereby insured her own death. Sometimes
the ward arrangements are such that patients are put in a
separate dying room (which is a more convenient place
for the patient to attem pt auto-euthanasia), and left unwatched
for comparatively long periods of time. Even when auto
euthanasia is anticipated, the ward may have no room available
with barred windows, for example, or staff shortages may pre
clude adequate surveillance.
Nurses understand and even sympathize with patients in
great pain who attem pt auto-euthanasia, since it is consistent
with their own collective desire to let him die. Their collective
reaction is usually relief, not m oral indignation or feelings of
professional incompetence. Just as letting the patient die seems
a realistic alternative for them, auto-euthanasia seems a realistic
alternative for the patient, as far as they are concerned. The
patients awareness of this alternative is a basic turning-point
in his passage toward death,10 for at this point he grasps the
most im portant implication of his condition that there is
nothing more to do and realizes that the medical staff is
merely managing this last phase of his dying as best they can.
When he also realizes that his physical condition will soon
prevent him from trying to manage his own dying, as well as
10 On turning points see Anselm Strauss, Mirrors and Masks (New
York: Free Press of Glencoe, 1959), Chapter IV. The essential idea
behind turning points is the awareness that I am not the same as I
was, as I used to be.
225
the staffs managing of it, he recognizes auto-euthanasia as

the only alternative to giving the staff complete control over
his fate.
Nurses are trained to motivate patients to recover, and
patients soon learn, if they do not know already, that they are
supposed to be willing to get well. But this goal, like many
other medical ideals, is inappropriate for a patient in the
nothing more to do phase. It makes more sense to motivate
him to die; that is, to get him to accept his impending death
and prepare for it. A nurse who is able to help motivate the
patient to die (m ost nurses we m et are not, however) can
teach him how to die gracefully. And this tactic may abet
the tacit permission for auto-euthanasia, for he may also learn
that it may be best, in the end, for him to manage his own
death, to arrange a painless, easy departure and forestall an
indefinite prolonging which, toward the end, he may be
powerless to prevent. In this way he can insure a shorter ordeal
for his family and a more presentable, undeteriorated physical
condition for the last farewells. Indeed, we have known active
people in the prime of life, many years before this stressful
turning point, to contemplate just such a plan when they ask
themselves, W hat would I do?
Awareness of the Nurses Composure
13
Awareness and
The Nurses Composure
Despite many adverse circumstances in the dying situation, a
nurse must make every effort to m aintain her professional
composure, for both her own work and the assistance she gives
other staff members depend on it.1 The combined requirements
of dying patients, their families, and the staff for appropriate
levels of involvement in and attitudes toward patients, and also
their kinds of collective moods and degrees of composure,
create and define an overall sentimental order in the ward
and each nurses composure is vital to maintaining that order.
Nevertheless, nurses can become very upset, and a nurses
great fear in the dying situation is that she might lose control
over herself.
While many structural conditions help a nurse m aintain
her composure, other conditions tend to break it down. To
cope with situations that threaten to destroy her composure, a
nurse m ust develop strategies to maintain it. Some strategies
are general avoiding dying patients, for example and there
fore appropriate under many different circumstances; other
strategies are m ore specialized. Some strategies involve inter
action; others consist of collective moods or patterns of be
havior. Some are standard and acceptable to all nurses, so
that anyone may use them ; others are developed privately and
may or may not be condoned by other staff members.
M aintaining professional composure is a complex problem,
in view of the multitude of disturbing conditions that nurses
1 The study of professional composure is, as yet, an unresearched
area in the sociology of professions.
226
227
encounter. To delimit the following discussion and integrate

it with the theme of the book, we shall focus on the conditions
and strategies related to the nurses awareness. We shall show
how the means by which a nurse maintains her composure
are largely determined by her expectations about the certainty
and timing of the patients death.
The general defense strategy by which a nurse tries to
maintain her composure, during the entire time it takes a pa
tient to die, is to manage her emotional involvement with him
according to her expectations regarding his death. This general
defense strategy is a developmental process, comprising a
progressive accumulation of strategies (pertaining to work, to
talk with the patient, and to collective moods that develop
among the nurses), which serve to reduce involvement in the
patient as he approaches death. By reducing involvement as
the ordeal increases, the nurses are less likely to become upset
and possibly lose composure; if they do lose composure, it
is easier to regain. O ur purpose in this chapter is to describe
this process.
DYING IS U N C ER TA IN
In this early stage, when nurses are not yet certain that
the patient will die, and do not know when this uncertainty
will be resolved, they perceive little need to protect their com
posure. They therefore do not avoid conditions of working and
talking that induce or encourage involvement with the patient,
though, during the later stages of dying, they will use counter
acting tactics in the same conditions.
Nurses have a general conception of the death ratio i.e.,
the proportion of patients who die on their ward. When
the chances are slim that any patient will die unexpectedly
( We dont have very many patients who die ), nurses permit
themselves to become involved quite freely, to the limit im
posed by the sentimental order of the ward. They do not need
to think in terms of possible death. On wards with a relatively
228
PART
high death ratio, such as cancer wards or intensive care units,

the prevailing sentimental order limits involvement more, as
a general defense against the higher probability that death will
occur. In short, the death ratio typical of the ward is articulated
in its sentimental order, determining the degree to which a
nurses involvement in a patients fate is considered appropriate.
A t this stage of dying the goal of recovery is in full force,
though some staff members may have begun to doubt that
recovery is possible. In working for recovery, nurses are apt
to become involved with a patient to the allowable limits of the
wards sentimental order, and sometimes to a greater extent
than is warranted by the real probability of his death. High
involvement is also more likely when the patient has a high
potential social loss value. A youthful patient, or one whose
occupation is prestigious or im portant, or whose family is
devoted to him, or who is gifted or beautiful, has a high
value and the slightest possibility that such a patient will die
distresses nurses and makes them work harder to save him.
For example, the thought that he might be dying caused one
nurse to become over-protective of a 15-year-old at this stage
of dying. Nurses are usually well aware of the loss a patients
death would entail, because at this stage they listen freely to
family members who discuss his value.2
Working on the evening shift also tends to induce involve
ment with patients at this stage of dying. The patients are
awake, and since the bulk of the days work is over, nurses
have more time for them. Only a few auxiliary personnel and
doctors are present to dilute the nurses attention. Few nurses
are on duty, so that although each nurse has more patients
to care for, each patient sees the same nurse more often. And
s In listening to family members they learn characteristics of the
patients social loss which serve to correct and to balance out apparent
characteristics of his social loss. See Barney Glaser and Anselm Strauss,
The Social Loss of Dying Patients, American Journal o f Nursing, 64
(June 1964), pp. 119-121. Occasionally, also, a nurse gets involved
with a recoverable and highly valued patient, but then meets him on
another floor at a later stage of his illness when she then finds it
difficult emotionally to disengage herself from the relationship.
Awareness of the Nurse's Composure
229
members of the family usually visit during evening hours. All

these factors combine to produce cooperative, talkative, in
volving relations with the patient and his family. Nurses on
the night shift, when the patients sleep, are least likely to be
come involved, but even these nurses have an opportunity to
involve themselves with dying patients if the hospital policy
is to rotate nurses among shifts.
Since the fear of not being able to avoid the topic of
death is minimal, conversations can be frequent and talk not
usually closely controlled at this stage of dying, so that nurses
may also become involved with patients through conversation.
Long-range future plans for work and family, to be carried
out when the patient has recovered, can be discussed. The
nurses feel free to motivate the patient to get well which
further involves them in his fate even if they feel slightly
dubious about his prospects. These conversations may also
increase the patients social loss potential, in the eyes of nurses,
since they learn more about him and his importance to others.
Nurses and patient can also discuss the latters physical condi
tion with little or no anxiety, since even realistic interpreta
tions are unlikely to lead to talk of dying. Should a nurses
interpretation of a physical symptom point to the possibility of
dying, the nurse can pass it off as inconsequential without
losing the patients confidence. Treating a symptom as unim
portant is such a standard tactic in all nursing care that it need
create no term inal suspicions on the patients part.
EX PEC TED R ESO LU TIO N TO U N CER TA IN TY

When the patients fate is uncertain, and the nurses can
expect certainty at no definite time, their collective mood
becomes unsettled. The situation is indeterminant: should hope
for recovery or concern for comfort be the appropriate stance?
B ut when they know the time when the patients fate living
or dying will be determined, this mood of indecision is re
lieved. Now they can begin to prepare themselves for the out
230
p a r t t h r e e : p r o b l e m s o f a w areness
come by managing involvement so as to m aintain their

composure while they await more definite news. They begin
by applying the standard strategies avoiding both the patient
and topics of dying and death as well as starting to develop
strategies that can be applied if the patients death becomes
certain.
Intense Work. The basic characteristic of this stage of
uncertainty is that it gives staff members a specified am ount
of time for saving the patients life. Though this stage of dying
may occur on all wards, certain wards are organized with
reference to it intensive care units, wards for premature in
fants, emergency rooms and operating rooms with their adjunct
wards. H ere the best available in talent and equipment is con
centrated. On these wards the nurses have a fairly definite idea
of the time it will take a patient to get through a crisis, so
that his fate can be predicted accurately. Operations take
several hours; a premies chances can usually be determined
in 48 to 72 hours. If a patient does not die within one day in
the emergency room he is sent either to a medical ward, if
he is expected to live, or to an intensive care unit for further
treatment. The fate of an intensive-care patient, however, may
not be predictable for as long as several days, or it may take
only a few hours.
In contrast, on cancer, medical, and neurological wards
patients often pass directly from uncertainty, with no expected
time of resolution, to the certainty stage of dying. The period
during which the nurses know when uncertainty will be resolved
is usually very short, if not imperceptible, because on these
wards certainty that the patient is dying usually reflects a sud
den, unexpected change in physical condition. Consequently,
staff members have little or no time to intensify their efforts to
save the patients life.
Nurses on the wards that are devoted to this stage of dying
are usually totally engaged in the fight for life during this
waiting period, and the intensity of their effort may involve
231
them very deeply in the patients fate. As one doctor put it,
Intensive care makes for intense nurses. This situation is
one of the most challenging for nurses, because, as we have
mentioned, they have been trained to regard saving a patient
as one of the highest achievements in nursing, and failure to
save may seriously threaten their composure. In the words of
one intensive care unit nurse: Staff feels different about the
seriously ill [uncertainty] and the terminal [certainty] patients.
They might have saved the seriously ill so they feel worse,
perhaps, that they didnt. A t the same time, however, the
intensity of the work tends to take a nurses mind off the pa
tients fate, reducing her conscious involvement and helping
her to maintain composure during the crisis. Even during
periods when intense work is not necessary, they keep them
selves as busy as possible in order to m aintain their composure
while awaiting the outcome.
However long the crisis period lasts, the prospect of failure
that is, when it seems likely that the patient will be certain
to die also causes the nurses to work even harder, to fore
stall any accusation of negligence (a direct threat to composure)
from themselves, other staff members, or the patients family.
They want to be sure that no stone is left unturned, that
there is no implication that they could have done m ore. To
withdraw their involvement from the patient they engage in
object switching (changing the object of their involvement)
by concentrating on helping patients who have a chance to
live, or on the poor doctor who has tried so hard to bring
the patient through. They also turn to each other for mutual
support in giving up the patient. To each other, they say that
there is nothing more to do, or they rationalize the patients
social loss. Even if he had lived he would have had brain
damage may be said of a young father with everything to live
for.
The risk of losing composure through excessive involvement
is especially high for nurses on the prem ature infant wards.
Nurses tend to become very involved in saving a prem ature
232
baby who lives through its first days, when they know the
m other wants it, especially if the baby has no physical defects,
but they can become equally involved in saving a premie that
the m other does not want, because it is alone and helpless in
the world. They adopt the babies, give them names, and
watch over them closely to the end. While the nurses may be
well insulated from the mother, they cannot avoid being aware
of her wishes because she either telephones the ward incessantly
or does not phone at all.
Involvement in the babys fate may even lead a nurse
to phone when she is off duty to keep track of its condition.
This radically departs from the standard strategy to reduce
involvement in patients seldom following a case on which
they are not immediately working, and seldom following their
own cases when not on duty. During the crisis period the very
involved nurse may become quite upset should the premie turn
bad, and some loss of composure is readily understood by
her coworkers. If the premie dies they fall back on the standard
strategies of object switching, absolving themselves of negli
gence, reducing social loss, and referring to Gods will.
Handling the family during this crisis period is also strategic
for maintaining composure. The intense work on all these wards
is somewhat shielded, by doors and door norms, from the
f a m ily This structural arrangement reduces nurses involvement
in the family, and prevents family members from disturbing the
nurses composure by creating a scene or asking many ques
tions. Since the family is unlikely ever to know who the nurses
on the case were, they cannot bother the nurses afterwards
by asking why the patient could not be saved. This means,
however, that any nurse passing through the door m ay be
approached for information (as we mentioned in C hapter 10).
Nurses also try to insulate themselves from the family by
putting relatives in distant waiting rooms, empty treatm ent
rooms, offices, or main lobbies to wait out the crisis. A glasswalled office is a preferred place to insulate family members
waiting out a crisis because they can be watched without actual
233
contact and thereby be prevented from drifting back to the

ward or nursing station. In isolating relatives, a nurse often
finds it helpful to tell them that she will bring news of any
change in the patients condition, but as likely as not she then
either forgets that they are waiting or simply avoids them when
she passes the waiting room. Forgetting, like avoiding, is a
standard strategy for m aintaining composure.
Talk. During this stage the nurses start to control their
conversations with the dying patient. They may discuss his
future plans with him, but they tend to restrict the scope of
their talk to the short range. They talk about getting back to
his work and family when all this is over, not about the patients
future occupational career. They interpret physical symptoms
less freely, making sure that what they say points to recovery
to keep up the unaware patients motivation to get well. During
the crisis they refrain from asking the patient, How do you
feel? to avoid prom pting him to ask about his chances. They
also avoid the subject of dying if the patient starts testing his
suspicions. Even when the awareness context is open, as before
surgery to explore for cancer, they try to deny that death is
a possible outcome, saying, for example, No one knows yet
for sure, We all die sometime, or Dont be so m orbid.
Or, the nurse keeps very busy in the patients presence, and
she responds to invitations to talk with a professional brushoff a silent, pleasant smile, a W ere doing all we can or
Well pull you through.
All these strategies enable nurses to m aintain composure
by keeping both themselves and the patient away from the
subject of dying, for a patients loss of composure is sometimes
as upsetting to nurses as the fact of possible dying itself.
During a crisis the patient is often under anesthesia or
sedation, or is comatose, so the nurses need not employ conver
sational strategy. Once they have determined whether the
patient can still hear even though apparently comatose they ac
cordingly can or cannot discuss, in his presence, his condition and
234
PART
his chances for recovery, in order to mutually support each

others composure.
The premie ward, however, does offer a unique conversa
tional strategy for maintaining composure that is not available
on other wards where patients can listen and respond. The
nurses tell their adopted infant patients anything; they can
talk of love, act possessive, reprove them for not being more
cooperative, get angry, joke, and so on. All of this adds up
to a catharsis that helps maintain composure, since stored
feelings may accumulate to break the nurses down, as does
happen frequently on other wards.
Collective Moods. A t this stage of dying the nurses col
lective mood amounts to hope that the outcome for the patient
will be recovery. Hope keeps them working hard to save the
patient and to motivate him to get well, and hope keeps up
their own morale. If the patients condition is vacillating between
certainty and uncertainty, so that the outcome cannot be pre
dicted, the nurses are not sure whether to accept that the
patient is dying or continue to hope any outcome would be
a relief. As one nurse put it, You know, she is looking much
better the last couple of days. I came back after the weekend
expecting to find her looking bad but shes looking better, she
really is. She goes back and forth. We would like her to die
and get it over with, yet if there is something going on that
she is getting better, maybe. . . . When cues fail them and
maybe perplexes them, they seek certainty from the doctors:
If the doctor knows that a patient is going to die, I prefer
to know it. They need to know what to expect so that they
can control their composure. Some doctors will tell the nurses
if they know, but others, as we have seen, will say nothing,
wishing to keep the nurses working hard to save the patient
even when there is nothing more to do for him.
235
D Y IN G IS C ER TA IN BUT TIM E O F D EA T H
IS UNKNOW N
W hen the nurses are finally certain that the patient will
die, but at an unknown time, the conditions of work and talk
that threaten their composure multiply rapidly. The progressive
building up of strategies begins to snowball. Previous strategies
to manage involvement and maintain composure, are refined
and used more often, and new strategies must be developed.
Work. Since there is nothing more to do for the patient
in this stage of dying, nurses begin now to design their work
to make the last days, weeks, or months of the patients life
comfortable and painless. Nurses work especially hard to make
the patient comfortable at this stage, to overcome their feelings
of negligence by achieving the only goal of patient care remain
ing. The need to feel justified was clearly expressed by the
nurse who said, I feel especially that I want to know I did
everything I could to make the patients last days comfortable.
So if anything I see it as more necessary rather than less.
Self-accusations of negligence are common during this stage
of dying. F or example, one nurse said, Sometimes a death
will linger in my mind. I feel I could have done more to make
her comfortable. In this case, self-accusations of negligence
made it impossible to carry out the m ajor composure strategy
of forgetting the patient and thus reducing involvement in his
death.
Family members may also accuse nurses of neglecting to
keep the patient comfortable; they may start a status-forcing
scene to get the nurses to do what they feel will insure comfort
the nurses job. To avoid such scenes the nurses isolate
the family in waiting rooms o r avoid them and call in other
helpers a social worker, chaplain, orderly, policeman, or ward
clerk. Although the strategy of role switching (referred to
in Chapter 9 ) turning family members over to others to
236
PART
avoid disturbing scenes and to reduce involvement in their

sorry plight is begun at this stage, the strategy becomes more
im portant at the last stage of dying. A t the present stage com
posure is also m aintained by turning the relative into either a
helper or patient, for this gives him a position on the ward
with which a nurse can deal in a composed manner, in her
capacity as the relatives supervisor or nurse. To avoid being
exposed to scenes or bom barded with questions by a freefloating family member with unlimited visiting privileges, nurses
try to structure the situation so as to maximize their control
of family members, and reduce threats to their own professional
composure.
Providing comfort, which reduces the feelings of negli
gence that threaten composure, is not always easy. Flexible
sedation orders from the doctor make it fairly easy as we
have seen, nurses give potentially lethal dosages to insure
painless comfort. Sometimes, however, they must overdose
without the sanction of flexible orders, using this invisible,
uncondoned strategy to cope with their own emotional reaction
to a patients pain. Thus, a nurse can m aintain her own com
posure by ensuring complete painlessness for the patient. Over
dosing may be against the expressed wishes of the patient, who
wants to be alert to settle his affairs, an d /o r the doctor, who
wants sedation used sparingly to prevent too early tolerance
of it. M ost often, however, no one knows this very occasional
strategy has been used.
The location of the pain, or the limited scope of the
doctors orders, sometimes makes it impossible to keep the
patient comfortable. A t this point the nurses collectively ex
perience an overwhelming m ood of helplessness, and to main
tain their composure, they try to avoid the patient and reduce
involvement with him. They wish for his death: We cant
relieve her pain. Everybody wants her to die for her sake,
meaning her suffering. Once we cant help them, let them die
for G ods sake and our sake as well as theirs. If the patient
is a research subject, the insuperable obstacles, both to letting
237
him die and to relieving his pain, combine to distress the nurses.
To help regain their composure, some nurses develop a
miracle rationale, hoping that something miraculous will arise
from the research to save the patient after all. A t best, slender
hope offers weak consolation.
A nother way of making dying patients comfortable is to
give some routine care in a somewhat expressive and considerate
manner. When nurses employ the composure strategy of avoid
ing the patient, dying patients sharply feel the nurses pulling
away. But when it comes to the duties of medical care, nurses
cannot literally avoid the patient. They can, however, employ
expressive avoidance ; that is, they avoid him as a person
while they are fulfilling their medical duties. They ignore him,
treat him as a body (that is, socially d ead ), wear bland pro
fessional facial expressions, exude dignity and efficiency, and
refuse or evade conversation, o r else do their chores quickly
and get out before the patient can say anything.
One nurse expressed her failure to make the patient com
fortable in this regard as negligence: I felt guilty after she
died because I could have made her life a little bit more
com fortable. A nother nurse engaging in expressive avoid
ance, while accomplishing routine duties with a 13-year-old
patient, was asked by the patient to play cards with him. This
request for continuous, relatively intimate contact forced her
literally to avoid him; she mumbled something about being very
busy and ran. But even her feeling of negligence later on was
less threatening to her composure than was the prospect of
close contact with a patient whose capacity to involve her
emotionally was, because of his high social loss value, virtually
irrestistible.
In another case a nurse involved with a young patient who
was dying began to avoid him expressively, but he countered
her strategy:
He began to make demands that she would come back
into the room and that she would spend more time with
238
PART
him. Then he began to accuse her, saying that she

wasnt as much fun as she used to be o r you hate me,
and she said that was really very rough, not only that
but he would say this to the doctors and then urge them
to ask her to spend more time with him. She handled it
by just spending more time with him.
To keep from being judged negligent by the doctors, this nurse
was forced to abandon her strategy and spend more time with
the patient.
Expressive avoidance tactics contradict the impulse to
provide good will comfort that we described earlier. While
comfort of this kind does make a nurse vulnerable to excessive
involvement, it may also help some nurses keep their composure
in the face of helplessness to save the patient. It gives them
something, however ineffective, to do for the patient.
In general, a nurse confronted with a patient known to be
dying, for whom there is nothing more to do, finds it easiest
to m aintain her composure when the patient can be made
comfortable with sedation and a minimum of expressive care.
Under these conditions, nurses can forget and avoid the patient
except for routine duties, thus minimizing both their feelings
of helplessness and negligence and the risk of emotional in
volvement.
These ideal conditions for maintaining composure are not
always easy to achieve, however, for many circumstances tend
to induce involvement with a dying patient. Some nurses cannot
help feeling upset about a patient who is left alone to die, for.
example, when no family member is in attendance. It is also
hard for any nurse to remain detached when the patients
social loss value is very high.
Occasionally some aspect of a case the patients age, sex,
marital status or occupation, for example, or the similarity of
his illness to one the nurse herself or someone close to her has
had leads a nurse to over-identify with the patient. As an
extreme form of involvement, this is disturbing to composure.
I found it so hard to talk to him, and he, of course, knew
239
enough about medicine to realize what his condition was. The

interviewer asked: He was close to your age? The nurse an
swered, Yes, he was, I think this was part of what made me
feel so badly. And the dying patient whose illness is similar to
one the nurse or a mem ber of her family actually has, or is sus
ceptible to, provides a distressing rehearsal of a potential fate.
A dying patient who lingers on can present special prob
lems: he cannot be avoided or forgotten all the time, and nurses
tend to become involved simply as a consequence of long-term
contact with him and his family. We get so attached to our
patients because they are long-term. They expect to see the
patient each day as a part of coming to work. They may become
good friends with him, expecting regular exchanges with him
and enjoying the daily job of making him comfortable. If
he is alternately going home and returning to the ward as his
condition varies, he is welcomed as an old friend each time
he returns. After a time, some nurses may develop a never
will die expectation for the patient, which weakens their
emotional defenses so that when he finally dies, loss of com
posure is more likely.
Nurses are also likely to become upset at the long arduous
struggle of some lingering patients, a struggle which in the
end only makes the dying harder on everyone. They may be
come particularly concerned about the family members who
must go through an emotional ordeal while their financial
resources are being drained away for naught. Patients deaths
may leave their family destitute financially as well as socially
This attachment to lingering patients and their families makes
it as difficult to avoid or forget patients as it would be if they
were friends known prior to hospitalization. When one such
patient died, the nurses were all with her, mutually supporting
each others composure. Such a collective form of involvement
reduces the strain on composure that each nurse would other
wise suffer alone. In the end, as one nurse put it, The longer
you know the patient, the harder it is to forget. A nother nurse,
talking about death in general, said, We talk over death here,
240
PART
get our feelings out, then forget it and go home. But she also
said, referring to a lingering patient. These are the kinds of
deaths that stay with you.
Talk. O utright avoidance and expressive avoidance are
both strategies for precluding conversation as well as control
ling work. However, once a conversation has begun, other
strategies can be used to protect a nurse from involvement or
distress. With patients approaching death, nurses avoid the
subject of death in general. This subject may easily come up
in discussions about another patient further along with the
same disease, thus allowing the first patient to use another
dying or deceased patient as a reference for gauging his
own chances.3 When the awareness context is closed, or one
of suspicion, a nurse cannot discuss another patients impending
death without virtually giving the first patients fate away. When
the context is one of m utual pretense, she might thereby destroy
the comfortable illusion that others, including the patient,
have carefully established. In an open awareness context, she
m ust still decide how much the patient should be told, or wants
to know, about how he o r another patient is going to die. Since
at this stage neither the nurse nor patient knows when he is
going to die, telling the aware patient when another patient
might die or did die may lead him to false and distressing
assumptions. Thus, giving him too many details may cause him
to break down or withdraw from all communication; either
outcome threatens her composure.
Selective listening is another conversational tactic. Nurses
listen only to information from the patient and his family that
will help them with their work, without increasing their involve
ment. They turn a deaf ear, if possible, to details about the
indicators of social loss children, spouse, job, talent that
make the impending death so distressing.
3 The use by patients of other patients as references for ascertain
ing facts about their own condition has been thoroughly studied for
the tubercular patient. See Julius Roth, Timetables (Indianapolis: BobbsMerrill, 1963).
241
Other strategies for controlling talk are designed to reduce

feedback in questions from the patient which may further
involve the nurse with him, thus threatening her composure.
Temporal references are sharply restricted, the nurse now
limiting herself to perfunctory descriptions of what will happen
to the patient in the following days or weeks, and eliminating
references to what he will do when he gets out of the hospital,
since these may occasion responses. Tomorrow you are going
to x-ray, or See you next week (when the nurse takes her
weekend off). Nurses depend more on cues they can gather
from the patients condition and less on asking him verbally
how he feels since the patient is likely to ask, in return, for
an interpretation of the way he feels. Interpretations of this
kind must be managed so as to mislead a patient in a closed,
suspicion or pretense awareness context, and to avoid mention
ing death in an open context; in either case they take an extra
effort to maintain composure. A standard strategy is not to
speak until spoken to and then to say as little as possible. This
strategy carries anticipatory rebuff of questions not yet asked,
and so discourages questioning.
A nother strategy to cut down feedback is to select from
a patients talk the least compromising subject and respond only
to it. A patient said: Yes, Im afraid, Im afraid, Im afraid
I cant breathe and I do a lot of coughing. The nurse, to
spare herself a discussion of the drastic consequences of in
ability to breathe, o r of fear of death, tried to focus on the
coughing. A nother nurse was asked directly, Am I going to
die? to which she responded with the over-generalizing
strategy: Well, we all go sometime.A nother tactic is to side
step the Am I going to die question by saying no, then
talking a blue streak about nothing in particular, steering clear
of his fate and covering her own feelings. A nother effective
strategy to cut off feedback is to parry this fateful question
with the standard role switching strategy: Talk to your
doctor. She told the interviewer that when asked this question
by a nine-year-old girl, You feel just like you swallowed your
242
PART
tonsils and you really dont know what to say. It is terribly

hard to handle. H er response was intended to stop further
discussion. Actually many nurses simply flee the room when
asked this question directly, sometimes seeking a private place
to cry and regain composure.4 On one pediatric ward the head
nurse prohibited crying on the ward; it had to be done in the
coffee room away from the children. Nurses in this stressful
situation often feel a need for self-examination, and they must
use tactics to maintain composure with the patient until they
can find privacy and perhaps a friend to talk it out with.
Death Talk and Awareness Context. We have already indi
cated that the current awareness context largely determines
the strategy a nurse uses to maintain her composure. M any of
the strategies just mentioned are also used in maintaining a
closed or pretense awareness context indeed, maintaining such
a context often depends on the nurses remaining composed.
But whether a closed awareness context helps or hinders the
nurse depends on her personal tolerance for talk about death.
A nurse who cannot tolerate talk about dying or death
feels less threat to her composure in a closed or pretense aware
ness context because the patient cannot force her to discuss
the subject. She can use the strategies we have described to
evade death talk without feeling that she is evading a conver
sation that would be legitimate in an open awareness context.
In an open or even in a suspicion context, evasion of this type
may risk damaging the patients confidence in the staff, but
in a closed awareness context, disclosure is the doctors respon
sibility.
W hen a nurse, who wishes not to, feels that an open aware
ness context requires her to talk about death, she may be able
to turn the patient over to a nurse who is noted for her ability
to discuss death a death talker. Thus, sometimes a division
4 Back regions" are structures provided by the hospital, one use
of which is for regaining composure; see Erving Goffman, Presentation
o f Self in Everyday Life (New York: Doubleday Anchor Books, 1959),
Chapter II.
243
of labor may develop among nurses to protect those who are

unable to handle this aspect of an open context. One nurse
reported, I can talk to patients about death. In fact, nurses
call me in to talk to patients because I am able to do this.
In a Catholic hospital, three of the older nuns, retired as nurses,
were on call for talking with patients about death and dying.
Chaplains, social workers, and psychiatrists are sometimes able
to take over this task. In these instances the nurse employs
the role switching strategy to maintain composure.
The nurse who is able and even wants to talk with the
patient about the implications of his dying may feel quite
uncomfortable in a closed awareness context. To keep the
patient unaware, she is forced to put up a false front, which
is a burden on her composure because it affords protection
she neither needs nor wants. Unless she receives instructions to
the contrary, she must not indulge in talk or behavior that
would indicate to the patient that he is dying. Yet, to display
some involvement, to talk about the death and to act natural
would enhance her composure. F o r this reason some nurses
often feel frustrated and distraught in the closed context, and
in order to stay composed, avoid the patient, although they
would rather give him com panionship.5
This threat to composure that a closed context gives to
nurses who wish to talk is increased if the patient lingers, starts
to suspect his fate, or signals that he wants to talk about
dying. The nurses begin to debate among themselves whether
the patient really knows, and, if so, how awful it is that they
cannot help him by talking about it. Further, they do not like
having to risk losing his trust by acting as if he will recover
when, possibly, he knows he will not. They are also upset by
family problems relatives moving in on the patients estate,
or a child who will not be able to say goodbye that could
Nurses in this situation were highly likely to go to a sociologist
interviewing on the subject to vent their frustration. We later used this
as a tactic to obtain information, for most often we were the only
people available to nurses for listening to the stresses and strains of such
a sensitive situation.
244
PART
be resolved if they could discuss the patients death with him.

U nder this sort of pressure, nurses occasionally break the rules
and tell a patient his true fate in time for him to settle his
affairs. These nurses appear rather composed after this trans
gression.
Collective Moods. W hen the nurses are certain that the
patient is dying, they begin developing several collective moods
(italicized below) which serve to reduce their involvement in
the patient and m aintain their composure. The moods continue
with refinements and reinforcements through the last stage of
dying (when time of death is known) until death. The nurses
derive continual mutual support and self-control from these
collective moods.
First, nurses begin, collectively, to prepare themselves for
the ordeal ahead by viewing the patients death as inevitable
and even desirable. A t the same time, they develop collective
rationales to explain and condone their acceptance of his death
( It is best for him and his family ). Respites in these moods
may occur when the patient rallies for a time before proceeding
downhill again ( So you feel better, even though you know
she couldnt live ). A lingering patient who can be made
comfortable and never seems to die also provides a respite
as he becomes familiar to the nurses they begin to forget that
he is dying, thereby altering their moods for a time. A patient
whose social loss value is high ( He was a young boy who had
everything to live for ), however, may reduce the composure
drawn from acceptance by lingering on interminably. The
ordeal becomes too much to bear and a m ood of sadness over
comes the nurses. Their collective impulse is to "let him die,
if he is being kept alive with drugs or equipment (see Chap
ter 12.)
The collective mood of acceptance is accompanied by one
of relief a m ood with many aspects. Nurses are relieved to
know the patients fate; it at least makes the situation definite.
Since there is nothing more to do they no longer need to hope
for recovery or be unduly disturbed about failing to save the
245
patient. They can anticipate additional relief after they know

when to expect his death and still more after he dies. These
anticipations counteract potential distress during the ordeal of
these last two stages of dying. They are also relieved when
a patient is sent home o r to a nursing home, another hospital
or another ward. This mood of relief compensates for any guilt
generated by the wish somehow to be rid of the dying patient.
A t death, relief results in a collective mood of gladness that
the ordeal is over for themselves, the patient and the family.
If they have had trouble making him comfortable and free of
pain the ordeal is increased, and the relief is greater when he
finally leaves the ward or dies.
A t this stage of dying when time of death is not yet known
the nurses morale now becomes a critically im portant col
lective aspect of the sentimental order of the ward, and there
fore an im portant underpinning of their composure. Their
morale is intimately linked to the patients. If he is withdrawn
or depressed, the nurses m ust sustain their morale on their own
through mutual support with no support from the patient. (This
becomes a strong reason for avoiding the patient.) If the pa
tients morale is good, however, he becomes the pacesetter for
the morale of the ward, for it would not do for a nurse to be
less cheerful than a patient. Nurses are literally forced to oneup the patient by having higher morale. This is one very
good reason why nurses are so grateful and appreciative when
a patient dies gracefully and cheerfully. Such patients make
it comparatively easy to maintain composure, especially when
they could, because they are known to be dying, legitimately
exert an emotional drag on staff morale.
A patient wavering between cheer and depression may be
controlled by the strategy of professional calm and mild
sympathy. Too much sympathy might dissolve the patients
tenuous control over himself, thus straining the nurses morale.6
Too much hope might m ake the patient still more depressed,
8 On the inundation of expressive boundaries, see Fred Davis,
Deviance Disavowal, Social Problems, 9 (Winter 1961), p. 124, et
passim.
246
pa rt t h r e e : pr o b l e m s o f a w areness
should he later suspect or discover his true fate. By remaining

calm, nurses hope to settle the patient into a middle-range
morale, which in turn will help them m aintain their own
composure with a minimum of strain.
E X PE C T E D T IM E O F D EA TH
In this, the last stage of dying, the strategies used in the
previous stage are still very useful, but to cope with the un
settling effects of the death watch, presence at death, the
death itself, and handling the family, the nurses develop addi
tional composure strategies. F or patients who pass directly
from either stage of uncertainty to this last stage, the strategies
reviewed directly above must be developed now in the remaining
time before death. In many cases, the specific time when death
is expected is known for only a few hours or days at the end
of the unknown-time stage, though it may be expected for a
longer period.
T he doctors participation in this last stage may help or
hinder the nurses in maintaining their composure. Nurses can
more easily ascertain that a patient will die than they can tell
when he will die, but knowing when death can be expected
will affect their m anagement of the patient, his family, and
their own behavior. By telling the nurses when to expect death,
and thus initiating this stage for them, the doctor helps them
prepare themselves to face the last days or hours with maximum
composure. They reassure each other not to feel badly when
death comes, reinforcing their collective moods of acceptance
and relief and reducing their feelings of helplessness and failure
due to inability to save the patients life. They also start plan
ning strategies to avoid imposing on a single nurse the strain
of the death watch and presence at death.
The Death Watch. Nurses may implement the standard
strategy of creating an informal team for the death watch. All
members participate in the death watch, no m atter which nurse
was assigned the patient. Each of the nurses on the team will
247
look into the room as she goes by and all will take turns
answering the patients calls. In some cases, however, one nurse
may have to sit with the patient for eight hours, being relieved
only for meals and coffee. This trap can be quite distressing,
especially if the patient has a high social loss value which
exacerbates the feeling of helplessness. One nurse watching a
14-year-old girl during her last hours said: I felt badly about
it, you know, here was a young girl and just to see her lying
there, and there really wasnt too much you could do for her,
you know, only to keep her comfortable and then she wasnt
even aware of that. I think more than anything, its just a help
less feeling, sort of a loss feeling, you just dont know what to
do and you couldnt do too much in this particular case.
One strategy to reduce the strain on composure imposed
by the death watch is to ask the doctor for flexible sedation
orders. Once the doctor has told the nurses when the patient
is likely to die, they may legitimately request flexible orders,
to insure a comfortable ending for the patient. F o r the nurse
who m ust participate in the death watch, this gives her an
opportunity to alleviate her own distress as well as the patients
sometimes final agony.
Nurses can also find ways to delegate the death watch,
usually to someone who is not quite aware of the task he is
being asked to perform (another use of role switching ). If
the dying patient is in a room with an alert patient, the nurse
may leave the room with a pressing work excuse, asking
the alert patient to call her immediately if he notices a change
in the other patient. Nurses will also ask an ever-present family
member, or perhaps a chaplain, to sit with the patient. If no
one is available, a patient may be left to die alone, between
periodic checks, though nurses find this outcome most disturbing
unless he is already comatose.
W hen the patient approaches death, a num ber of pre
announcements have to be made; keeping a death watch is
a necessary preliminary to these announcements. A doctor must
be called, and calling him either too soon o r too late is em
barrassing, and hence threatening to composure. (Doctors can
248
pa r t t h r e e : pr o b l e m s o f a w areness
relieve nurses of this source of strain by making recurrent visits

to the patient.) If the patient is a Catholic, the Catholic chap
lain m ust also be called. Family members must be notified
if they exist, but nurses may also have to interrupt or
prevent a routine family visit to the bedside, in order either
to avoid announcing the imminent death to those unable to
bear the shock o r to keep relatives from witnessing the death.
Timing and handling these announcements with com posure is
possible only if the doctor has given the nurses a fairly precise
indication of when death is likely to occur. Otherwise a sudden
change for the worse may disrupt the ward and upset the
nurses. Some nurses, therefore, will ask the doctor when to
expect death when they are not told (if they can find him ).
Presence at Death. The nurses composure is crucial during
the entire final stage of dying, but, especially at the mom ent
of death, when the likelihood of breaking down is greatest.
F o r this reason nurses usually prefer to avoid the death scene.
During the first stage of certainty, a collective m ood of wishing
not to be present at the patients death tends to develop among
them. Since they do not know, at that stage, when to expect
death, they can do little but sustain themselves hopefully on
the fiction that patients die at night.
Once they know when to expect death, however, they can
make active efforts to be absent. They can take vacations, or
request ward changes or different patient assignments, or they
can simply disappear when the time comes. A nd sometimes staff
nurses are saved by a routine rotation off the ward. If the
doctor or chaplain is present, the nurse can sometimes excuse
herself for other work. On some wards special nurses take care
of the moribund during the death watch and death, relieving the
staff nurses of responsibility for the dying situation altogether.
One staff nurse in this situation, however, did say that she
was bothered by not being able to take care of h er patient at
the end.
F o r the many nurses who cannot avoid being present at death,
several conditions affect composure. O n wards that seldom ex
A wareness of the Nurses Composure
249
perience death, such as obstetric and some pediatric or medical

wards, nurses lack experience and may be terribly upset -by
watching a patient die. These nurses are also likely to become
upset because death represents a failure among many successes,
and because the wards sentimental order has let them become
deeply involved with the patient and his family. As mentioned
earlier, these wards allow much involvement with patients, not
only because of the low death ratio but because when a
patient starts to die he will often be sent to another ward,
such as an intensive care unit, for saving. On wards with a
high death ratio, where involvement is not condoned and ex
perience is greater, nurses m ake less effort to avoid being
present at death. In general, just as they do after failing to save
a patient in the second stage of dying, nurses who m ust partici
pate in a death watch and be present at death try to transfer
their involvement (object switching) to family members, doc
tors, or other patients, maintaining composure by h e lp in g the
living.
A t the time of death, it is the doctors duty to announce
it to family members, but in some cases they appear on the
scene unexpectedly, unprepared for the death, and demand
to know what is happening. The nurse may have no time to
remove the relatives or to get a doctor to pronounce death.
If she lies about the patient and removes the family she risks
a scene in which all parties might lose composure. She is,
therefore, forced to announce the death herself, while the
family actually watches it take place, hoping her own com
posure, if she can maintain it, will help the family members
keep theirs. One nurse, horrified to find an unsuspecting aunt
coming to inquire about her nephew after his brain tum or
operation, told the aunt directly that he had died. H a v in g
regained her own composure, she was then able to help the
stunned aunt control herself.
Forgetting. A t death, another collective m ood develops to
help the nurses maintain composure forget the patient. A
patient is easiest to forget if his dying follows the pattern ex
250
PART
pected for his condition. (Ease in forgetting a dying patient

is one of the more useful consequences of death expectations.)
But if the patient dies in great pain, or after prolonged linger
ing; if the death involves some negligence or failure on their
part; if he had unusual habits such as drinking shaving lotion;
or if his social loss value was high, he may be very hard to
forget. In interviewing nurses about their past experiences with
dying patients, they invariably mention first those patients who
were hard to forget.
In many cases the collective effort to forget will preclude
further discussion of the- patient, and each nurse is therefore
left to find her own individual means for giving up the patient
and regaining composure. In extreme cases of failure an
unanticipated suicide or negligence the collective m ood may
be curiosity instead of forgetfulness. The nurses discuss the
case among themselves and with doctors to figure out what
went wrong, why the patient died, or died as he did. To forget
the patient once and for all, they m ust rid themselves of feelings
of guilt or negligence. It is not uncommon for nurses to be
reminded with a sense of distress of a patient who is hard to
forget, when passing the room where only a few days before
he was alive. Requesting transfers to another ward alleviates
these reminders, and it is done occasionally.
U N EX PE C TE D D EA TH
We have described the progressive accumulation of strategies
by which a nurse tries to maintain her composure as guided by
her changing death expectations for the patient as he approaches
death. This developmental building up of strategies does not
occur for nurses who have (1 ) no death expectation; and it
is compromised to a degree for nurses with either (2 ) radically
changing expectations or (3 ) an inaccurate expectation. Thus,
these are three principal sources of the unexpected or surprise
death of a patient, which in turn become sources of reduced
control over both involvement with the patient and professional
251
composure. The underlying point is that any death expectation

but the accurate one leaves the nurse open to the shock of
an unexpected death, though the threat to her composure varies
with the degree of inaccuracy.
When death is really unexpected, and the nurses have no
death expectation at all, they have no cause to develop strategies
to m aintain composure. Sudden death therefore catches them in
a most vulnerable state. Lack of awareness among nurses is
due, as we have indicated, to an absence of appropriate cues,
or from the doctors failure to reveal his own awareness, or
both. Insufficient experience is also a factor, of course: students
must learn, first, to recognize an impending death and, then,
to manage involvement in terms of death expectations, so as
to m aintain their composure. From her shock at an unantici
pated suicide, one student nurse realized that she had started
to cope with dying and death according to her death expecta
tions: Yes, boom [a suicide], therefore I am frustrated because
it happened so quickly. Last time I wasnt frustrated because I
knew it would come sooner or later. I wasnt much prepared for
it the other time either, but emotionally I was more prepared
for it.
Inaccurate expectations may cause a nurse to be too care
ful, or not careful enough, in managing her involvement with
the patient. If she is careless, she is liable to be very upset by
a death that could have been anticipated correctly. If she is
too careful, her strategies for managing involvement may do
unnecessary damage to her relations with the patient and his
family. F or example, she may avoid a patient or limit talk
with him to such an extent that he does not trust her, and
this in turn m ay make her less effective in patient care and
unnecessarily hard on the patient and his family. Again, failure
to form an accurate death expectation may be due to inex
perience, inadequate information from the doctor, or misreading
cues.
Radically changing expectations also make it difficult for
a nurse to manage her involvement. She may never quite know
252
PART
whether involvement will ultimately destroy her composure:

at one m om ent the patient is going to die, at the next, he will
linger some time, the next he may recover, and so forth. In
these cases, the patients physical condition changes constantly,
m aking it difficult for either doctor or nurse to establish a
reliable death expectation. Similarly, a vacillating doctor can
leave the nurses at a loss as to how much they should guard
their composure. In such an unstable situation a nurses
relations with the patient and his family suffer because her talk
and avoidance patterns fluctuate in accordance with her chang
ing expectations. And the death m ay catch her off guard,
devastating her composure.
Negligence in W ork. The surprise death is particularly up
setting to nurses composure because it implies that they might
have been negligent in nursing care. This implication varies with
the three sources of the surprise and with the nearness of the
patient to death. If the patient was only seriously not fatally
ill, o r if his death was still not certain, his unexpected death
immediately raises a question in the nurses minds as to whether
they could have saved him by watching him m ore closely.
Having h ad no expectation of death o r having been, inac
curately, so certain of death that they concentrated on comfort
rather than recovery, nurses m ay in fact have been negligent.
As one nurse put it, The initial shock of a sudden death
bothers you. W hat could I have done? Could I have noticed
something that would have been a warning? She will be very
upset if she remembers something that should have alerted her
to the possibility of saving the patient.
A fter an unexpected death, the nurses may go over the
patients last days in great detail, to see if any of them were
negligent in any way. In one case a wom an who committed
suicide b u t who otherwise would have lived this discussion
went on for weeks and was linked to a coroners inquest and
an autopsy. Usually, however, the post-mortem discussion is
finished in a few days and any negligence accounted for.
253
W hen the patients death was in fact inevitable, his comfort

may have been neglected somewhat by a nurse who did not
expect death o r who believed that death was not certain. She
might have focused on recovery to the detrim ent of comfort
care. W hen such a patient dies, she regrets not having done
m ore to m ake the patient comfortable, instead of insisting on
a treatm ent, perhaps a rigorous one, designed to cure him. Of
course, a nurse in this situation is still less upset than she
would be if she had neglected recovery care in favor of comfort.
W hether their feelings of negligence refer to recovery o r to
comfort, a surprise death m ay very well lead distressed nurses
to blame the doctor for not telling them what was going on.
To some extent, they can thus attribute any apparent negligence
on their part to his negligence o r incompetence in keeping them
informed. To forestall such accusations, the doctor can try
doubly hard to obtain an autopsy in cases for which he cannot
give an adequate explanation. A nother tactic is to accuse him
self first. In one very dram atic case a m other hem orrhaged
on the delivery table and both she and her baby died. The
husband was shattered and absolutely lost his m ind. The
nurses were terribly upset, and the doctor broke down and
cried. While his crying was genuine, its effect was to forestall
accusations of negligence or incompetence based on his failure
to foresee trouble beforehand. T hat is, he was not hiding his
awareness from them. Having to support the doctor also
helped the nurses regain composure.
Forgetting. Collective forgetting after death is inhibited
and delayed when the patient dies unexpectedly. The nurses
m ust discuss what happened and whether any negligence was
involved. In this way they help each other cope with their
distress, and regain some composure, but the restorative process
of forgetting the patient is delayed. Once the question of negli
gence is settled, however, they can begin to forget the death,
which, as we have seen, forces the discussion to end. Telling
nurses on other wards about the surprise death also enhances
254
PART
forgetting and restores composure. The other nurses may

find the circumstances of the death interesting, or they m ay have
known the patient themselves. Telling another nurse, especially
one who was not involved in the case, allows the nurse to organ
ize the tragedy and account for it to herself; this helps her
forget it and regain her norm al composure.
This m ethod of regaining composure may, however, have
unpleasant consequences. Sometimes the news is shocking
enough to reverberate through the hospital and even to the
newspapers. Others in the hospital may be disturbed because
they are in some way implicated, as the head of nursing services,
for example, would be implicated as the overseer of negligent
nurses in a preventable suicide.
In the process of forgetting, the nurses composure is least
threatened when they can remember only the general circum
stances, not the individual or the agonizing details of his few
last days and his death. In a surprise death, the details
of the last days are constantly mulled over afterward, making
it difficult to forget them. These details m ay remain in the
nurses memories to disturb them much later, sometimes to the
point th at they cannot face other dying patients with composure
and have to arrange for transfer to wards where they are un
likely to have to deal with them.
Overdue Death. A nother surprise inherent in the dying
situation occurs when a patient does not die at the expected
time. H e may linger on interminably, beyond all reasonable
expectations for his condition. F or example, a child with a
congenital defect was supposed to die at five or six years of
age and lived on to thirteen. One nurse found this so disturbing
that she avoided the patient, especially his continual requests
to play cards. A fter he died, his seven years of lingering were
very hard for her to forget. (Unfortunately this particular pa
tient had been taken care of for six years by nurses who had
well-developed guards against involvement. His life as a patient
would have been much easier if the nurses had known that
255
his death would be so delayed, for then they could have per
mitted a more appropriate degree of involvement during the
years of lingering.)
One strategy to handle an overdue death is simply to avoid
and forget the patient. This is sometimes aided by the collective
mood: Its senseless. When the patient lingers on indefinitely
in a comatose state physically alive, socially dead, nurses may
just stop thinking about him. One nurse said in such a case, It
m akes no sense for her to live beyond the point where she
went on living and that was by some m onths.
The patient who is expected to die on the operating table,
b u t instead lives on for three o r four days, is another example
of overdue death. In one such case, during this short but
intensive overdue period, nurses who had participated in the
operation became quite upset. They felt that perhaps more
could have been done for the patient if the doctors had ap
proached him with the expectation th at he was going to
recover, and that everyone had been negligent. Each day, as
they came to work, they asked whether the patient had died
yet. Only after he died could the nurses who were involved
convincingly regain their norm al composure. Thus some overdue
patients are forgotten before death, and others never forgotten.
PESSIMISM AND C OM POSU RE
Built into the development of strategies to manage involve
m ent and maintain composure while the patient is dying, is
an element of pessimism. The nurses base their defenses on a
conservative estimate of the patients chances, preferring ex
cessive caution to the risk of breaking down at the wrong
moment. As we have indicated, this too-careful guarding of
composure may be hard on a patient who is really not as
close to dying as the nurses believe. The nurses pessimism,
therefore, is a source of slightly inaccurate death expectations.
If the patient recovers comes alive a nurse with a
well-developed strategy against losing composure may be dis-
256
FART
tressed by the realization that she failed him. She was negligent.
If she had not been so careful to guard her composure, she
could have treated the patient as if he were merely seriously
ill, talking with him about almost anything and motivating
him to get well. N ot only could she have applied her training
more effectively, but the patient would have had m ore com
panionship.
In general, this implicit pessimism causes nurses, too often,
to pull away from patients sooner than necessary and, in
deed, why should it be necessary at all? If the patient has
been unaware that he is expected to die, this pulling away may
risk a prem ature, abrupt disclosure of his impending death
which can be very distressing to the patient and unnecessary
if he is still in the uncertain stages of dying. One uncertain-todie patient who later recovered told us, W hen I felt the
nurses pulling away from me, I knew I was dying and I became
terribly depressed.
Part Four
Conclusions
14
The Practical
Use of Awareness Theory
In this chapter we shall discuss how our substantive sociological
theory has been developed in order to facilitate applying it in
daily situations of terminal care by sociologists, by doctors
and nurses, and by family members and dying patients. The
application of substantive sociological theory to practice requires
developing a theory with (a t least) four highly interrelated
properties. (A s we have dem onstrated in this book and will
discuss explicitly in the next chapter, a theory with these
properties is also very likely to contribute to formal i.e., gen
eral sociological theory.) The first requisite property is that
the theory m ust closely fit the substantive area in which it
will be used. Second, it m ust be readily understandable by
laymen concerned with this area. Third, it m ust be sufficiently
general to be applicable to a m ultitude of diverse, daily situa
tions within the substantive area, not just to a specific type of
situation. Fourth, it m ust allow the user partial control over
the structure and process of the substantive area as it changes
through time. We shall discuss each of these closely related
properties and briefly illustrate them from our book to show
how our theory incorporates them, and therefore why and how
our theory can be applied in term inal care situations.1
1 Applied theory can be powerful for exactly the reasons set forth
by John Dewey, some years ago: W hat is sometimes termed applied
science . . . is directly concerned with . . . instrumentalities at work in
effecting modifications of existence in behalf of conclusions that are
reflectively preferred. . . . Application is a hard word for many to
accept. It suggests some extraneous tool ready-made and complete which
259
260
PART FOUR: CONCLUSIONS
FITNESS
T hat the theory m ust fit the substantive area to which it
will be applied is the underlying basis of the theorys four
requisite properties. It may seem obvious to require that sub
stantive theory m ust correspond closely to the data, but actually
in the current ways of developing sociological theory there
are m any pitfalls that may preclude good fitness.2 Sociologists
often develop a substantive theory theory for substantive areas
such as patient care, delinquency, graduate education that
embodies, without his realization, the sociologists ideals, the
values of his occupation and social class, as well as popular
views and myths, along with his deliberate efforts at making
logical deductions from some form al theory to which he became
committed as a graduate student (for example, a theory of
organizations, stratification, communication, authority, learning,
o r deviant behavior). These witting and unwitting strategies
typically result in theories too divorced from the everyday
realities of the substantive area, so that one does not quite
is then put to uses that are external to its nature. But . . . application of
science means application in, not application to. Application in some
thing signifies a more extensive interaction of natural events with one
another, an elimination of distance and obstacles; provision of oppor
tunities for interactions that reveal potentialities previously hidden and
that bring into existence new histories with new initiations and endings.
Engineering, medicine, social arts realize relationships that were unreal^
ized in actual existence. Surely in their new context the latter are
understood or known as they are not in isolation. Experience and
Nature (Chicagor-Open Court Publishing Company, 1925), pp. 161-162.
* For many years, Herbert Blumer has remarked in his classes that
sociologists perennially import theories from other disciplines that do
not fit the data of sociology and inappropriately apply sociological
theories developed from the study of data different than that under
consideration. Cf. The Problem of the Concept in Social Psychology,
American Journal o f Sociology (March, 1940), pp. 707-719. For an
analysis of how current sociological methods by their very nature often
result in data and theory that does not fit the realities of the situation
see Aaron V. Cicourel, M ethod and Measurement in Sociology (New
York: Free Press of Glencoe, 1964).
261
know how to apply them, or in what part of the social structure

to begin applying them, o r where they fit the data of the
substantive area, o r what the propositions m ean in relation
to the diverse problems of the area. The use of logical deduction
rests on the assumption that the formal theory supplies all
the necessary concepts and hypotheses; the consequences are a
typical forcing and distorting of data to fit the categories of the
deduced substantive theory, and the neglecting of relevant data
which seem not to fit or cannot be forced into the pre-existing
sociological categories.* In light of the paucity of sociological
theories that explicitly deal with change,4 logical deduction
usually is carried out upon static theories which tends to ensure
neglect, distortion, and forcing when the deduced theory is
applied to an ever-changing, everyday reality.
Clearly, a substantive theory that is faithful to the everyday
realities of the substantive area is one that is carefully induced
from diverse data gathered over a considerable period of time.
This research, usually based prim arily on qualitative data
gathered through observations, interviews and documents and
perhaps later supplemented by surveys, is directed in two ways
toward discovering new concepts and hypotheses, and then
continually testing these emerging hypotheses under as many
diverse conditions as possible. Only in this way will the theory
be closely related to the daily realities (w hat is actually going
on ) of the substantive area, and so be highly applicable to
dealing with them. A fter the substantive theory is sufficiently
formulated, formal theories can be scrutinized for such models,
concepts and hypotheses as might lead to further formulation
O u r position may be contrasted with that of Hans L. Zetterberg
who, after some exploratory research to determine problems, bypasses
development of substantive theory and goes directly to formal theories
for help. He says, We must know the day-by-day issues facing the
practitioner and then search the storehouse of academic knowledge to
see whether it might aid him. Social Theory and Social Practice (New
York: Bedminster Press, 1962), p. 41.
This is noted by Wilbert Moore in Predicting Discontinuities in
Social Change, American Sociological Review (June, 1964), p. 332,
and in Social Change (Englewood Cliffs, N. J.: Prentice-Hall, 1963),
preface and Chapter I.
262
PART
f o u r : c o n c l u s io n s
of the substantive theory.5 We have described in the appendix

on m ethod how we have proceeded in developing our theory
to fit the realities of terminal care in hospitals. Readers who
are fam iliar with this area will readily be able to judge our
degree of success in that enterprise.
UND ERSTA N D IN G
A substantive theory that corresponds closely to the
realities of an area will be understood and make sense to
the people working in the substantive area. This understanding
is very im portant since it is these people who will wish either
to apply the theory themselves or employ a sociologist to apply
it.6 Their understanding the theory tends to engender readiness
5 Thus, in contrast to Zetterberg who renders his data directly with
a formal theory, we first develop a substantive theory from the data
which then becomes a bridge to the use of what formal theories may
be helpful. By bridging the relation of data to formal theory with a
carefully thought out substantive theory the forcing, distorting and
neglecting of data by rendering it with a formal, usually thought-up,
theory is prevented in large measure. See Zetterberg, op. cit., Chapter
4, particularly pp. 166-178.
In contrast, both Zetterberg and Gouldner imply by their direct
use of formal theory that the practical use of sociological theory is the
monopoly of the sociologist as consultant, since, of course, these
formal theories are difficult ehough to understand by sociologists.
Zetterberg, op cit. and Alvin W. Gouldner, Theoretical Requirements
of the Applied Social Sciences, American Sociological Review, Vol.
22 (February, 1959). Applying substantive theory, which is easier to
understand, means also that more sociologists can be applied social
theorists than those few who have clearly mastered difficult formal
theories to be competent practitioners of them. Zetterberg, op. cit.,
p. 18.
Another substantive theory dealing with juvenile delinquency, in David
Matza, Delinquency and Drift (New York: Wiley, 1964), provides a
good example of our point. This is a theory that deals with what is
going on in the situations of delinquency. It is not another rendition
of the standard, formally derived, substantive theories on delinquency
which deal intensively with classic ideas on relations between culture
and subculture, conformity, opportunity structures, and social stratifi
cation problems, such as provided in the formal theories of Merton and
Parsons and as put out by Albert Cohen and Richard Cloward and
Lloyd Ohlin. As a result two probation officers of Alameda County,.
California, have told us that at last they have read a sociological
theory that deals with what is going on and makes sense and that
will help them in their work. Thus, they can apply Matzas theory in
their work!
263
to use it, for it sharpens their sensitivity to the problems that

they face and gives them an image of how they can potentially
make m atters better, either through their own efforts or those
of a sociologist.7 If they wish to apply the theory themselves,
they m ust perceive how it can be readily mastered and used.
In developing a substantive theory that fits the data, then,
we have carefully developed concepts and hypotheses to facili
tate the understanding of the theory by medical and nursing
personnel. This, in turn, has ensured that our theory corresponds
closely to the realities of term inal care. O ur concepts have
two essential features: they are both analytic and sensitizing.
By analytic we mean that they are sufficiently generalized to
designate the properties of concrete entities not the entities
themselves and by sensitizing we mean that they yield a
meaningful picture with apt illustrations that enable medical
and nursing personnel to grasp the reference in term s of their
own experiences. F or example, our categories of death ex
pectations, nothing more to do, lingering, and social
loss designate general properties of dying patients which un
questionably are vividly sensitizing or meaningful to hospital
personnel.8
To develop concepts of this nature, which tap the best of
two possible worlds abstraction and reality takes consider
able study of ones data.9 Seldom can they be deduced from
formal theory. Furtherm ore, these concepts provide a necessary
bridge between the theoretical thinking of sociologists and
the practical thinking of people concerned with the substantive
area, so that both parties m ay understand and apply the theory.
The sociologist finds that he has a feeling for the everyday
realities of the situation, while the person in the situation finds
7 See Rensis Likert and Ronald Lippitt, T h e Utilization of Social
Science, in Leon Festinger and Daniel Katz (eds.), Research Methods
in the Behavioral Sciences (New York: Dryden Press, 1953), p. 583.
* On sensitizing concepts see Herbert Blumer, W hat is Wrong with
Social Theory, American Sociological Review, 19 (February, 1954),
pp. 3-10, quote on p. 9.
* Zetterberg has made this effort in choosing concepts with much
success, op. cit., p. 49 and passim.
264
PART
he can m aster and manage the theory. In particular, these

concepts allow this person to pose and test his favored hy
potheses in his initial applications of the theory.10
W hether the hypotheses prove somewhat right or wrong,
the answers still are related to the substantive theory; use of
the theory helps both in the interpretation of hypotheses and
in the development of new applications of the theory. F or
example, as physicans (and social scientists) test out whether
or not disclosure of terminality is advisable under specified
conditions, the answers will be interpretable in terms of aware
ness contexts (C hapters 3 and 6 ) and the general response
process (C hapter 8 ). This, in turn, will direct these people
to further useful questions as well as lead to suggestions for
changing many situations of term inal care.
In utilizing these types of concepts in our book, we have
anticipated th at readers would almost literally be able to see
and hear the people involved in term inal situations but see
and hear in relation to our theoretical framework. It is only
a short step from this kind of understanding to applying our
theory to the problem s that both staff and patients encounter
in the dying situation. F or instance, a general understanding
of what is entailed in the m utual pretense context, including
consequences which may be judged negative to nursing and
medical care, may lead the staff to abandon its otherwise un
witting imposition of mutual pretense upon a patient. Similarly,
the understanding yielded by a close reading of our chapters
on family reactions in closed and open contexts should greatly
aid a staff mem bers future management of not to say com
passion for those family reactions. A good grasp of our theory,
also, will help hospital personnel to understand the charac
teristic problems faced on particular kinds of hospital services,
including their own, as well as the typical kinds of solutions
that personnel will try.
10 Gouldner (op. cit., pp. 94-95) considers in detail the importance
of testing the favored hypotheses of men who are in the situation.
However, we suggest that the person can test his own hypotheses too,
whereas Gouldner wishes to have a sociologist do the testing.
265
G E N E R A L IT Y
In deciding upon the analytic level of our concepts, we
have been guided by the criteria that they should not be so
abstract as to lose their sensitizing aspect, but yet must be
abstract enough to make our theory a general guide to the
multi-conditional, ever-changing daily situations of terminal
care. Through the level of generality of our concepts we have
tried to make the theory flexible enough to make a wide variety
of changing situations understandable, and also flexible enough
to be readily reformulated, virtually on the spot, when necessary,
that is, when the theory does not work. The person who applies
our theory will, we believe, be able to bend, adjust, or quickly
reform ulate awareness theory as he applies it in trying to keep
up with and manage the situational realities that he wishes to
improve. F o r example, nurses will be able better to cope with
family and patients during sudden transitions from closed to
pretense or open awareness if they try to apply elements of
our theory (see Chapters 3, 8, 9 ), continually adjusting the
theory in application.
We are concerned also with the theorys generality of scope.
Because of the changing conditions of everyday terminal situa
tions it is not necessary to use rigorous research to find precise,
quantitatively validated, factual, knowledge upon which to
base the theory. F acts change quickly, and precise quantitative
approaches (even large-scale surveys) typically yield too few
general concepts and relations between concepts to be of broad
practical use in coping with the complex interplay of forces
characteristic of the substantive area. A person who employs
quantitatively derived theory knows his few variables better
than anyone, but these variables are only part of the picture. 11
Theory of this nature will also tend to give the user the idea
that since the facts are correct so is the theory; this hinders
the continual adjustm ent and reform ulation of theory necessi Zetterberg, op. cit., p. 187.
266
PART
tated by the realities of practice. Because he is severely limited

when facing the varied conditions and situations typical of
the total picture, the person who applies a quantitatively derived
theory frequently finds himself either guideless or applying
the inapplicable with (potentially) unfortunate hum an and
organizational consequences. This kind of theory typically
does not allow for enough variation in situations to take into
account the institution and control of change in them. Also,
it usually does not offer sufficient means for predicting the
diverse consequences of any action, done with purpose, on
those aspects of the substantive area which one does not wish
to change but which will surely be affected by the action. Who
ever applies this kind of theory is often just another voice to
be listened to before the decision is reached or announced
by those who do comprehend the total picture.12
Accordingly, to achieve a theory general enough to be
applicable to the total picture, we have found it m ore im portant
to accumulate a vast num ber of diverse qualitative facts on
dying situations (som e of which may be slightly inaccurate).
This diversity has facilitated the development of a theory that
includes a sufficient num ber of general concepts relevant to
most dying situations, with plausible relations among these
categories that can account for much everyday behavior in
dying situations. Though most of our report is based on field
observations and interviews, we have used occasional data
from any source (newspaper and magazine articles, biographies
and novels, surveys and experim ents), since the criterion for
the credibility and potential use of this data is how they are
integrated into the emergent substantive theory.13
The relations among categories are continually subject to
qualification, and to change in direction and magnitude due
to new conditions. The by-product of such changes is a correc
tion of inaccuracies in observation and reintegration of the
12Ibid.
13 This theme on integration into a theory as a source of confirming
a fact or a proposition is extensively developed in Hans L. Zetterberg,
On Theory and Verification in Sociology (New Jersey: Bedminster
Press, 1963).
267
correction into the theory as it is applied. The application is

thus, in one sense, the theorys further test and validation.
Indeed, field workers use application as a prime strategy for
testing emerging hypotheses, though they are not acting as
practitioners in a substantive area. In the next section, by
illustrating how our theory guides one through the multifaceted
problem of disclosure of terminality, we indicate how one con
fronts the total picture with a theory that is general enough in
scope to be applicable to it.
This m ethod of discovering and developing a substantive
theory based on a m ultitude of diverse facts tends to resolve
two problems confronting the social scientist consultant, who,
according to Zetterberg, is dependent on what is found in the
tradition of a science and, when this fails, is apt to proceed
on guess work so as not to lose respect and future assign
m ents. 14 O ur m ethod resolves these problems in large measure
because it is not limited by the dictum that Zetterbergs con
sultant m ust follow: Only those details were assembled by
the consultant and his co-workers that could be fitted into
the categories of sociology, i.e., phrased in sociological termi
nology. 15 As stated earlier in the section on fitness, we do
not believe that the categories of sociology can at the outset
be directly applied to a substantive area without great neglect,
forcing, and distortion of everyday realities. A substantive
theory for the area m ust first be induced, with its own general
concepts; and these concepts can later become a bridge to more
formal sociological categories if the latter can be found. As
Wilbert M oore has noted, however, we still lack the necessary
formal categories to cope with change adequately.
C O N TRO L
The substantive theory must enable the person who uses
it to have enough control in everyday situations to make its
14 Zetterberg, Social Theory, op. cit., pp. 188-189.
13 Ibid., p. 139. This dictum is based on the idea: The crucial act
here is to deduce a solution to a problem from a set of theoretical
principles." Theoretical principles refer to laws of formal theories.
268
PART
application worth trying. The control we have in mind has

various aspects. The person who applies the theory m ust be
enabled to understand and analyze ongoing situational realities,
to produce and predict change in them, and to predict and
control consequences both for the object of change and for
other parts of the total situation that will be affected. A nd as
changes occur, he must be enabled to be flexible in revising
his tactics of application and in revising the theory itself if
necessary. T o give this kind of control, the theory m ust provide
a sufficient num ber of general concepts and their plausible
interrelations; and these concepts m ust provide him with under
standing, with situational controls, and with access to the
situation in order to exert the controls. The crux of controlla
bility is the production and control of change through con
trollable variables and access variables.
Controllable variables. O ur concepts, their level of generality,
their fit to the situation, and their understandability give who
ever wishes to apply them, to bring about change, a controllable
theoretical foothold in the realities of terminal situations. Thus,
not only m ust the conceptual variables be controllable, but
their controllability m ust be enhanced by their integration into
a substantive theory which guides their use under m ost condi
tions that the user is likely to encounter. T he use of our
concepts may be contrasted with the unguided, ad hoc use
of an isolated concept, or with the use of abstract formal
categories that are too tenuously related to the actual situation.16
A t a lower level of generality, in much consulting done _by
sociologists to industrial firms, hospitals, social agencies, and the like,
what is usually offered by the sociologists is understanding, based upon
an amalgam of facts intuitively rendered by references to formal theory
and some loosely integrated substantive theory developed through con
tact with a given substantive area over the years. (Sometimes this is
abetted, as in consumer research, by relatively primitive but useful
analyses of data gathered for specific purposes of consultation.) Pro
viding that the amalgam makes sense to the client and that he can
see how to use it, then the consultation is worthwhile. Conversely, no
m atter how useful the sociologist may think his offering is, if the
client cannot see it then he will not find the consultation very use
ful. See also Zetterberg, op. cit.. Chapter 2.
269
F or example, the prime controllable variable of our study

is the awareness context. D octors and nurses have much
control over the creation, maintenance, and change of aware
ness contexts; thus they have much control over the resultant
characteristic forms of interaction, and the consequences for
all people involved in the dying situation. Also, the interactional
modes we have specified are highly controllable variables;
doctors and nurses deliberately engage in many interactional
tactics and strategies.
If a doctor contemplates disclosure of terminality to a
patient, by using our theory he may anticipate a very wide
range of plausibly expected changes and consequences for
himself, patient, family members and nurses. By using the
theory developed in C hapter 8, he m ay judge how far and in
what direction the patients responses may go and how to
control these responses. By using the theory in C hapter 3, he
may judge what consequences for himself, nurses and patients
will occur when the context is kept closed; and by referring
to C hapter 6, he may weigh these against the consequences
that occur when the context is opened. Also, he m ay judge how
advisable it is to allow the characteristic modes of interaction
that result from each type of awareness context to continue or
be changed. F rom these chapters he also may develop a wider
variety of interactional tactics than ordinarily would be in his
persona] repertoire. If maintaining a closed context will result
in too great a m anagement of assessment (an interactional
m ode) by the nurse which might decrease the patients trust
in the whole staff when he discovers his terminality it may
be better to change the context to allow the nurse to respond
differently.
The doctor m ay also review C hapters 9 and 10 for judging
to what degree opening the context by disclosure will lead to
problems in controlling family members, and how the disclosure
may affect their preparations for death. Resting this decision
upon our theory allows him much flexibility and scope of action
precisely because we have provided many general concepts
and their probable interrelations closely linked to the reality
270
PART
of disclosure, in order to guide the doctor in considering the

many additional situations that will be affected by disclosure.
Simply to disclose in the hope that the patient will be able to
prepare himself for death is just as unguided and ad hoc as
to not to disclose because he may commit suicide. To disclose
because the patient m ust learn, according to formal theory,
to take the role of a term inal patient, is too abstract a notion
for coping with the realities of the impact of disclosure for
all people concerned.
This example brings out several other properties of con
trollable variables and, thus, of our substantive theory. First,
the theory m ust provide controllable variables with m uch ex
planatory power: they m ust make a big difference in what
is going on in the situation to be changed. We have discovered
one such variable awareness contexts. As we have reiterated
m any times, much of what happens in the dying situation is
strongly determined by the type of awareness context within
which the events are occurring.
Second, doctors and nurses, family and patients are already
purposefully controlling many variables delineated in our sub
stantive theory. While the doctor exerts most control over the
awareness context, all these people have tactics that they use
to change or m aintain a particular awareness context. The
patient, for example, is often responsible for initiating the
pretense context. However, all these people are, in our obser
vation, controlling variables for very limited, ad hoc purposes.
O ur theory, therefore, can give staff, family and patients a
broader guide to what they tend to do already and perhaps
help them to be m ore effective.
Controllable variables sometimes entail controlling only
ones own behavior and sometimes primarily others behavior
the more difficult of the two. But, as we have tried to show,
control usually involves the efforts of two parties; that is con
trol o f the interaction between two people by one o r both. In
the dying situation it is not uncommon to see patient, family,
doctor, and nurse trying to control each other for their own
271
purposes. Those who avail themselves of our theory may have

a better chance in the tug-of-war over who shall best control
the dying situation.
In the hospital, material props and physical spaces are of
strategic importance as variables which help to control aware
ness contexts and peoples behavior.17 We have noted how
doctors and nurses use spatial arrangements of rooms, doors,
glass walls, rooms and screens to achieve control over aware
ness contexts. By making such controllable variables part of
our theory we have given a broader guide to the staffs pur
poseful use of them. Thus, to let a family through a door or
behind a screen may be more advisable than yielding to the
momentary urge of shutting out the family to prevent a scene.
Letting in family members may aid their preparations for
death, which in turn may result in a more composed family
over the long run of the dying situation.
Access variables. The theory m ust also include access
variables: social structural variables which allow, guide and
give persons access either to the controllable variables o r to
the people who are in control of them. To use a controllable
variable, one m ust have a means of access to it. F o r example,
professional rules give principal control over awareness con
texts to the doctor; therefore the nurse ordinarily has a great
deal of control in dying situations because of her considerable
access to the doctor, through or from whom she may try to
exert control over the awareness context. Professional rules
forbid her to change the context on her own initiative; they
require her to maintain the current one. Thus the organizational
structure of the hospital, the medical profession, and the ward
provide degrees of access to control of awareness contexts by
both doctor and nurses and our theory delineates this matter.
Family members have more access to a private physician than
17 Elements of material culture should not be neglected in de
velopment of substantive theory. Gouldner suggests they are the for
gotten man of social research : op. cit., p. 97.
272
PART
to a hospital physician; thus they may have m ore control over

the form er. Sometimes they can dem and that their private
physician keep a closed awareness context because of the
control they exert over him through the lay referral system
(upon which he may depend for much of his p ractice).18 The
patient has little access in the closed context to a doctor in
order to control changes of context. However, like the nurse,
he has m uch access to everyday cues concerning his condition
they exist all around him and he learns to read them better
and better. Thus, his access to strategic cues gives him an
opportunity to control his situation and we have discussed
at length how he can manage cues to gain controls. Access
variables also indicate how best to enter a situation in order to
m anage a controllable variable while not otherwise unduly dis
rupting the situation. Thus, we have delineated the various
alternatives that a nurse m ay use to gain control over the
nothing m ore to do situation in order to let a patient die.
CONCLUSION
Throughout our m onograph we have indicated many stra
tegic places, points and problems in dying that we feel would
profit from the application of our theory. By leaving these
short discourses on application in context we trust they have
had more meaning than if gathered into a single chapter.
We have made this effort to establish a practical theory
also because we feel, as many sociologists do and as Elbridge
Sibley has written: The popular notion that any educated
m an is capable of being his own sociologist will not be exorcised
by proclam ation; it can only be gradually dispelled by the
visible accomplishments of professionally com petent sociolo
gists. 19 By attem pting to develop a theory that can also be
applied, we hope to contribute to the accomplishments of
18 On the lay referral system see Eliot Freidson, Patients' Views of
Medical Practice (New York: Russell Sage Foundation, 1961), Part
Two.
18 The Education of Sociologists in the United States (New York:

Russell Sage Foundation, 1963), p. 19.
273
sociology. Social theory, in turn, is thereby enriched and linked

closely, as John Dewey rem arked thirty years ago, with the
pursuit and studied control of practical m atters.20
Two properties of our type of an applied theory m ust be
clearly understood. First, the theory can only be developed by
trained sociologists, but can be applied by either laymen
or sociologists. Second, it is a type of theory which can
be applied in a substantive area which entails interaction vari
ables. W hether it would be a useful type of theory for areas
where interaction is of no powerful consequence (th at is,
where large scale param eters are at issue, such as consumer
purchase rates, birth control, the voting of a county, desegragation of a school system, and audiences for T V ) remains
unanswered.
See Social Science and Social Control in Joseph Ratner (ed.),
Intelligence in the Modern World, John Deweys Philosophy (New

York: Modern Library, 1939), pp. 949-954.
Awareness and The Study of Social Interaction
15
Awareness and
the Study of
Social Interaction
We noted in the first chapter that this book would focus upon
a set of related questions: W hat were the recurrent kinds of
interaction between dying patient and hospital personnel?
W hat were the kinds of tactics used by personnel in dealing
with the patient? W hat were the conditions of hospital organi
zation under which interaction and tactics occurred? And in
what ways did they affect the goals and stakes of patient, family,
and personnel in the dying situation? In anticipation of the
pages to follow, we remarked that in finding answers to such
questions we had discovered that most of their variations could
be accounted for by what each party to the dying situation
was aware of about the patients fate. In consequence, we
would deal with these questions as they related to the powerful
explanatory variable of awareness, which we conceptualized
as awareness context. ( W hat each interactant knows of the
patients defined status, along with his recognition of the others
awarenesses of his own definition the total picture as a so
ciologist might construct it an awareness context. It is the
context within which these people interact while taking cogni
zance of it. )
In part II of the book, we discussed four types of awareness
context of great im port for what goes on around the patient :
closed, suspicion, m utual pretense, and open awareness. We
assumed, for simplicitys sake, that only two interactants were
274
275
involved in the given awareness context: the patient and the

hospital staff (taken as though all its members shared the same
aw areness). Each discussion was organized around a paradigm:
(1 ) a description of the given awareness context; (2 ) the social
structural conditions that enter into the context; (3 ) the conse
quent interaction, including various tactics and countertactics;
(4 ) the change of interaction from one type of awareness to
another; (5 ) the ways in which various interactants engineer
changes of awareness context; and (6 ) various conseqences
of the interaction for the interactants, for the hospital, and for
further interaction. The discussion in P art III was organized
around various general problems associated with awareness
context as they bear upon the patient, his family and the
hospital staff. In that latter discussion, the paradigm noted
above was left implicit but was the analytic bedrock upon which
discussion itself rested.
M any readers, notably the health professionals, have un
doubtedly read our chapters as realistic, albeit highly organized,
descriptions of the multi-faceted dying situation. But as we
have previously emphasized, the chapters can also be read as
an integrated substantive theory about awareness contexts as
those contexts relate to interaction in the dying situation. We
did not, in this book, concern ourselves with everything that
happens around dying patients, but only with the varied events
which occur in relation to awareness of the patients death.
Every page touches on some aspects of awareness as it affects
interaction around or toward the dying patient.
This kind of theory is, of course, grounded in research on
one particular substantive area (dying), and might therefore
be taken to apply only to that specific area. Theory at such a
level, however, may have im portant general implications. We
shall now detail the more general significance of our theory of
awareness contexts.
276
PART
FR O M SUBSTANTIVE TO FO R M A L TH EO R Y
In the preceding chapter we remarked that substantive
theory faithful to the empirical situation cannot be formulated
by merely applying a formal theory to the substantive area. A
substantive theory must be first formulated in order to see which
parts of diverse formal theories (originally developed for such
conceptual areas as deviance, status congruency, reference
groups, stigma, or hierarchy) can then further the substantive
formulation. A nd in its turn, substantive theory may help in
formulating formal theory. It may also contribute to the formu
lation of new formal theory grounded on careful comparative
research.
In developing our substantive theory of awareness contexts,
we utilized the strategy of choosing multiple comparison groups
to study (see A ppendix). Those groups were chosen by virtue
of the logic of our emerging analytical framework. The reader
will recall that we studied, simultaneously or in succession, a
series of hospital services designed to maximize differences and
similarities in the dying situation according to our emerging
substantive theory of awareness contexts. We suggested in our
opening chapter, and discussed in a recent article,1 that aware
ness contexts are not confined to the dying situation, but are
found generally throughout the full range of social interaction.
Consequently, if one wishes to develop a systematic formal
(o r general) theory of awareness contexts, he m ust analyze
data from m any substantive areas.
When advancing a substantive theory to a formal one, the
comparative analysis of groups is still the m ost powerful
method for formulating credible theory. The logic for dis
covering substantive theory which provided an efficient guide
to multiple groups in one substantive area also will provide
a guide for obtaining m ore data from many kinds of substantive
areas, in order to generate formal theory and to verify or
1 Glaser and Strauss, Awareness Contexts and Social Interaction,
American Sociological Review, 29 (October, 1964), pp. 669-679.
277
negate its hypotheses. Again, the constant com parison of many

groups quickly draws attention to their many similarities -and
differences, the analysis of which now generates a formal
theory.2
How this might be done with the theory of awareness con
texts can be briefly illustrated. Situations where awareness con
texts exist are, for instance, clowning at circuses, buying and
selling cars, hustling in pool halls, the passing of Negroes as
whites, spying as practiced by nations, and the m utual suspicion
of prisoners in Chinese war camps. Quick scrutiny of these situ
ations, as well as our earlier preliminary analysis of differences
between some of them and the dying situation, suggests several
properties useful for comparing groups within these situations.
The signs or indicators of an interactants status may vary in
how visible they are to the other interactants. Different numbers
of interactants can be involved: two, three, or m ore. Different
numbers o f groups can be represented by the interactants. The
ratios of insiders and outsiders present during the interaction
may vary: one patient and dozens of staff m em bers; five
hustlers and one m ark; one Negro, five wise people in the
know, and millions of white persons not in the know. The
statuses of interactants m ay also vary in hierarchical position
(sam e o r different status levels) as well as in degree of hierarchy.
And of course the stakes of the interaction may vary tre
mendously.
Comparisons of each property for diverse substantive
groups, as listed above, quickly leads to the development of
general categories and the form ulation of associated hypotheses.
Suppose that one focuses, for instance, upon the identifying
signs of status. Some signs are physical (skin co lo r), some are
behavioral (speech or gesture), some are m arks of skill (the
agility of the card shark), some are insignia (uniform s and
clothing), and so on. F or any given interactional situation
1 Barney Glaser and Anselm Strauss, Discovery of Substantive
Theory: A Basic Strategy Underlying Qualitative Research, American
Behavioral Scientist, 8 (February, 1965), pp. 5-12.
278
p a r t f o u r : c o n c l u s io n s
certain signs of status may be thought of as prim ary and others

as secondary: in America, skin color is the prim ary indicator of
Negro, just as genitalia are of respective sex. The secondary
signs those that strongly suggest status, especially in con
junction with prim ary signs would be, for Negro, kinky
hair and perhaps southern-style speech, and for sex they would
be clothing, hair style, and gesture.
The visibility of such signs depends on learned ability to
recognize them; for instance, many people have never learned
to recognize homosexuals, and others would not know an A m er
ican Indian if they saw one. Understandably, some interactants
may not even recognize the signs of their own status: for
instance, the dying person kept in the closed awareness context.
Signs can be manipulated, both crudely and with subtlety. F or
instance, they may simply be removed from vision, as when
stigmata are concealed. They can be disguised, as when kinky
hair is straightened or, as John Griffin did when passing for
Negro, skin color is changed temporarily with chemicals. Signs
can also be suppressed, as when an interactant chooses not to
flash signs that he is really an Am erican spy, or a JapaneseAmerican visiting in Japan chooses to talk in the native tongue
at a departm ent store so as not to be recognized as a rich
Am erican. These, of course, are all tactics for minimizing
potential recognition by other interactants.
Countertactics consist of eliciting im portant give-away
signs, for otherwise one may have to wait for and hope to
recognize those signs. Some countertactics for recognizing
persons who are suppressing identity depend upon passing
as a mem ber of that group (an FBI m an posing as a Com
m unist) o r on informers from within the group. Among persons
of similar status, conventional signs may be used to further
recognition; the deliberate use of these signs will vary depending
on whether outsiders are present or absent, and whether if
present they are wise (sym pathetic insiders) or not. There
are typical spatial locales where the gathered insiders can forgo
3 See Erving Goffman, Stigma (New York: Prentice-Hall, 1963).
279
efforts to disguise or suppress identifying signs. But they may

need (as with drug addicts) countertactics to avoid betrayal
even in such secluded places. It is worth em phasizing that
sometimes identifying signs need to be rectified as when a
m an m istaken for a thief must prove his innocence to bystanders,
or even to police and later to a court of law. Sometimes rectifi
cation is made, but it false! A nd sometimes, as we have seen
in mutual pretense, the dying patient in some sense rectifies
that notion by acting as very much alive and, given the am
biguity of m ost signs, other people act up to his rectification
until the signs are either so unambiguous th at the game is hard
to play, or he drops the pretense and admits who he really is.
A subjective, and subtle, accompaniment of such ambiguities
is when an interactants rectification is rejected and he himself
begins to doubt who he is, as in Nazi Germ any when gentiles
with faint Jewish lineage came to doubt their true identities.4
Such comparisons of diverse groups on the dimension of
signs quickly lead to both useful concepts and hypotheses
about this facet of a formal theory of awareness context. Just
as when one is developing a substantive theory, the hypotheses
will be concerned with such m atters as tactics and countertactics, as well as with the structural conditions for them, their
consequences, and so on. But it is im portant to understand
that this kind of inquiry can be furthered immensely by sys
tematic analysis not only of a single property but of combina
tions of properties', signs and stakes, for instance; or signs,
stakes, ratios of insiders-outsiders, and numbers of group repre
sentatives present at the interaction. (In essence, this is how
we proceeded when developing substantive theory in P art III
of our book.) This kind of analysis becomes increasingly en
riched, because it leads the researcher to ask, W here can I
find another com parison group that differs in one more speci
fied respect? W hen he finds that group, its examination leads
4 A rthur Millers perceptive novel, Focus (New York: Reynal and
Hitchcock, 1945), portrays this dilemma for a man who looked Jew
ish but had no Jewish lineage whatever.
280
PART
him to further verification or to fruitful qualification. By

such means, exceedingly complex and well-grounded formal
theory can be developed. It is precisely by such means that
one could extend our substantive theory of awareness contexts
upw ard (in abstractness) and outward (in generality). In
doing so, many m ore useful types of awareness context would
be generated and related to interactants behavior.
281
Unquestionably most sociologists tend to avoid the formu

lation of grounded formal theory; they stay principally at the
substantive level. Aside from the inherently greater difficulties
in working with high level abstractions o r in feeling confident
about broader generalities, we believe there are several reasons
why sociologists prefer to stay with substantive affairs.
First of all, a researcher tends to know one or two sub
stantive areas well, and feels increasingly comfortable as he
learns more about them over the years. The internal satisfac
tions and securities of such specialization are augmented by
the further rewards of being considered a m ature expert in a
specialized field rewards which emanate from colleagues and
the wider public.5 Furtherm ore, sociologists learn very early
the dictum that there is great difference between a dilettante
and a true pro. The latter knows his data inside out. This
conviction tends to keep sociologists from researching more
widely, and certainly from working more abstractly, than they
might. One perceives this vividly in the writings and remarks
of colleagues who feel they have to amass and comprehend
great amounts of data before they can safely claim findings.
(T he historians critical reaction to invasion of any well-worked
historical field by a social scientist is an extreme instance of the
attitude that the outsider does not know the field well enough
not to commit errors.)
Insofar as the sociologist concerns himself with formal
theory, he currently tends to handle it in several alternative

ways. He may set out to test, in a given substantive area, some
small portion of one or more formal theories, derived often
from prom inent theorists. Such studies are legion. He may
apply several formal theories to a substantive area which he
already knows well, seeking to give his m aterials greater mean
ing. Sometimes he does this as a post hoc enterprise, after the
data is collected; but sometimes the theories do direct at least
portions of his data collection.
A third aproach is to check out upon comparative materials
an im portant body of theoretical writing, as when R obert
Blauner systematically scrutinized a num ber of industries with
respect to their degree of alienation. 6 This latter type of
research is typically confined to careful verification or invalida
tion of the central guiding theory. B ut except where the old
theory seems to require qualification, this very m andate to
check out the existing formal theory under diverse conditions
tends to block the chances for developing new theory that are
afforded by comparative analysis. In short, such conventional
use of comparative analysis shows variations in an established
general theory and explains them. In contrast, our use of com
parative analysis generates and generalizes a new theory;
variations and explanations become p art of the process, not
the product.
Probably the most widespread use of formal theory is, how
ever, that when initiating specific researches a sociologist begins
with a loose framework of ideas, hunches, notions, concepts,
and hypotheses about the substantive area under consideration.
This framework is often linked with the researchers graduate
training (Parsonians from H arvard, M ertonians from Columbia)
and with his further experiences since graduation.
None of these uses of formal theory is to be decried; they
are, however, rather different enterprises than the formulation
of grounded formal theory through systematic study of multiple
5 Cf. Fred Reif and Anselm Strauss, The Impact o f Rapid Dis
covery Upon the Scientists Career, Social Problems, 1965.
Robert Blauner, Alienation and Freedom (Chicago: University of

Chicago Press, 1964).
ON TO F O R M A L TH EO R Y ?
282
PART
com parison groups. Perhaps the closest relative to such formula

tion is the kind of essay writing established m any years ago by
Georg Simmel, and nurtured by such contemporaries as Erving
Goffman and David Riesman, where the essayist with or
without systematic data before him develops a series of
general propositions of relatively high abstraction. Such writing
tends to be criticized as, at best, full of insights and, at worst,
as pure speculation. (Some insights may get tested by
more rigorously m inded social scientists.) From our viewpoint,
such writing is exceedingly valuable, but as theory it does
lack both integration of well defined concepts and sufficiently
credible grounding in careful comparative research.
The more prestigious style of logico-deductive, systematic
grand theorizing, in the hands of its most brilliant practitioners
is more than merely aesthetically satisfying; it also gives impetus
to considerable useful, precise testing of hypotheses. B ut it
provides no directive any more than a century ago when
Comte and Spencer were its spokesmen to a closing of that
embarrassingly noticeable gap between highly abstract theory
and the multitude of miniscule substantive studies so charac
teristic of sociology. It should be evident that we put greater
faith in grounded formal theory, of the kind outlined above,
to close that gap, for it is a type of formal theory that readily
fits whats going on in everyday situations. Possibly the main
benefit yielded by grand theories is their use of abstract models
(such as the various mathematical, system and equilibrium
m odels). The formulation of formal theory often requires guid
ance from such explicit models more than does substantive
theory, if only because the greater level of abstraction of its
concepts requires integration according to such models.
AW ARENESS CONTEXTS, IN TER A C TIO N A L
T H E O R Y AND R ESEA R C H
As we noted in the preceding chapter, the theory outlined
in this book entails interaction variables, because it was de
veloped to gain a greater understanding of and control over
283
interaction in the dying situation. M uch earlier we had remarked

that a central sociological problem is how to conceptualize and
account for social interaction and that many influential theorists
(W eber, Durkheim, Thomas, M ead, Parsons, and Goffman
among them ) had grappled with the problem.
If now we focus upon the relation of awareness contexts
to social interaction, what can be said about the writings of
such men? In general, they have tended, when discussing the
kinds of phenomenon which we have term ed awareness con
text, to attend only to one or two of the many possible
contexts. F or instance, M ead was consistently interested in
open awareness and Erving Goffman, though not solely con
cerned with the closed context, is perhaps at his best when
discussing the phenom ena which we would associate with
closed awareness.7 Of equal importance, such theorists tend
to emphasize only certain portions of the basic paradigm
developed in this book for the study of interaction. Sometimes
they discuss a given awareness context in relative isolation
from, or in only loose connection with, social structure. And
the relevant interactional tactics may not be fully explored.
Sometimes the consequences are barely noted, or only as they
bear on one or the other interactant, not necessarily in relation
to the social or organizational structure within which interaction
occurs.
Essentially the same strictures hold for sociological research
when awareness is central to the situation investigated.8 Also
in many research reports the readers are given passing inform a
tion which suggests that the paradigm of awareness might have
provided a very useful addition, even within the framework
of the researchers own analysis.9 A nd some substantive theories,
based upon extensive bodies of research, turn upon awareness
7 See our paper, Awareness Contexts and Social Interaction, op.
cit.
8 Ibid.; also see Ned Polskys description of hustler tactics in pool
halls: The Hustler, Social Problems, 12 (1964), pp. 3-15.
8 See M. Daltons description of organizational cheating in his Men
Who Manage (New York: Wiley, 1959); also Albert J. Reiss, Jr., The
Social Integration of Queers and Peers, Social Problems, 9 (1961),
pp. 102-120.
284
PART
contexts: in our estimation, these sometimes excellent writings

require m uch more extensive analyses of the specific and central
awareness contexts in order to increase plausibility and make
verification more operational.10
T o be sure, not all interactional research and analysis
requires a central focus upon awareness. We urge, however,
that such studies, irrespective of their specific focus and central
variables, include in their analysis a consideration of awareness
as a strategic general variable. Additional data gathering on
awareness is not thereby required, for most interactional
studies inevitably collect data bearing on awareness; the data
need only be analyzed in term s of awareness. Form al interac
tional theory, in general, could profit from such specific sub
stantive analyses, just as the latter might benefit from an
increasingly sophisticated interactional theory. In interactional
studies which do not centrally focus on awareness, the specific
paradigm which underlies the analysis in our book need not
be used in toto. Presum ably different enterprises require some
what different guiding schemes. However, we wish to emphasize
that o u r paradigm forces ones gaze upon two im portant fea
tures of interaction that are frequently underplayed. One
feature is the developmental character of interaction. The
second is the social structural context within which interaction
occurs. Perhaps a fitting close to this book is a re-emphasis upon
that second feature. In so m uch writing about interaction there
has been such neglect or incomplete handling of relationships
between social structure and interaction that we have no fear
of placing too much emphasis upon those relationships. The
careful reader of our book may even have noted that these
relationships are not entirely one-sided. O n occasion, the
course of interaction m a y partly change the social structure
within which interaction occurred, as when the transform ation
of an awareness context alters the structural conditions operat
i c / . David Matza, Delinquency and D rift (New York: Wiley,
1964), Chapter 2; Alfred Lindesmith, Opiate Addiction (Bloomington,
Indiana: Principia Press, 1947).
285
ing at a given hospital service so as to make it a different

place. This interplay of social structure and social interaction is
precisely what makes the study of interactional development
so richly rewarding, thus binding together two basic concerns
of our book and our theory.11
11 And, as we suggested in an earlier paper ( Awareness Contexts
. . . op. cit., p. 678), this is why our paradigm is so useful in relating
development and structure: how does one type of context lead to
one another; what are the structural conditionsincluding rules in
the relevant institutions that facilitate or impede existence of a context,
and changes in it; what are the-effects of a changing awareness con
text on the identity of a participant; why does one party wish to change
a context while another wishes to maintain it or reinstate it; what are
the various interactional tactics used to maintain or reinstate change;
and what are the consequences for each party, as well as for sustain
ing institutional conditions?
Methods of Collection and Analysis of Data
APPENDIX
Methods of Collection
and Analysis of Data
In our book, we have attem pted to develop an integrated
theory about awareness contexts as they pertain to dying in
hospitals. A fter a preliminary stage of eleven months this aim
governed further collection and analysis of data. As noted in
C hapter 1, fieldwork data were gathered at a num ber of hospitals
in the Bay area of San Francisco over a period of three years.
Fieldwork procedures are well known to sociologists, but the
health professionals may be less acquainted with them. We shall,
therefore, write this appendix with both audiences in mind,
emphasizing especially the features of field m ethod which make
it extremely appropriate for developing integrated theory about
a substantive topic like dying. 1 In presenting our views, we
shall give information about what we did and how we did it,
and at the same time we shall give a more general commentary
on the characteristics of a credible and pragmatically useful
substantive theory.
First, we should note that our concept of awareness
context was foreshadowed by personal experiences of both
authors. Five years before the study began, Strauss participated
in what was at first a closed and then a mutual pretense aware1 For extended discussions, from which this Appendix is drawn, of
discovering and developing substantive theory through the collection
and analysis of qualitative data, see Barney G. Glaser and Anselm L.
Strauss, Discovery of Substantive Theory: A Basic Strategy Under
lying Qualitative Analysis, American Behavioral Scientist, 8 (1965),
pp. 5-12;.and Barney G. Glaser, The Constant Comparative Method of
Qualitative Analysis, Social Problems, 12 (1965), pp. 436-45. Extensive
citation of related source materials on qualitative analysis may be found
in these papers. See also the remarks on method in the Preface.
286
287
ness context surrounding the dying of his mother. Two years

later, he found himself deeply involved with others in what
he perceived as an elaborate collusive game designed to
keep a dying friend unaware of his impending state (closed
aw areness). Both personal experiences brought to his attention
the problems and consequences of lingering deaths as com
pared with speedy deaths. Still later, during brief observation
on a geriatrics ward at a state mental hospital, he was impressed
by the immense importance of the ward structure for the staffs
handling of dying patients, m ost of whom were comatose or
senile. Consequently, when he initiated a study of dying in
hospitals he started by focusing on expectations about certainty
and timing of dying and awareness of these expectations
as they affect how the staff manages and how the hospital
organization impinges on the dying of patients. A fter a half
year of preliminary fieldwork the project was funded by the
Public Health Service, and five months later Glaser joined the
project. Glaser, whose father had recently died, had been
especially impressed by the importance of death expectations
and who was aware of them as they affected how family
members discussed his fathers illness, how the members were
handled by the doctors, and how everyone treated his father.
He was also struck by the hopelessness of the nothing more
to do phase of his fathers dying.
Shortly after Strauss and Glaser joined forces, they sys
tematically worked out the concepts (and types) of death
expectations and awareness contexts, and the paradigm for
the study of awareness contexts. Thus, a concern with death
expectations and awareness guided the preliminary data col
lection; the systematic formulation of these concepts and of
the paradigm governed further data collection and the ensuing
analysis; and this book completes the formulations on the
theory of awareness of dying. These formulations should guide
others both in research on dying and in developing other
theories which must take into account the awareness of people.
We shall begin our discussion of m ethod by noting that
288
APPENDIX
fieldwork is admirably adapted to the development of systematic

substantive theory. Fieldwork allows researchers to plunge
into social settings where the im portant events (about which
they will develop theory) are going, on naturally. T(ie
researchers watch these events as they occur. They follow them
as they unfold through time. They observe the actors in the
relevant social dramas. They converse with or formally inter
view the actors about their observed actions. O ur own fieldwork
procedures did not differ significantly from those described
in the literature.
Num erous fieldworkers have also rem arked that the form u
lation and testing of hypotheses inevitably begins early in the
research. The earliest hypotheses quickly tend to become
integrated, to form the basis of a central analytic framework
such as is developed in our book. The analytic framework gets
expressed in field notes in two forms: analytic comments are
written directly into the field notes; and they are written into
occasional memos addressed specifically to m atters of analysis.
A process of coding also begins at or near the outset of the
research, as the researchers think systematically about the data
in accordance with their basic analytic categories. In such
qualitative analysis, there tends to be blurring and intertwining
of coding, data collection and data analysis, from the beginning
of the investigation until near its end. As various segments
of the analytic framework get built during chronologically
different stages Of the fieldwork, more data need not be
gathered nor analysis be rethought for the segment, unless fur
ther theoretical work m akes it necessary.2 W hen the researchers
are convinced that their analytic framework forms a systematic
substantive theory, that it is a reasonably accurate statement
of the m atters studied, and that it is couched in a form
possible for others to use if they were to go into the same
field, then they are ready to publish their results. This is a
schematic but accurate description of how our own analysis
was accomplished.
See Glaser, Constant Comparative . . . , ibid., for specific
procedures of coding and analysis.
289
The formulation of substantive theory is both facilitated

and made additionally credible by the strategy of systematically
studying several comparison groups. This strategy which in
our opinion should be more extensively employed is guided
by the logic of the emerging analytic framework. In our study,
we began by observing on a prem ature baby service, intention
ally minimizing the patients awareness of dying and maximizing
the expectedness of deaths (m ost premie deaths are.expected).
We then observed on a cancer service, intentionally maximizing
the lingering and nothing more to do aspects of dying as well
as several varieties of awareness of dying. In this fashion, we
gradually studied various wards at different hospitals, some
of them simultaneously.3 O ur choice of these sites and the
directions of our observation there accorded with our aims
of verifying (in diverse settings) our initial and later hypotheses,
of suggesting new hypotheses, and providing new data either
on categories or combinations of categories.
Using comparison groups maximizes the credibility of
theory in two fundamental ways. First, by detailing precisely
the many similarities and differences of the various comparison
groups, the researcher knows, better than if he had studied
only one or a few groups, under what sets of structural condi
tions his hypotheses are minimized and maximized; and hence
to what kinds of social structures his theory is applicable. In
increasing the scope and delimiting the generality of his theory,
he saves his readers work. Frequently, readers of fieldwork
must figure out the limitations of a published study by making
comparisons with their own experience and knowledge of
similar groups. By comparison,, they figure that the reported
material jibes just so far and no further for given structural
reasons. By using multiple comparison groups, much of this
burden of delimiting relevant boundaries for the theory is
T h e hospitals were: a teaching hospital, a VA hospital, two county
hospitals, a private Catholic hospital, a state hospital. The services, or
wards, included: geriatrics, cancer, premie, intensive care, medicine,
pediatrics, neurosurgery, urology, emergency, etc. The field workers
spent between two and four weeks on each ward. In addition, a third
researcher later contributed extensive interview data on nursing stu
dents encounters with death.
290
APPENDIX
lifted from the readers shoulders. In short, replication is built

into the research. A second way that multiple comparison groups
maximize credibility is by helping the researcher to calculate
where a given order of events or incidents is m ost likely to
occur or not to occur (in our study, on what specific w ards).
This calculus provides him with an efficient logical guide to
groups whose study will yield more data to fill in gaps in
theory and to verify hypotheses.
Multiple comparison groups also perm it and generate the
speedy development of analysis. The constant comparison of
many groups rather quickly draws the observers attention to
many similarities and differences among groups: these con
tribute to the generation of theoretical categories, to their full
range of types or continuum, their dimensions, the conditions
under which they exist more or less, and their m ajor conse
quences. In addition, the observed differences and similarities
speedily generate generalized relations among the theoretical
categories, which in turn bcome the hypotheses soon integrated
into the substantive theory.
When the field researcher decides to write for publication,
then he faces the problem of conveying the credibility of his
discovered theory so that readers can make sensible judgments
about it. The problem of conveying credibility is divisible into
two sub-problems. The first is that the researcher m ust get
readers to understand his theoretical framework. This is gen
erally done by giving an extensive abstract presentation of the
framework and its associated theoretical statements, usually
at the beginning and end of the publication but also in seg
ments throughout the publication. We have written theory on
almost every page of our book.
The related sub-problem is how to describe so vividly the
social world studied that the reader can almost literally see
and hear its people but in relation to the theoretical frame
work. To do this, the researcher ordinarily utilizes several of
a considerable armam entarium of standard devices and we
have done so (using pertinent phrases dropped by informants,
291
briefly describing scenes, occasionally quoting sentences or

segments of on-the-spot field notes).
The first and second sub-problems of conveying credibility
are reflected in the types of concept chosen for writing sub
stantive theory. With regard to the first problem, the researchers
concepts are analytic sufficiently generalized to designate the
properties of concrete entities, (not the concrete entities themseves). With regard to the second problem, his concepts also
are sensitizing they yield a meaningful picture with apt
illustrations which enable one to grasp the reference in terms
of ones own experience. O ur readers will readily recognize that
our concepts are both analytic and sensitizing.
Several aspects of the presentation enter into how the
reader, in turn, judges the credibility of the theory that the
writer is trying to convey. First of all, if a reader becomes
sufficiently caught up in the description to feel that vicariously
he has been in the field, then he is likely to be more kindly
disposed toward the theory than if the description seems flat or
unconvincing. In the case of our theory, many readers will
have had similar experiences to the people depicted in our
book, and will come to the reading armed with numerous, and
usually realistic, images. Second, a judgment of credibility will
also rest upon assessments concerning how the researcher came
to his conclusions. Readers will note, for instance, what range
of events the researcher saw, whom he interviewed, who talked
to him and under what conditions, and perhaps what experiences
he had and how he might have appeared to various people
whom he studied. That is, readers will assess the types of data
utilized from what is explicitly stated as well as from what
can be read between the lines. It is absolutely incumbent to
make such judgments about reports based on fieldwork, partly
because the publication could conceivably be a complete fabrica
tion, but more usually because any analysis may require some
qualification.
Such qualification, we may term the discounting process.
Readers surely discount aspects of many, if not most, analyses
292
APPENDIX
which are published (whether resting upon qualitative or

quantitative d ata). This discounting takes several forms: the
theory is corrected because of one-sided research designs,
adjusted to fit the diverse conditions of different social struc
tures, invalidated by the conditions for other structures through
the readers experience or knowledge, and deemed inapplicable
to yet other kinds of structures.
This ongoing discounting process of qualification by the
reader allows the researcher to write his theory in general form,
because the researcher knows that the reader will make the
necessary corrections, adjustments, invalidations, and inapplica
tions when thinking about o r using the theory. These are
qualifications that the researcher could not begin to cover for
even a small percentage of one type of reader and, more
im portant, they are qualifications which the researcher must
gloss over or ignore in order to write a substantive theory of
some generality. O ur own theory, for instance, m ust be
qualified to fit various types of hospitals and wards that we
did not study their num ber is legion. On the other hand,
the theory is of sufficient generality, and sufficiently grounded
in careful fieldwork, so that the discounting process can usefully
proceed. In sum, we and our readers share a joint responsibility.
Probably we need to say a final word about the question
of rigorous findings versus credible theory. The presenta
tion of substantive theory, developed through analysis of
qualitative data, is often done at a sufficient level of plausibility
to satisfy most readers. The theory can be applied and adjusted
to many situations with sufficient exactitude to guide thinking,
understanding and research. Given certain structural conditions
under which sociologists work (such as designing specific
action programs, or working in well-developed substantive
areas), then more rigorous testing may be required to increase
the plausibility of some hypotheses. Under these conditions, it
should be a m atter of empirical determination as to how the
further testing can best be accomplished whether through
more rigorous or extensive fieldwork, o r through other methods.
293
The two essential points in this decision about m ethod are:

first, the testing should be more rigorous than previously (but
not necessarily the most rigorous of all m ethods); and second,
the more rigorous aproach should be compatible with the
research situation in order to yield the m ost reliable findings.
M any researchers share an ideological commitment to quanti
tative methods, but an increased, required level of plausibility
should be based on the m ethod o r methods best suited to
the socially structured necessities of the specific research situa
tion.
We wish to emphasize that the type of substantive theory
developed in our book is often of great practical use long before
the theory is tested with great rigor. Both informed laymen and
social scientists often manage to profit quite well by the kind
of analysis represented in these pages. In the case of our own
theoretical analysis, readers should regard it as only the
beginning of the search for fruitful ways of understanding,
explaining, and controlling interaction th at pertains to dying
patients.4
4 As we remarked in the first chapter, this book does not represent
all that we know about such interaction.
Index
Abstract models: mathematical system,
282; equilibrium, 282
Acceptable style of dying, 86, 89-93,
103, 105; see also Appropriate,
Behavior, Improper conduct, Proper
conduct
Accepting death, 3, 62, 111, 121, 126,
127-132, 133; not accepting, 5
Access variables, 271-272
Accident patients, 48, 83, 170
Accountable, 5, 169
Acknowledgment: of awareness, lOfn,
79, 134, 198; not pretense, lOfn; pre
tense, lOfn, 11; see also Awareness
context
Action, 11, 13, 18, 34, 52, 54, 62, 68,
70, 74, 79, 81, 102, 104, 120, 181,
182, 186, 196, 200; see also Behavior,
Improper conduct, Proper conduct,
Response
Acute disease, 7
Addict, 210, 279
Admission of truth, 40
Age: nurse-patient same age, 38, 239;
see also Child, Geriatric patients,
Pediatric ward, Premature babies,
Young patient
Agreement, 10, 39; see also Consensus,
Disagreement
Aide, nurses, 59, 63, 143, 166, 207
Alertness: of nurse, 23, 40, 48; of
patient, 49
Alienation, 281
All right, 36
Allies, 32, 34, 52
Ambiguities, 36, 79-106, 176, 182, 279
Analytic theory, 263, 265, 291
Anger, 57, 88, 91, 92, 106, 110, 161,
234
Announcement, 45, 53, 54, 55, 94, 108,
124, 149, 151, 152, 153, 154, 168,
170, 176, 196, 248, 249; see also
Disclosure, Formal declaration
Anxiety, 55, 100, 126, 142, 145, 229
Apathy, 89, 91
Appropriate, 80, 82, 93, 149
Approval, 103
Asia, 31, 46, 85, 162fn
Ask your physician, 36, 57, 241
Assessment, 55, 95
Assessment management, 34-39, 41
Attention, 157-158, 165, 185, 186

Attitude, 81, 83, 84
Authority, 143, 148; see also Control,
Power, Status
Autopsy, 26, 150, 252, 253
Avoidance, 5, 226, 227, 229, 230, 233,
237, 238, 240, 245, 251, 252, 254;
see also Contact, Emotional distance,
Inaccessibility of physician, Isolate
Awareness context, 10, 11, 12, 13, 26,
47, 60, 61, 64, 67, 137, 180, 184,
191, 196, 210, 217, 218, 269, 271,
274, 276, 280, 283-285, 287; see also
Closed awareness, Mutual pretense,
Open awareness, Suspicion awareness
Babies, 29
Back regions, 242fn
Badly, taking death, 45, 96
Bargain, 94, 96
Bates, Alan P., 161fn
Be brave, 121
Becker, Howard S., 4fn, 19fn, 178fn
Behavior, 39, 63, 82, 83, 84, 85, 86,
87, 88, 91, 92, 96, 107, 175, 217,
226, 243; see also Action, Crying,
Improper conduct, Proper conduct,
Response
Bensman, I., 66fn
Betrayed, feels, 46
Biderman, Albert, 63fn
Biography, fictional, 32-34, 38
Blauner, Robert, 281, 281fn
Blessing he died, 85, 215, 216; see
also Collective mood
Blumer, Herbert, 14fn, 260fn, 263fn
Body changes, 48, 165, 168, 233; see
also Deterioration
Breakdown, 147, 150, 163, 171, 172,
174, 184, 222, 240, 248, 255
Breaking through, 91
Bucher, Rue, 14fn
Calm resignation, 130

Cancer, 18, 20, 24, 25, 49, 70, 78, 84,
94, 112, 120, 124, 125, 128, 132,
134, 136, 188, 193, 194, 210, 215,
217, 228, 230, 289
295
296
Cant talk, 213-214
Cardiac patient, 44, 48, 98, 134, 154fn
Carpenter, K. M., 165fn
Case, 185-188, 189, 195, 201
Catch on, 48
Catholic: hospital, 20, 64, 78, 171, 201,
243; patient, 39fn, 248; sacrament,
39fn
Caudill, William, 33fn
Censor, 37
Ceremonies, 59, 90, 104
Certain suspicion, 48, 55
Certain to die on time, 24, 25
Certainty of death, 18, 47, 55, 89; see
also Death expectations
Change, 15, 25, 32, 49, 60-61, 75, 76,
80, 89, 93, 103, 111, 122, 127, 135,
139, 155, 179, 196-203, 204, 205,
206, 217; see also Social change,
Transitions of awareness
Chaplain, 21, 25, 61, 70, 75, 91, 96, 97,
98, 103, 121, 127, 143, 147, 171,
172, 212, 214, 235, 243, 247, 248
Charity Hospital, 166
Charts, medical, 53, 55, 222
Chattering, 54, 57
Cheerful, 57, 86, 89, 125, 141, 207, 245
Child, 33, 65, 82, 100, 108, 137, 188,
191, 237, 241, 254
Chinese, 81
Chronic disease, 7, 22, 48
Cicourel, Aaron, 260fn
City of Hope Hospital, 166
Claim, 58, 59
Clinical experience, 30
Closed awareness, 10, 11, 29-46, 48, 49,
56, 59, 60, 64, 67, 80, 105, 111, 132,
134, 137, 139, 144, 153, 154fn, 155,
157, 158, 176, 181, 182, 186, 190,
199, 207, 210, 216, 217, 240, 242,
243, 269, 272, 283, 286, 287; see also
Awareness context, Transitions of
awareness
Closed suspicion, 47, 55, 61, 241; see
also Suspicion awareness
Cloward, Richard, 262fn
Coaching, 40, 91, 92
Coax, 91, 148
Cobb, Beatrix, 131fn
Coding, 288
Cohen, Albert, 262fn
Coleman, James, 178fn
Collaborate, 141, 144, 197
Collection, methods, 286-293
Collective mood, 190, 204, 208, 215,
216, 217, 224, 226, 227, 229, 234,
236, 244-246, 248, 255; see also
Accepting death, Forgetting, Sadness
AWARENESS OF DYING
Collier, Richard, 52fn

Collusive, 31, 54, 60, 62, 161fn, 287
Comatose, 9, 26, 37, 40, 42, 62, 76, 82,
84, 85, 99, 107, 108, 110, 123, 174,
184, 191, 204, 233, 255; see also
Socially dead
Comfort, 5, 23, 37, 40, 41, 64, 76,
125, 144, 157, 159, 166, 177, 178,
179, 180, 184, 185, 186, 187, 193,
194, 195, 196, 197, 200, 204-225,
229, 235, 236, 237, 238, 239, 245,
247, 252, 253; see also Medication,
Nursing care, Sedation, Snow a
patient under
Coming along fine," 36
Command, 90, 91
Commitment, 46, 53, 96
Communication, 176, 214, 240; block
ing of, 132, 134; see also Conversa
tions, Talk
Comparative analysis, 279, 281, 289,
290
Compassion, 41
Complaints, 73, 86, 88, 91; no com
plaints, 213
Composure: of doctor, 149; of family,
136, 144, 155, 167, 168; see also
Scenes; of nurse, 226-258; of patient,
86, 93; see also Breakdown, Crying
Comte, A., 282
Condition, physical, 20-32, 35, 36, 41,
42, 43, 48, 60, 69, 75, 78, 108, 140,
141, 144, 145, 151, 154, 155, 159,
171, 174, 176, 196, 204, 229, 234,
239, 241; see also Body changes,
Status
Conditions to maintain awareness con
text: of family, 138-139, 145-146
Confidantes, 52
Confirm, 11, 31, 50, 51, 54, 55, 56, 60,
61, 63, 154; see also Validating
Conflict, 100, 160, 201; see also Con
test, Disagreement
Confrontation, 5, 10, 53, 56, 67, 80,
104, 124, 125
Congruence, 80, 105
Conjecture, 55, 60, 61
Conscious, 29, 30; see also Sentient
Consensus, 10, 103; see also Agree
ment, Disagreement
Consequences, 11, 13, 47, 80, 82, 83,
122, 154, 160, 199, 202, 239, 250,
254, 269, 287; of closed awareness,
42-46; mutual pretense, 77-78; open
awareness, 103-106; suspicion aware
ness, 62-63
Contact, 40, 45, 46, 119, 130, 134, 139,
144, 167, 208, 233, 237, 239; see
In d ex
also Avoidance, Emotional distance,
Inaccessibility of physician, Isolate

Contest, -47, 50-53, 57, 59, 60, 76, 124,
134, 145
Contingencies, 100-103, 146, 166, 168,
180
Contradictions, 53, 54
Control, 47-63, 92, 93, 110, 121, 122,
124, 129, 131, 141,142; of patient,
7, 207, 211; of staff, 39, 40,41, 136,
137, 151, 156, 157, 164, 174, 208,
223, 225, 236; of structure, 259, 267272
Controllable variables, 268-271
Conversations, 3. 4, 5, 11, 37, 38, 41,
42, 44, 50, 53, 54, 55, 56, 58, 62,
67, 74, 82, 98, 104, 111, 162, 176,
229, 232, 237; see also Talk
Cooley, C. H., 13
Cooperate, 43, 87, 88, 92, 94, 103, 122,
234
Cope, 5, 137, 146, 155, 246, 253; be
havior, 63, 64; idiosyncratic modes,
5; standardized modes, 5
Corpse, 113, 114
Coser, Rose Laub, 156fn, 178fn
Counter-cues, 23, 181
Counter-tactics, 11, 53, 55-58, 141,
143, 211, 227, 237, 278, 279
Course of disease, 80
Covert suspicion, 48, 61
Crisis, 3, 182, 230, 231, 232, 233
Critical list, 20, 21, 147, 204, 206
Critical patient, 20, 150; see also Dying
patient, Imminent death, Impending
death
Crying, 73, 88, 91, 105, 139, 159, 164,
242, 253
Cues, 11, 19, 20, 21, 22, 23, 26, 140,
168, 205, 207, 217, 272; behavioral,
38; collective, 20; consistent, 39;
deny, 22, 60; doctors, 19, 20, 23;
giving of, 23, 36, 37, 38, 42, 139,
181; inconsistent, 39, 234; nurses, 21,
22, 56, 60, 61, 95, 251; obvious, 20,
42, 50, 153, 154, 155; patient, 22,
54, 55, 60, 95; physical, 21, 22, 36,
127, 132, 241; reduce, 37; seek, 51;
significant, 53, 54; social, 127; spa
tial, 54, 56; temporal, 21, 22, 36, 54,
127, 133, 241; validating, 60
Cure, 40, 87, 170, 171, 186, 190, 198,
253
Dalton, Melville, 283fn
Data analysis, 286-293
Data processing, 9, 13
297
Davis, Fred, 13, 19fn, 26fn, 41fn,
132fn, 150, 172fn, 182fn, 245fn
Death: long term, 41, 136, 137, 138,
170; open, 43, 76; short term, 137;
unexpected, 26, 44, 151, 152, 168,
177, 200, 250-255; see also Linger
ing, Shock, Surprise
Death expectations, 16-28, 58, 79, 80,
93, 102, 125, 171, 180, 188, 206,
220, 250, 255; certain death (known
time), 19, 24, 82, 123, 151, 170, 173,
227, 287; certain death (unknown
time), 19, 24, 168, 169, 179, 185,
188, 190, 191, 194, 235-246; chang
ing, 23-26, 147, 148; uncertain death
(known time question resolved), 19,
25, 230; uncertain death (unknown
time), 19, 25, 153, 177, 179, 227234; see also Certainty of death
Death sentence, 154
Death talker, 242
Death watch, 158, 246-248
Decision making, 166, 182, 191, 197,
199; see also Disclosure
Deduction, 261, 282
Defensive, 47
Definition of dying, 9, 10, 11, 17, 51,
82; nurses', 23; patients, 23; social,
16-28
DeGramont, Sanche, 52fn
Degree: of awareness, lOfn; convinced,
18; of preparation, 167-174; of sus
picion, lOfn, 11, 60; of unawareness,
lOfn
Demand, schedule of medication, 211,
212, 213
Deny, 56, 59, 60, 63, 121, 126, 127,
132-135, 171, 174, 208
Depressed, 94, 121, 126, 135, 186, 245
Deserves to die, 82, 83, 110
Deterioration, physical, 75, 79, 85, 101,
102, 103, 105, 184, 198, 213, 221;
see also Body changes
Dewey, John, 259fn-260fn, 273
Diagnosis, 4, 7, 9, 18, 19, 20, 112, 148,
177
Died without knowing, 44
Diet, 41, 87, 134, 184, 204
Difficult patient, 85; see also Noncooperative
Dignity, 80, 86, 103
Direct disclosure, 20, 22, 30, 31, 127135, 151; see also Announcement,
Disclosure
Dirty work, 148
Disadvantage, 103
Disagreement, 9, 10, 86, 89, 195, 221,
222, 223; see also Conflict
298
Disapproval, 81, 84, 93
Disclosure, 5, 6, 12, 19, 22, 26, 30, 35,
41, 42, 59, 119-135, 145, 146, 150,
193, 269; decision for, 6, 39, 42, 48,
166; doctor's, 19, 20, 30, 42, 48, 50,
139, 143, 147, 179; family, 31, 42,
167, 196; gentle, 125, 126, 148, 151,
168, 176; guard against, 29, 31, 32,
35, 36, 41, 44, 48, 62, 78, 101, 110,
243; nurses, 19, 41, 42, 151, 256;
sharp, 125, 126, 150, 152, 153, 176;
unwitting, 37, 39, 40, 158; when, 6,
112; see also Announcement
Discounting, 35, 107-118, 200-203
Divert, 37
Division of labor, 63, 204, 206, 242,
243
Do something care, 186-188
Doing all we can, 149, 188, 196, 231
Doubt, 33, 195, 217
Drift routes, 164, 233
Drowned out, 159
Dulles, Allen, 52fn
Dumping ground, 178
Dying patient, 82, 84, 85, 87, 88, 89,
92, 94, 132, 139, 146, 151, 155, 160,
165, 166, 172, 177, 178, 188, 197,
210, 226, 230, 239
Education, 146
Ehrlich, Danuta, 14fn
Elicit, 51, 53, 54, 95
Emergency: conditions, 146, 165, 200;
measures, 112, 196, 218, 221
Emergency ward, 18, 25, 39, 83, 139,
152, 154, 164, 170, 230
Emotional care, 143, 149, 172, 210
Emotional distance, 46; see also Avoid
ance, Contact
Empathy, 90, 104
Equipment, 186, 190, 191, 195, 197,
200, 201, 202, 219, 220, 230, 244
Ethnic status, 146
Etzioni, Amitai, !22fn
Europe, 85
Euthanasia, 3, 209, 222
Expert testimony, 4, 159, 195
Extended kin, 31
Facial expression, 33, 37, 39, 42, 56,
71, 142, 237
Facing, 82, 86, 87, 88, 103, 104
Failure to discount, 113-115; see also
Discounting
Family, 4, 6, 8, 10, 11, 20, 21, 24, 25,
42, 43, 44, 46, 49, 56, 62, 70, 72,
73, 75, 76, 90, 92, 93, 95, 96, 97,
AWARENESS OF DYING
98, 100, 103, 104, 110, 113, 114,

127, 128, 134, 188, 191, 192, 196,
197, 199, 212, 226, 229, 232, 233,
235, 239, 243, 271; aware, 19, 108,
155-176, 182, 183, 184, 189; sus
picions, 139-145; unaware, 136-154,
186, 196
Farberow, Norman L., 3fn, 119fn,
129fn
Fatal illness, 49, 123-125, 128, 131, 193
Fate, 8, 18, 31, 62, 78, 80, 130, 131,
152, 222, 228, 229, 230, 231, 232,
239, 240, 243, 244, 246
Favorable information," 142
Fear, 75, 87 , 88, 91, 101, 102, 104,
226
Feedback, 241
Feifel, Herman, 3fn, 5fn, 30fn, 31fn,
119fn, 126fn
Festinger, Leon, 263fn
Fieldwork, 8, 266, 286, 288-293
Finances, 24, 122, 198, 199, 239
Fitness of theory, 259, 260-262
Flash cues, 22, 49, 54; see also Signal
Foote, Nelson, 59fn
Forgetting, 249-250, 253, 254, 255
Formal declaration of patient status,
20, 23; see also Announcement
Formal theory, 276-285
Fox, Renee, 19fn, 30fn, 43, 78fn, 92fn,
105, 131fn, 184fn, 185fn, 186fn,
190fn, 192fn, 207fn
Freidson, Eliot, 14fn, 181fn, 187fn,
272fn
Frustration, 160, 166, 191, 202, 215,
243, 251
Full life, 129
Fulton, Robert, 6fn, 126fn, 129fn,
146fn, 183fn
Funeral preparations, 113, 150
Future-oriented, 38, 69, 72, 128, 132,
134, 233
Garver, I., 66fn
Geer, Blanche, 4fn, 19fn, 178fn
Generality of theory, "2$9r~265-267
Geriatric patients, 48, 65, 70, 78, 84,
86, 102, 109, 124, 155, 164, 174, 287
Gestures, 50, 54, 65, 67 , 74, 76, 93,
111, 152, 157
Given up," 45, 76, 84, 89
Glaser, Barney G., 5fn, lOfn, 24fn,
29fn, 38fn, 130fn, 178fn, 216fn,
219fn, 228fn, 276fn, 277fn, 283fn,
286fn, 287, 288
Go to pieces, 30, 69, 121; see also
Breakdown
Index
Goals, 176, 177, 179, 188, 190, 193,
194, 195-203, 204, 205, 206, 208,
215, 218, 225, 228, 235; see also
Comfort, Recovery, Save the patient
Go-between, 52
Goffman, Erving, 9, 13, 14, 14fn, 29,
29fn, 52, 108, 122fn, 161fn, 219fn,
242fn, 278fn, 282, 283
Going fast, 21
Good will comfort, 206-208, 238
Goudeket, Maurice, 104fn
Gouldner, Alvin, 66fn, 216fn, 262fn,
264fn, 271fn
Gouldner, Helen, 66fn
Gracefully, 225, 245
Gracelessly, 44, 62, 101, 105, 180
Grand theorizing, 282
Grief, 44, 139, 141, 168, 169, 171, 172,
173-174, 184, 199
Griffin, John, 17, 278
Guilt, 173, 237, 245, 250
Gunther, John, 74fn, 131fn
Hagstrom, Warren O., 178fn
Handle, 45, 63, 82, 135, 162, 167, 168,
196, 212, 242, 243, 248, 255, 287
Hard, taking death, 45
Helpers of the family, 171-172
Helplessness, 191, 202, 214, 236, 238,
247
Heroics, 25, 111, 159, 178, 197, 198,
200, 201, 219
Hint, 42, 48, 53, 59, 94; see abo
Disclosure
Hobbes, Thomas, 14
Hope, 24, 108, 168, 169, 170, 186, 187,
190, 192, 212, 229; lose all, 121, 122;
none, 24, 84, 126, 131, 177, 186, 200,
201, 223, 244, 287
Hospital, county, 45, 48, 78, 81, 164,
187, 289fn
Hospital, private, 24
Hospital administrators, 24, 198
Hospital milieu, 54
Hospital organization, 8, 18, 19, 23fn,
31, 39, 49, 50, 57, 64, 78, 172, 212,
271,,287
Hospital stay, 21, 41, 49, 50, 54, 84,
166, 182
Hostile, 88, 91, 156
Hughes, Evqrett C., 4fn, 19fn, 148fn,
178fn
Icheiser, G., 13
Ideal, medical, 201, 202, 215, 216, 219,
225
Identification, 38
299
Identity, 13, professional 59
Ideology, 30, 64, 99, 100
Illness, 82, 84, 133, 170, 239
Imminent death, 38, 64, 95, 102, 113,
126, 192, 196, 248
Impatience, 57
Impending death, 4, 5, 11, 18, 20, 26,
30, 31, 32, 34, 38, 42, 50, 55, 67,
70, 75, 76, 79, 82, 98, 101, 103, 107,
109, 120, 127, 132, 133, 134, 135,
141, 147, 151, 154, 155, 166, 167,
174, 189, 196, 225, 240, 251, 256
Improper conduct, 83, 84, 85, 86, 96;
see also Action, Behavior, Proper
conduct
Inaccessibility of physician, 41, 45; see
also Avoidance, Contact
Indian, 81
Indicators, 53, 55; of pain, 213; see
also Cues, Signs
Induction, 261, 267
Ineptness, 5
Information, 49, 50-53, 124, 136, 137,
138, 142, 143, 145, 153, 155, 159,
174-176, 197, 204, 232, 251; see abo
Disclosure
Informer, 52
Inquiry tactics, 23-55, 57; see abo
Questions, Tactics
Insiders, 8, 9, 51, 277
Insight, 87
Intense work, 230-233
Intensive-care ward, 25, 48, 49, 54, 64,
137, 141, 142, 144, 164, 178, 201,
219, 228, 230, 231, 249
Intensive living, 131, 132
Interactants, 16, 274, 277
Interactional theories, 282-285
Interesting case, 185-186
Interpretation, 35, 36, 54, 60, 133, 233,
241
Interruption of life by illness, 43
Interviewing, 8, 266
Intrusion, 38, 73
Invalidate, 11
Invisible act, 198, 199, 202, 210, 216,
222, 236
Invisible system, 202, 216
Involvement, 37, 78, 160, 174, 178, 200,
206, 218, 219, 226, 227, 228,
229, 231, 232, 235, 236, 237, 238,
239, 243, 244, 249, 250, 251, 252,
254, 255
Isolate, 41, 132, 134, 233; see abo
Avoidance, Contact
Jaco, E. Gartly, 126fn, 131fn
Japan, 49, 81
300
AWARENESS OF DYING
Mayo Clinic, 187

Mead, George H., 9, 13, 14, 283
Meaning, 12, 13, 18, 22, 37, 53, 54
Medical center, 187, 190, 289fn
Medical ward, 230, 249
Medication,37, 40, 41, 94, 101, 102,
103, 124, 127, 128, 162, 177, 188,
189, 190, 191, 194, 197, 198, 208,
209, 210, 212, 215, 222, 236, 244;
see also Sedation
Kastenbaum, Robert, 65fn
Mental hospital, 109, 139, 164, 183,
Katz, Daniel, 263fn
187, 287
Knowledge: doctor of patient, 135; Merton, Robert, 19fn, 219fn, 220fn,
patients, 48, 49, 55, 61, 101, 159
262fn, 281
Messinger, Sheldon, 215fn
Miller, Arthur, 279fn
Language, 146; technical, 149
Minimize, 94, 95, 105
Last chance operations, 48, 102; see Minimized
encounter, 5; see also
also Operation, Surgery
Avoidance
Lasting longer, 21
Miraculous, 170, 237
Leave taking, 184; weekend, 157
Misperceived pretense, 140fn
Lederer, Henry D., 126fn
Misread, 36, 251; see also Cues, Read
Legitimate, 19, 20, 191, 206, 208, 242,
ing signs, Signs
245
Misrepresentations, 36, 40; see also
Let him die, 190, 191, 193, 194, 196,
Disclosure
204, 214-225, 236, 244
Models, 12, 120
Lie, 36, 124
Modes: characteristic interaction, 10,
Likert, Rensis, 263fn
11, 37; of death, 79, 80, 81, 93, 102,
Lindesmith, Alfred R-, 284fn
106, 124, 176, 182
Line," 63
Mood, 25, 36, 78, 190
Lingering," 21, 24, 25,
100,169,173,Moote, Wilbert, 261fn, 267
177, 183, 187, 188, 189, 190, 191,
attitude, 3, 6
195, 197, 215, 239, 240, 243, 244, Moral
Morale, 130, 178, 234, 245
250,254-255, 287, 289; see also
Morbid, 217
Death
Moribund, 166, 248; see also Dying
Lippitt, Ronald, 263fn
patient
Locke, John, 14
Motivation, 178, 210, 225, 229, 233,
Long, Norton, 13
234, 256
Movement, restricted, 52
Mutual
pretense, 10, 11, 61, 63, 64-78,
Machinery, on, 102; see also Equip
90fn, 106, 125, 132, 139, 155, 157,
ment
158, 182, 240, 242, 264, 279, 286,
Maintain, 29, 32, 47, 49, 54, 59, 67,
287
71, 74, 75, 143, 149, 158, 188, 190,
Myrdal, Gunnar, 16, 17fn
210, 226, 227, 242
Mythology, 30, 49
Malaya, 81
Malpractice, 120, 180, 196
Manage, 4, 22, 41, 47, 56, 63, 73,
89-93, 99, 102, 103, 122, 139, 155, National Institutes of Health, 190
168, 175, 214, 224, 225, 255, 287; Natural causes, 26, 194, 197
Negligence, 26, 97, 98, 99, 100, 160,
engineer changes, 11, 42
166, 195, 209, 212, 218, 221, 222,
Mandate, 19
231, 232, 235, 236, 237, 238, 250,
Maneuver, 80, 162, 211; see also
252-253, 255, 256; non-medical, 96Counter-tactics, Manage, Strategies,
100
Tactics
Marginal doctors, 131, 138, 161, 186, Negotiation, 65, 94-96, 105, 189, 198,
211,
212, 213
187
Neurological ward, 230
Matza, David, 140fn, 262fn, 284fn
No change, 176
Mauksch, Hans O., 49fn
Jewish hospital, 201

Joseph, Florence, 158fn
Judgment, 82, 83, 84, 86, Ji9, 100, 196,
201, 211; ethical, 82, 216; medical,
82; moTal, 85; social, 82
Just in case, 62
Just like home, 72
Just as well, 107
Index
No-recovery problem, 177-203; see also

Cure, Recovery, Research
Non-cooperative, 41, 44, 57, 87 90
126, 134
Non-interacting, 126; see also With
drawal
Non-person, 26, 108, 164, 165, 171
Non-rescue tactics, 218-221
Non-verbal, 130
Normal, 74
Norms, 81, 105, 146, 161, 162fn, 164;
door norms, 164, 175, 232; see also
Rules
Nothing to live for, 85
Nothing more to do, 37, 102, 124
152, 166, 177-225, 231, 234, 235,
272, 287, 289
Nurses station, 46, 56, 141, 142, 154,
175, 233
Nursing care: adverse, 62; better,
62, 83; physical, 21, 23fn, 37, 41, 56,
58, 62, 68, 72, 88, 90, 94, 97, 108,
109, 144, 159, 161, 162, 165, 166,
178, 235-240, 264
Nursing home, 109
301
179, 184, 185, 190, 191, 192, 195,

197, 198, 201, 204, 205, 206, 215,
216, 217, 220, 222, 224, 225, 236,
237, 245, 250
Painless comfort, 208-214
Panic, 75, 101, 143, 160, 165, 175
Paradigm, 11, 12, 275, 283, 284, 285fn,
287
Parent participation program, 166
Parsons, Talcott, 9, 19fn. 126fn, 129fn,
262fn, 281, 283
"Pass off, 186
Pathology reports, 125
Pediatric ward, 33, 166, 174, 242, 249,
People, number involved, lOfn, 12;
see also Interactants
Periodic schedule, 213, 214
Personal techniques, 58
Personal values, 6
Perspective, 81, 84
Pessimistically, 54, 182fn, 255-256
Phone calls, 175, 176, 232
Physically dazed, 42, 49, 82
"Picking up the pieces, 149
Pinto, Oreste, 52fn
Plan, 43, 73, 103, 121, 122, 128, 229
Plays ball, 96; see also Cooperate
Object switching, 160, 166, 218, 231,
Plead, 102, 217, 223
232, 249
Polarity game, 133
Obligation, 82, 83, 191, 195, 199
Oblique disclosure, 20, 22, 42, 56, 125; Polio patients, 40
Polsky, Ned, 283fn
see also Disclosure
Post-mortem care, 4, 113, 114
Observations, 57, 85, 103, 126, 154
Power, 161; see also Authority, Control
Obstetrics ward, 156, 249
Practical use, 7, 8, 12, 259-273
Offensive, 47
Premature babies, 18, 25, 85, 107, 137,
Oken, Donald, 30, 128fn
144, 166, 220, 230, 231, 232, 234,
Olin, Lloyd, 262fn
289
Open awareness, 32, 43, 64, 67, 74-77,
Prepared for death, 42, 121, 138, 143,
79-106, 90fn, 91, 111, 132, 134, 137,
146, 149, 151, 153, 166; active, 121,
155, 158fn, 176, 180, 189, 191, 197122, 127, 128, 129, 135; family, 151,
200, 207, 212, 214, 216-218, 233,
153, 166, 169, 170, 171; passive, 121,
240, 242, 243
129-132; see also Plans; Review life
Open suspicion, 48, 50, 58, 60, 61, 63
Presence at death, 248-249
Operating room, 26, 137, 142, 154,
Present-future orientation, 37
230
Operation, 41, 48, 102, 112, 146, 166, Pretense, 90fn, 158, 241; initiating, 6770; interaction, 70-74; see also
189, 192; see also Surgery
Mutual pretense
Optimistic, 36, 50, 125, 182fn, 220
Pride, non-medical, 96-100
Organization of work, 63
Priorities, in care, 180-181
Oscillation, 61, 62
Privacy, 80, 93, 141, 153, 155, 161,
Out of character, 66
162, 167, 242
Out of our hands, 149
Private reaction, 38, 172; see also
Outsiders, 8, 277
Reactions
Over-generalizing strategy, 241
Privileges, 41, 94, 139, 155, 156, 157,
184-185, 207, 236; see also Rules
Pacing, 94, 208, 245
Problem, 3-16, 17-28, 41, 54, 56, 58,
Pain, 3, 5, 12, 37, 40, 74, 75, 84,
90, 92, 156, 160, 162, 175, 179, 180,
85, 96, 100, 102, 104, 124, 129, 159,
206, 287
302
AWARENESS OF DYING
Procedures, 35, 41, 48, 50fn, 54, 56, Reiss, Albert J., 283fn
Reissman, Leonard, 165fn
94, 95
Rejection, 93, 94, 98, 174
Process, 3, 11, 17, 227
Relative; role of patient, 164-167, 236;
Professional ethics, 6, 143, 148, 151,
role of worker, 164-167, 236; see
222
also Family, Visitor
Prognosis, 7, 18, 19, 20, 24, 25, 42,
Religion, 146; see also Chaplain,
138
Spiritual care
Progression of disease, 21, 26, 36, 147
Prolong life, 82, 84, 99, 104, 159, 166, Repetitive interaction, 15
Report, 4
179, 180, 182, 185, 186, 190, 191,
193, 194-203, 205, 206, 215, 216, Repress, 80
Reprieve, 170, 177; short term, 24, 127,
219, 220, 223, 225
138
Promise, 97, 98
Reprimand, 90, 91
Proper conduct, 82, 84, 85, 88, 90,
102, 103, 104, 105; see also Action, Reproach, 89
Research, 39, 43, 105, 120, 131, 145,
Appropriate, Behavior, Improper
170, 171, 188-193, 194, 198, 201,
conduct
213, 215, 236
Protection, 46, 63, 70
Resigned, 45
Psychiatrist, 31, 86, 91, 243
Resources, 51, 57, 60
Psychological aspects, 4, 5, 50, 58, 62,
Response, 122, 123, 125, 126, 130, 242;
75, 78, 83, 99, 120, 127, 166
see also Reactions
Public Health Service, 287
Pull away, 256, see also Avoidance Responsible, 82, 130, 141, 166, 197,
199, 204, 223, 248
Purposeful, 61
Rest, 156, 167
Retrogress, 22, 32, 40, 41, 75
Qualitative facts, 266, 292; diverse, 266 Revealing, 53, 56, 142, 145, 149, 157;
see also Disclosure
Quantitative facts, 265, 292, 293
Questions, 5, 7, 18, 32, 39, 51, 53, 58, Review life, 43, 104, 127, 157; see also
Prepared for death
100, 127, 132, 133, 144, 174, 175,
Reward, 178, 205
176, 232, 241, 242
Riesman, David, 282
Quint, Jeanne C., 4fn, 5fn, 57fn, 58fn,
Rights, 58, 59, 175, 192
129fn, 216fn
Rituals, 42, 59, 64-78, 85, 106
Rohrer, John H., 165fn
Role, 71, 86, 140, 159, 164-167
Race, 16, 17, 81, 84
Role switching, 143, 241, 243, 247;
Rapport, 146
see also Ask your physician"
Ratio; certain/uncertain death, 178179, 228; insiders/outsiders, 279; Rose, Arnold, 14fn
Rotation of patient, 136
patient-personnel, 46
Roth, Julius A., 19fn, 23fn, 33fn,
Ratner, Joseph, 273fn
132fn, 240fn
Reach, 126
Reactions, 38, 63, 81, 102, 103, 114, Rough patient, 63
Routine work, 45, 76, 94, 162, 164,
122, 123, 205
167, 174, 185, 207, 237, 238
Reading signs, 19, 54
Rules, 15, 32, 57, 67, 70, 72, 73, 78,
"Really knows, 47, 50
86, 114, 122, 135, 139, 155, 156,
Reassure, 36, 41, 42, 48
157, 158, 175, 199, 220, 221, 244,
Reciprocal, 82; isolation, 132, 134
285fn; invoked, 158, 161, 167;
Recognition, 10, 11, 48, 49, 50, 63,
relaxed, 155, 156, 157, 167; see also
79, 101, 102, 224, 225
Privileges
Recovery, 178, 179, 182, 186, 188, 190,
191, 194, 196, 197, 200, 201, 202, Runaround, 57, 62
204, 205, 206, 228, 229, 233, 234,
252, 253; see also Goals, No-re
Sabshin, Melvin, 14fn
covery problem
Sadness, 37, 174; see also Collective
Rehabilitation, 41
mood
Rehearsal, 80, 239
Sanction, 236
Reif, Fred, 280fn
Index
Save the patient, 176, 177, 178, 186,
188, 196, 201, 202, 215, 216, 228,
230, 231, 234, 244, 245, 252
Scenes, 73, 88, 91, 105, 139, 150, 151,
152, 158-161, 163, 186, 196, 212,
232, 236; see also Status forcing
Schatzman, Leonard, 14fn
Schedule, 94, 211, 212
Scold, 90, 91
Screaming, 88, 102, 208
Screening measures, 178
Secret, 29, 31, 35, 37, 52, 57, 79, 141,
145
Sedation, 21, 40, 75, 126, 159, 166, 182,
184, 204, 208, 210, 211, 212, 213,
233, 236, 238, 247; tactics of, 22,
40, 102, 213, 214, 221-223; see also
Medication
Selective listening, 240
Self, 16, 89
Send the patient home, 181-184
Senile, 42, 109
Senselessly, 85, 215, 219, 255
Senses, 52, 53
Sensitive, 172, 174
Sensitizing theory, 263, 265, 291
Sentient, 9, 42, 85, 94, 108, 115, 182
Sentimental, 226, 227, 228, 245, 249
Serene, 78
Settle affairs, 168, 169, 171, 236, 244;
social, 127; financial, 127, 128; see
also Plan, Prepared for death
Shadow, 51, 52
Sheff, Thomas J., 120
Sherif, M 59fn
Shift, changes, 40, 49, 228, 229
Shniedman, Edwin S., 129fn
Shock, 26, 44, 50, 152, 182, 197, 248,
251, 252, 254; see also Death, un
expected
Shopping around, 87
Signal, 42, 65, 69, 74, 76, 91, 134, 149,
243; see also Flash cues
Signs, 12, 13, 30, 48, 51, 52, 55, 277,
278, 279
Si'ence, 39, 47, 53, 54, 59, 62, 64, 75,
78, 91, 102, 125
Simmel, Georg, 13, 282
Situational discounting, 111-113; see
also Discounting
Skin, color, 13, 16, 17, 277, 278
Slips, 54
Snow a patient under, 40, 102, 213,
214
Sociability shield, 36
Social change, 14
Social consequences, 3, 5
Social definition, 16-28
303
Social interaction, 7, 8, 9, 10, 14, 15,
16, 49, 58, 59, 60, 63, 64, 65,
73, 75, 82, 90, 94, 100, 108, 121,
143, 145, 156, 158, 159, 160, 167,
179, 180, 183, 206, 207, 209, 269,
274-285
Social issues, 6
Social loss, 38, 130, 219, 228, 229,
232, 237, 238, 244, 247, 250; actual,
38; potential, 38
Social order, 13, 14
Social-psychological dimensions, 12
Social regulations, 14, 15
Social structure, 11, 13, 271, 283, 284,
285, 289
Social workers, 21, 121, 127, 128, 130,
133, 134, 135, 143, 147, 153, 171,
172, 183, 214, 235, 243
Socially alive, 109, 110
Socially dead, 109, 110, 215, 237, 255;
see also Comatose, Premature babies
Socioeconomic status, 81; high, 57, 123;
low, 57, 64, 84, 123, 171
Sociological research, 282-285
Space, 37, 56, 111, 153, 158; arrange
ment of, 57, 154; management of,
162-164
Spencer, 282
Spiritual care 99, 128; see also Chaplain
Spotter, 52
Staff rotation, 40
Stakes, 12, 111, 190, 277, 279
Status, 9, 13, 16, 23, 24, 30, 32
47, 48, 53, 54, 68, 82, 122, 145,
148, 154, 207, 217, 218, 277, 278;
professional, 57, 58, 59
Status forcing, 158-161, 163, 173, 212,
217, 235
Status passage, 24
Stewart, J. M., 165fn
Stinchcombe, Arthur, 161fn
Strain, 45, 59, 62, 88, 99, 105, 149,
151, 164, 172, 173, 184, 239, 243fn,
246, 247, 248
Strategies, 226, 227, 230, 232, 233, 235,
236, 237, 238, 240, 241, 242, 243,
246, 250, 251, 255, 256, 269;
defense, 227, 228
Strauss, Anselm L., 4fn, 5fn, lOfn,
14fn, 19fn, 24fn, 29fn, 33fn, 38fn,
59fn, 130fn, 158fn, 178fn, 216fn,
219fn, 224fn, 228fn, 276fn, 277fn,
280fn, 283fn, 285fn, 286, 286fn
Stress, 9, 25, 105, 130, 138, 164, 182,
185; 199, 225, 242, 243fn
Structure, 63, 85, 161fn, 181; condi
tions, 29-32, 46, 48, 49, 67, 79, 220,
223, 226, 279, 284, 285fn, 289
304
Substantive area, 259-273, 275, 276,
281
Substantive theory, 259-273, 275, 276,
280, 282, 283, 284, 286, 288, 289,
290, 293
Suicide, 83, 84, 85, 93, 101, 102, 110,
120, 121, 129, 132, 186, 193, 210,
212, 223-225, 251, 252, 254
Support, 43, 60, 120, 143, 149, 164,
172, 197, 231, 234, 244, 245
Surgery, 97, 104, 177, 233; patients,
54, 125; see also Operation
Surprise, 26, 44, 150, 151, 152, 168,
170, 252, 253, 254; see also Death,
unexpected
Suspicion awareness, 11, 32, 33, 34, 35,
36, 38, 39, 40, 42, 45, 47-63, 64,
67, 90fn, 100, 101, 111, 139, 151,
154, 155, 157, 158, 168, 169, 180,
183, 240, 243
Suspicions, 47, 79, 108, 181, 182, 186,
229, 246; negate, 47; unfounded, 60;
verify, 47, 50, 54, 169, 233
Sympathy, 52, 71, 72, 75, 83, 85, 92,
95, 135, 149, 171, 174, 224, 245
Symptoms, 40, 41, 42, 48, 49, 50, 53,
73, 194, 229, 233; see also Cues,
Signs
System, 51, 53
Taboo, 3, 184
Tactics, 8, 11, 13, 15, 26, 36, 41,
51, 54, 63, 77, 91, 94, 102, 105,
123, 125, 140, 142, 145, 147, 149,
153, 163, 166, 167, 209, 210, 211,
253, 269, 279, 283; avoidance, 154;
confirmation seeking, 56; patient,
212, 213; staff, 55-58, 90, 137, 143,
144, 155, 159, 160, 161, 174, 175,
225, 238; see also Strategies
Take, 61, 105, 119, 141, 154fn, 183
Talk, 3, 4, 5, 11, 37, 38, 41,
42, 44, 50, 57, 68, 72, 73, 78,
84, 91, 95, 98, 103, 104, 107, 130,
148, 214, 233-234, 238, 240-244;
of dying, 41, 45, 58, 59, 67, 69, 105;
see also Communication, Conver
sation, Disclosure
Task, 148, 160fn, 161, 162, 165
Team, 12, 34, 51, 52, 53, 57, 78,
161fn, 165, 173, 246
Technical care, 4, 5, 6, 97
Teenagers, 32, 52, 92
Temporary illness, 43
Terminality, 4, 5, 7, 8, 12, 17, 20,
26, 32,
33, 38, 40, 41, 43, 45,
48, 49,
50, 53, 54, 55, 58, 61,
AWARENESS OF DYING
67, 73, 75, 77, 78, 80, 94, 100,

102, 109, 111, 119-135, 153, 179,
182, 229
Testing, 50fn, 53, 56, 58, 62, 132,
145, 169, 233
Tests, 35, 38, 41, 112
Theoretical aspects, 7, 11, 12, 13, 259273
Thomas, W. I., 9, 283
Threat, 90, 91, 159, 212
Time, 16, 40, 47, 106, 124, 132, 194195, 241; expected period until death,
16, 24, 43, 44, 80, 147, 149, 150,
192; expected time of death, 18, 79,
100, 102, 147, 167, 168, 169, 170,
246-250; see also Death expectations
Too far gone, 42, 48
Training: doctors, 4, 19; informal, 63;
nurses, 4, 63, 99, 256
Transgression of rules, 244
Transitions of awareness, 11, 32, 39-42,
45, 60, 61, 62, 74-77, 123, 139, 145,
147-154, 155, 157, 158, 170, 180,
182, 185, 189, 196, 197, 204-206,
210, 217, 265; see also Change
Traumatic, 5, 26, 146
Treatment, 4, 35, 40, 41, 44, 73, 94,
112, 120, 143, 146, 154, 160, 166,
187, 188, 193, 194, 204, 217, 253;
superfluous, 160, 161
Trust, 33, 34, 50fn, 120, 140, 229, 243,
251
Truth, 42, 55, 57, 60, 62, 125, 180
Turn for the worse, 107, 108, 206,
218, 248
Turning point, 224, 225
Unaware, 44, 136-154, 185, 186, 189,
191, 220, 233; see also Closed aware
ness
Understanding, 13, 80, 85, 93, 130;
mutual, 13
Understanding of theory, 259, 262-264
Upsetting, 5, 101, 130, 226, 227, 233,
238, 239, 249, 251, 252, 253
Vacillating pattern, 24, 25, 158, 234,
252
Validating, 53, 54, 60, 84, 139, 200,
205
Values, 14, 38, 41, 180, 181, 182
Vegetable," 80, 101, 102, 174, 215
Very much alive, 218, 219
Veterans Administration Hospital, 39,
122, 123, 139, 164, 187, 289
VIPs, 61
Index
Visibility, 219
Visiting, 155-167, 173
Visiting attendant, 166; see also
Relative
Visiting nurses, 92
Visitor, troublesome, 156; see also
Family
Waiting room, 159, 163, 164, 182, 232,
233, 235
Ward clerk, 171, 235
We all have to go sometime, 36, 233
Weber, Max, 9, 283
Weissberg, Alexander, 66fn
303
Wertenbaker, Lael, 87fn, 93fn, 104fn,
113fn, 129fn, 162fn
Wilson, M-, 59fn
Wish to die, 80
Withdrawal, 89, 92, 126, 186, 192, 208,
240, 245
Work schedules, 21, 167
Work with," 78
Young patient, 155, 219, 228, 287, 247
Zetterberg, Hans L., 261fn, 262fn,
263fn, 265fn, 266fn, 267fn, 268fn

Barney G. Glaser and Anselm L. Strauss-Awareness of Dying-Aldine (1965) PDF

Uploaded by

Copyright:

Available Formats

Barney G. Glaser and Anselm L. Strauss-Awareness of Dying-Aldine (1965) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Barney G. Glaser and Anselm L. Strauss-Awareness of Dying-Aldine (1965) PDF

Uploaded by

Copyright:

Available Formats

IStCVAFICNS

A series edited by Howard S. Becker

Recently The N ew Y ork Tim es reported: v e r y i l l c h il d r e n

listens to conversations himself. Occasionally he queries the staff

we were told, remarked flatly that sociologists have nothing

Like all field researchers, we are especially indebted to many

1. The Problem of Awareness of Dying

PART I I : TYPES OF AWARENESS CONTEXTS

PART i n : PROBLEMS OF AWARENESS

Chapter S. Direct Disclosure of Terminality

The Practical Use of Awareness Theory

hospitals. Although trained to give specialized medical o r nurs

The Problem of Awareness of Dying

each other, or to their superiors, what they have talked about

somehow discover the truth without being told explicitly. Some

The Problem of Awareness of Dying

trend: because medical technology has vastly improved, fewer

As noted, our book is based upon extensive field obser

The Problem of Awareness of Dying

il, most useful sociological accounts are precisely those which

Our own approach to social interaction and awareness has

among themselves. W hat each interacting person knows of the

The Problem o f Awareness of Dying

chapters develop models for understanding and explaining

The Problem of Awareness of Dying

meaning is not always apparent, either, to all other interactants.

tist, recently wrote that there is a sociology of John Locke

The Problem of Awareness of Dying

we are cognizant of rules but are principally interested in

Expectations of Death: A Problem in Social Definition

the excursion. . . . [They] entertained me for more than

DIA GNOSIS AND PROGNOSIS AS

n o less im portant when the patient is fated soon to die, unless

Expectations of Death: A Problem in Social Definition

(2 ) certain death at an unknown time, (3 ) uncertain death

flashed, perhaps unwittingly or obliquely, by the doctor. Some

Expectations of Death: A Problem in Social Definition

Expectations of Death: A Problem in Social Definition

are less dependable but operate in conjunction with the other

The changing expectations of nurses and physicians actually

Expectations of Death: A Problem in Social Definition

alternation may occur sufficiently often to cause stress among

surprising changes are the sudden death, or onset of death,

experienced at recognizing the signs of impending death. Of

sense telling them directly. Presumably some, patients do not

is that ordinarily, the patient has no allies who reveal o r help

ting an accurate assessment of his condition from the medical

PART TWO: TYPES OF AWARENESS CONTEXTS

rather more complicated. Trust does not arise automatically;

that arouse his suspicions about his condition. Inevitably he

may also go out of their way continually to reassure him that

identities, conveying to the patient that he is not expected to

PART TWO: TYPES OF AWARENESS CONTEXTS

tight control over their reactions. Although the staffs control

compassionate distress when suddenly confronted by obviously

of rehabilitation programs because they do not anticipate being

perm anently); occasionally this can cue him into suspicion

touching on a contrasting situation. In her book Experim ent

t w o : ty pes o f aw areness co n tex ts