mess

W Illl’llllllllillfillilmllllil

3 1293 01559 2730

 

LIBRARY

Michigan State
Universlty

 

 

 

This is to certify that the
thesis entitled
OH Ttl-E C0“ C6?!" OF ‘FDWEQ 'IN
'DoQTDQ. .. TAT3€NT lNTELACfi 0N

presented by

KUET‘L' kARLSEN

has been accepted towards fulfillment
of the requirements for

MASTERS degree in Mars «1 Levels

 

 

 

 

Date ‘4 my {$56

0-7639 MS U is an Waive Action/Equal Opportunity Institution

 

 

 

 

 

PLACE N RETURN BOX to roman this chockoul from your record.
TO AVOID FINES return on or before date duo.

DATE DUE DATE DUE DATE DUE

  

 

 
    

 

 

-
___l:_l__j

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

THE CONCEPT OF POWER IN DOCTOR-PATIENT INTERACTION
By

Kjetfl Karlsen

A THESIS

‘ Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of ‘

MASTER OF ART
Interdisciplinary Programs in
Health and Humanities

Department of Arts and Letters

1996

ABSTRACT
THE CONCEPT OF POWER IN DOCTOR-PATIENT INTERACTION
By

Kjetil Karlsen

This thesis is a commentary on Howard Brody's book The Healer‘s
Power. I am not disagreeing with the substance of Brody's analysis or with his
advice for the ethical uses of power in doctor-patient relationships. But. I do
suggest that Brody‘s pragmatic use of the term power prevents us from gaining
certain important insights in the role of power in clinical encounters. I present
a more rigorous definition of power, based on Anthony Giddens' structuration
theory: Power arises from structural resources, produced in interaction
between people. Power is not something you have, rather it is something you
activate.

Power thus resides in relationships, and not within any person.
According to this definition, the healer is not powerful. Healing is the fortunate
outcome of instances where both the ‘healer' and the ‘healed' invest some of
their resources in a relationship, thereby activating the healing powers in a
doctor-patient relationship.

ACKNOWLEDGMENTS

This paper was written at Michigan State University (MSU), under the
supervision of Howard Brody, M.D., Ph.D. (Family Medicine and Medical
Ethics); Martin Benjamin, Ph.D. (Philosophy); and Laurie Medina. Ph.D.
(Anthropology).

Kjetil Karlsen is currently enrolled in a Ph.D. program at the Center for
Medical Ethics, University of Oslo, Norway. A scholarship from the Norwegian
Research Council covered the expenses for the stay at MSU.

TABLE OF CONTENTS

1: INTRODUCTION .................................... l
2: POWER IN SOCIAL SCIENCES ........................... 4
Agency ........................................ 7
Structure ....................................... l 1
Practice approaches ............................... 14
Diiferent ways of combining structure and agency ........... 22
3: POWER IN THE CLINICAL SETTING ....................... 27
To be powerful ................................... 28
To use power .................................... 32
4: POWER AND ETHICS ................................. 43
Views of power in schools of moral philosophy .............. 45
A deconstructivist critique of power ..................... 48
5: POWER AND ETHICS IN THE CLINICAL SETTING .............. 51
Use of power and moral accountability ................... 52
Practice approach view of power: a contradiction resolved ...... 59
REFERENCES ........................................ 67

iv

1: INTRODUCTION

There are two reasons why we should think about power in the doctor-
patient interaction. First, a practical one. Howard Brody claims that medical
ethics is most profoundly about power and its responsible uses (Brody 1992,
p.12). Several publicly discussed themes support this claim: The physician's
authority; patients' feeling of powerlessness in interaction with health
professionals; the potential of coercion and manipulation; but also, the potential
for positive forms of influence (which Brody calls “the healer‘s power”); all
indicating that power does indeed play a central role in the encounter between
physician and patient. In spite of the perceived importance of power, there is a
disturbing lack of consensus as to what this word should mean. This paper is
an attempt to bring some clarity to that question. A physician who employs a
coherent and adequate vocabulary to evaluate the role of power in her
interaction with patients, we may hope, will be more sensitive to what types of
power relationships she engages in. With this sensitivity, she would be more
likely to engage in power relationships which are constructive rather than
detrimental for her patients.

Second, there's a theoretical reason why we should think about power.
If the doctor-patient encounter often yields morally significant outcomes, and
what goes on in this encounter can (partially) be described in terms of power,
then an understanding of that power relationship will provide relevant
information for a moral assessment. The social sciences present a large body of
empirical research on the healer-patient interaction. It is not always easy to see
how insights from this literature could be reformulated to fit the language of

medical ethics. The original contribution of this paper is the way it forms a

2

‘link' between moral philosophy and social sciences. The device I use to create
that ‘link’ is the concept of power.

A rumor is spreading to the field of medical ethics these days that
postmodern critique challenges the very foundation of its reasoning. Central to
postmodern thinking is a critique of power, it is said. Nobody seems to be able
to explain precisely what implications the postmodern insights should have for
practical medical ethics. While the ‘pre-postmodern' adherents to medical
ethics are waiting for someone to formulate this critique in an intelligible way,
and to point out what its practical implications may be. we would do well to
make clear to ourselves what our own "modern" view of power is. That will be
my major task in this paper.

We could say that medicine is a powerful knowledge, that a surgeon has
the power to save peoples' lives, that he uses powerful tools and techniques,
and that he partly relies on the self-healing powers within the patient’s body.
By using the word ‘power' in this loose sense, we take the statement ‘A is
powerful' to mean simply ‘A has the capacity to cause effects, or to play a part
in the process of causing efi‘ects.’ In this paper I will present a more rigorous
definition, based on Anthony Giddens' structuration theory: Power arises from
structural resources, produced in interaction between people. Power is not
something you have, rather it is something you activate. With this definition.
we are able to use power as a conceptual tool to analyze interactional aspects
of the doctor-patient relationship with a high degree of precision. According to
this definition, the healer is not powerful. Healing is the fortunate outcome of
instances where both the ‘healer' and the ‘healed' invest some of their
resources in a relationship. The power of healing resides in this relationship.

and not within any person.

3

Based on a reading of central social science texts dealing with the
question of power, I will identify and discuss three distinct views of power:
models based on agency; structure; and a practice approaches (chapter two). In
spite of their differences, these models are all based on the assumption that
there is something definite to be said about where power is located and how it
can be exercised. In opposition to these views, is an approach I call theoretical
pragmatism. which holds that we should not base our empirical theories on any
fixed ideas about power.

In chapter three, I will asses three clinical cases, using a language
compatible with Giddens' version of the practices approach. In the following
chapter, I will briefly suggest how opposing views of power are reflected in
different theories of moral philosophy. In the last chapter, I assess Howard
Brody’s use of a pragmatic definition of power in a discussion of medical
ethics. I will claim that there are problems with Brody's analysis which could
be solved by applying the practice approaches view of power as a theoretical
starting point for our ethical analysis of the doctor-patient interaction.

In chapters two and three my approach is mainly descriptive, and I will
draw upon literature from the social sciences. In chapters four and five I
introduce normative questions. and there will be some references to the

literature of moral philosophy.

2: POWER IN SOCIAL SCIENCES

The question of where power is located is most profoundly a question of
where to look for the cause of the types of effects we are interested in. A
number of different reductionist models are available: the ultimate cause for
human behavior is found in the genetic material; in physiological processes in
the human body; in psychologically reinforced patterns of behavior; in the
struggle for survival; etc. Most of these reductionist models do not provide a
useful answer to the question we started out with: the question of power in
doctor-patient interaction. Let us take genetic reductionism as an illustration: If
it were true that our genetic material predetermined our behavior, we would
have to conclude that two persons meeting are like two fanciftu constructed
automata, whose interaction consists of a set of precoded types of adaptive
activities. Such claims have no practical implication for our understanding and
evaluation of the interactional aspects of the doctor-patient encounter. Most of
the models that theoretically could ofier an answer as to why we act the way
we do, do not need to be considered then, since the types of replies they
would yield would not serve our purpose.

Our scope is then limited to models of the same ‘scale’ as our subject
matter. Our focus is on the interaction between individuals in a certain social
situation. not in physiological processes or Skinnerian shaping of behavior.
Broadly speaking then, we will turn to theories that focus on individual
behavior within a social setting, and models that focus on the social setting in
which individuals interact.

Howard Waitzkin and Arthur Kleinman are two of the central authors in

the social science literature on doctor-patient interaction. Waitzkin draws on

5

structural explanations in his work; Kleinman focuses on agency. Gregory
Pappas claims that a disagreement about the nature of power is “at the crux of
the structure/ agency division in medical anthropoloy” (Pappas, p. 199). The
structural approach locates power outside individual actors. and sees it as a
constraining and determining condition. The structuralist sees human action
and experience as predetermined in a mechanical way by structural
regularities, existing independently of individual actors. Obviously, there is no
room for individual liberty and freedom of choice (agency) in such a scheme.
The agency approach, on the contrary, views individual actors as
knowledgeable entities, pursuing their goal-oriented strategies of action:
“Agency concerns events of which an individual is the perpetrator, in the sense
that the individual could. at any phase in a given sequence of conduct, have
acted differently” (Giddens, p.9). This approach thus locates power within each
individual, seeing power as the capacity or resource that gives agency its
momentum, the individual actor's capacity of “bringing about of consequences“
as Mark Philip formulates it (Philip, p.635).

Pappas argues that Waitzkin puts too much emphasis on structure in his
analysis. whereas Kleinman focuses too narrowly on agency. He suggests that
practice approaches offer a theoretical framework for a more adequate
understanding of the doctor-patient encounters. After reviewing the literature
in medical sociology. Per Méseide makes similar remarks: the structural
models miss “the local production and management of power relations,“
whereas models that focus on interactional aspects neglect “the structural
necessities of power in medical practice“ (Méseide, p.546-7). In brief, the
practice approaches hold that we cannot gve primacy to either structure or

agency, because neither one of them can be understood independently of the

6

other. Structure and agency mutually influence each other -- the mechanism
that ties them together is the practices in which individuals engage.

The points raised by Pappas and Méseide reflect a general shift of
theory within the social sciences in the 80's. After presenting an overview of
anthropological theory since the 50's, Sherry B. Ortner claims that
anthropological theory in the 70's found itself in a state of liminality, a period
marked by profound theoretical disagreements, out of which a “new and
perhaps better order“ could arise (Ortner, p. 127). She argues that the practice
approaches, formulated by Pierre Bourdieu and Anthony Giddens, represent a
synthesis of the better aspects of previous competing theoretical models.1

I will review some of the central works in general social science theory
that address questions of power. An article by Bertrand Russell and critics of
his view illustrate the agency approach to power; one version of structural
thinking (the Marxist version) is represented by Nicos Poulantzas; Anthony
Giddens' structuration theory illustrates the practices approach. My discussion
will be guided by the following three questions: 1) What types of interaction are
seen as significant for a power analysis; 2) What characterizes instances of

power; 3) How are power relationships maintained. challenged and altered.

 

1 This change in social scientific theory historically coincides with changes in prevailing
theoretical models of the doctor patient relationship. Méseide argues. A normative model
based on justified professional domination. the expert model. was challenged and gave
rise to its antithesis: a normative model of cooperation and equality. the social
psychological model. After identifying problems with each of these. Maseide claims that a
new model is now emerging. one that “instead of [prescribing] ideal forms.... tries to
understand the form and production of competent medical work“ -— the control model
(MAseide. p.547).

Macy
Russell defines power as “the production of intended efi‘ects“ (Russell,

p. 19). Instances of power take place when two or more actors have similar
desires, and one of them achieves what he desires to a larger extent than the
other. If A desires something difierent than B; or if both A and B have multiple
desires. of which A achieves some and B achieves others. there is no way of
estimating who has the more power, according to Russell. The locus of power
is an intentional actor: “Alexander the Great and Julius Caesar altered the
whole course of history by their battles“ (Russell, p.21). The structural
conditions merely provide a scene on which individual actors intentionally seek

to get what they desire:

Different types of organizations bring different types of individuals to the

top. and so do different states of society. An age appears in history

through its prominent individuals, and derives its apparent character

from the character of these men. (Russell, p.23)
When Russell says that an organization ‘brings' a person to the top. he does not
ascribe agency or power to that organization, the role of which remains
passive. Society provides a backdrop that makes it possible for a person with
those capacities possessed by Richard Lionheart to achieve prominence in his
time, whereas another historical period would have provided him with fewer
opportunities to exercise power others.

Still, institutions may be possessors of power in Russell’s model: a
political party. for example, can be seen as a ‘machine' which actively creates
effects: “Sometimes.... the machine is able to secure the victory of a man

without ‘magnetism’; in such cases, it dominates him after his selection. and he

never achieves real power.“ The successful politician. on the contrary, is able

8

to “dominate [the machine) and make it subservient to his will” (Russell, p.26).

The mechanisms by which individuals and institutions exercise power
over (influence) persons are classified in three groups: use of physical force;
use of rewards and punishments; influence of opinion, i.e. propaganda
(Russell, p.19). Central to Russell's account, is the view that power is exercised
in situations of conflict: two or more actors desire similar things. their intended
actions are aimed at getting that which they desire; the most powerful agent
gets it his way: “If you wish to be Prime Minister. you must acquire power in
your Party, and your Party must acquire power in the nation” (Russell. p.23).
Even though it is possible to define power this way, Russell's concept conflicts
with the meaning commonly ascribed to power in several important ways as
demonstrated by other authors:

By defining power as “the possibility of imposing one's own will upon the
behavior of other persons” (my emphasis), Weber reminds us that it is not the
actual exercise of influence over others we normally refer to when speaking of
powerful agents, but rather their capacity of doing so if they want to (Weber.
p.29). Weber. like Russell, maintains that an overt conflict of interests is
central to instances of power:

Power (Macht) is the probability that one actor within a social

relationship will be in a position to carry out his will despite resistance,

regardless of the basis on which this probability rests. (Weber 1922.

quoted by Dahl, p.406)

Steven Lukes expands the concept of power further, arguing that if we
only include situations with resistance (overt conflict), we will exclude those
situations where the more powerful agent uses his capacities to silence and

suppress the expressions of resistance, thereby avoiding an overt conflict:

Indeed. is it not the supreme exercise of power to get another or others
to have the desires you want them to hav -- that is, to secure their
compliance by controlling their thoughts and desires?

(Lukes 1974. p.23)

This point may be in accordance with our intuitive understanding of power, but
in the subsequent discussion Lukes runs into trouble when he defines power
as instances where one agent influences another contrary to that agent’s real
interests. Fo -- how are we to determine what would amount to a person's real
interest? Lukes does not provide us with an answer.

Pappas avoids this problem by excluding the idea of real interests from
his definition of power. Not only situations where somebody is influenced
contrary to their interest should count as instances of power. he argues. for
power can also be used to influence others in a way that is compatible with
their best interest. By doing so , he suggests that power should be understood
as a value-neutral term: “Power is involved in all that is repressive and
destructive, as well as all that is liberating and creative in human behavior”
(Pappas, p.200).

Several authors within the social sciences have used power as a morally
negative term, more often implicitly than explicitly. In such research, the
scientist has seen his or her role as locating where and how power is
exercised, and then assumed without additional argument that we ought to
neutralize or minimize such use of power in the future. There is a large body of
research on the doctor-patient relationship based on the assumption that an
equal power balance is a goal for the doctor-patient encounter (Meetuwesen et
al.; Street). David Nyberg points to a fundamental theoretical problem with this

negative view of power: Whether or not the effect of an action is compatible

10

with or contrary to a person's interests is not only difficult to determine, but
often it will be impossible to determine this at the time the relevant action
takes place. With the negative view of power, we would have to wait until we
know whether the efiects of an action are ‘good‘ or ‘bad' before determining if
that action was an instance of power or not. Information about the setting and
the actors that engage in interaction, no matter how precise and reliable that
information is, is not in itself sufficient for a power analysis. Based on the way

we commonly use the word power. this position has something awkward to it:

Stalin, Hitler, and Joe McCarthy are clear enough examples that power
corrupts, but what of other powerful people such as Lincoln. Gandhi.
Franklin Roosevelt, and Martin Luther King whose power did not
apparently corrupt them?... It will not do simply to say that if what we
took to be power did not actually corrupt, then it must have been
something else posing as power... (Nyberg, p.38)

The discussion so far has illustrated some definitions of power and the
ways in which they differ.The purpose of this paper does not require that we
pursue these disputes, as all they would yield are different modifications within
the agency views of power. Several additional disputes have not even been
addressed here: whether power is a ‘zero-sum' concept; what kind of effects
are relevant for a power analysis; questions of intentionality; and of causation.
Philip states that “there seem to be few, if any, convincing metatheoretical
grounds for resolving disputes between competing theoretical paradigms“
(Philip, p.636). Lukes concludes that “every attempt at a single general answer
to the question [of defining power] has failed and seems likely to fail,“ and Dahl
explains why this might be so: “[A] variety of simple alternative explanations
seem to fit the data equally well“ (Lukes 1986. p. 17; Dahl, p.41 1). In spite of

their differences, the views discussed so far have one trait in common: they

11

see power as a resource or capacity that characterizes individual agents,
whether they are defined in terms of persons, institutions or societies. Power
thus is something one agent exercises over another agent. Contrary to this view

are authors who locate power outside of individual agents.

Structure

According to Poulantzas, power should not be applied to inter-individual
relationships independent of the wider system within which they are set. What
happens in the interaction between single actors, unrestricted by structural
constraints, should rather be described in terms of might (in the original text.
this distinction is between pouvoir and puissance). This is more than a play of
words, for Poulantzas argues that in our social reality (the capitalist society)
what happens in the inter-personal relationship is typically determined by
structural power, not by might. This means that the true explanation of why
people interact as they do is found in conditions located outside them: “In this
sense. we can say that power is a typical phenomenon, traceable from the
structures, while that of might is a phenomenon characterized by a sociological
amorphy" (Poulantzas, p.146).

Whereas Russell had allowed power sometimes to be located within
individual actors, sometimes within institutions (like a political party),
Poulantzas sees a supra-individual entity (social class) as the only locus of
power: “By power, we shall designate the capacity of a social class to realize its
specific objective interests“ (Poulantzas, p. 144). It does not suffice to say that
the difference between Russell and Poulantzas is merely that the first locates
power within individuals or supra-individual entities; the latter only in supra-

individual entities. The difference between agency and structure arises from

12

more profound differences in theory. When Russell refers to organizations. he
describes them as goal-oriented actors that possess power and interests,
arising from the power and interests of th08e individuals of which the
organization consists (Russell, p.27). Poulantzas sees social class as an entity
that arises from the economic reality (the distribution of the means of
production). The interest of a social class is defined by the theory, and cannot
be described as a function of individual characteristics of the members of that
class: “[The] concept of interests can and must be stripped of all psychological
connotations” (Poulantzas. p. 151). In a similar fashion. structure is not a
product of the behavior of single actors, but rather the condition within which
the behavior of individuals must be understood. Again, structure is defined by
theory, and in a way that makes it a conditioning factor for human activity,
rather than a factor conditioned by human activity: “Marx goes as far as to say
that class interests, in the class struggle, have an existence somehow prior to
the formation itself, to the practice of a class” (Poulantzas, p.149).

The types of interaction that can be described in terms of power.
according to this view, are only situations of conflict and struggle between
social classes: “The capacity of a class to realize its objective interests, and so
its class power, depends on the capacity. and so on the power, of its
opponent“ (Poulantzas, p. 15 l ). Individuals have no power. what they
psychologically perceive as their ‘interests' is merely a function of the true
interests of the social class to which they belong.

Rather than pursuing the theoretical aspects of the structural view of
power. I will illustrate some of its features by referring to an example of how
Waitzkin employs his structuralist view in a critical analysis of the doctor-

patient encounter. He summarizes the encounter between a 55 year old male

l3

worker currently on sick leave, the patient's wife, and their doctor like this: “A
man comes to his doctor several months after a heart attack. He is depressed.
His period of disability payments will expire soon. and his union is about to go
on strike. His doctor tells him that he is physically able to return to work as a
radial drill operator and that working will be good for his mental health. The
doctor also prescribes an antidepressant and a tranquilizer“(Waitzkin, p. 76-7).

The following transcript is from this conversation:

Yeah, so if they arrange something. they'll know if by mid June.
They should.

Is that bugging you? The idea of going back to work?
Well...actually I think I want to go back.

Yeah, I think you should go back.

Actually, I think I want to go back, but then go back and go on
strike? That seems to bother me.

Yeah. But if you go back mid-June it won't, won’t bother you.
No ...........

79F? 79979979?

In Waitzkin’s analysis, this case illustrates how a well-meaning doctor
unintentionally becomes an instrument of the medicine's social control over
workers. In the doctor's utterances. Waitzkin sees a strong ideologic message -
- namely that to work is good for the patients health. The offering of
psychotropic medications to the patient represents a technical solution to a
contextual problem, and serves the purpose of marginalizing the social context
in which the patient developed a depression. The function this physician is
really serving. according to Waitzkin, is to “control a working-class patient's
role in economic production by withholding the continued certification of

illness” (Waitzkin, p.78).

14

The Marxist model illustrated by Poulantzas and Waitzkin is only one
among several structuralist models of human behavior. What all these models
have in common, is the view that the capacity of making decisions is
determined by conditions outside the control of single individuals. According to
structuralist thinking, human behavior and cognition is organized by a reality
outside themselves, consisting in a number of categories/classes. These
categories are related to each other in a systematic fashion -- this system is
what is described as ‘structure,’ In Lévi-Strauss’ account, the categories are
classified in binary oppositions (high - low; land - water; mountain hunting -
sea hunting; peak - valley; raw - cooked, etc.), the most fundamental
opposition being that between nature and culture (Lévi-Strauss, p.471). Ortner
advocated a structural model based on the opposition female - male (Ortner
1974). Another type of structuralism is the form of linguistics which views
language as a organizing (both enabling and restricting) condition for all human

reflection.

Practice approaches

There are two forms of reductionism we have to avoid if we want to
understand social behavior, Giddens argues. First, a type of reductionism
which suggest that we can fully explain the causes of human behavior by
reference to characteristics of single individuals, reflected in their intentions
and unconscious motivations. Second, the type of reductionism which sees
intention and motivation as mechanically governed by a structural reality
outside the individual actor. In his structuration theory, Giddens regards
agency and structure as levels “of equivalent interest and importance, aspects

of a duality rather than a mutually exclusive dualism“ (Giddens, p.30). He

15

recognizes that the task of formulating a theory which combines structure and
agency demands “a very considerable conceptual effort” (Giddens, p.xxi). This
partly stems from the complexity of the subject matter, but also from the fact
that in combining these two perspectives, he has to use words that are
embedded with assumptions Giddens wants to disagee with. For this reason,
he introduces a variety of neologisms, and he also defines already known terms 1
in a way that differs from their traditional meaning. I will give a brief account of E
the meaning he ascribes to the terms most useful for the discussion in this

paper.
Structure: “Rules and resources recursively implicated in reproduction of

 

social systems. Structure exists only as memory traces, the organic basis of
human knowledgeability, and as instantiated in action“ (Giddens, p.377).
Structure thus does not have an existence in and of itself, it’s only existence is
through the formative impact it has on social behavior (practices), such
behavior being “reproduced chronically across time and space“ (Giddens,
p.xxi). Partly structure function as rules, conceived of as (a) normative elements
which restrict the realm of permitted behavior; and (b) codes which ascribe
certain significance to certain types of behavior. Partly structure functions as a
resource, by which Giddens means (a) authority, arising form the organized and
coordinated features of those social activities where some exercise a legitimized
control over the other, and (b) allocative resources, stemming from control over
“material products or aspects of the material world“ (Giddens, p.xxxi).

At the core of Giddens theory is the concept of duality of structure.
Giddens sees structure both as the medium through which social interaction is

organized, and at the same time, as the outcome of such social behavior: “The

16

structural properties of social systems do not exist outside of action but are
chronically implicated in its production and reproduction“ (Giddens, p.374).
[Practices] are not brought into being by social actors, but continually
recreated by them via the very means whereby they express themselves
as actors. In and through their activities agents reproduce the
conditions that make these activities possible. (Giddens, p.2)
Stratification model: This is Giddens’ account of human agents,
described in terms of the grounds/ reasons for their behavior. Giddens points
out that structuralist models have treated individuals as if all human behavior
arose as a mechanical response to external stimuli, in the same fashion as
involuntary neurological reflex loops. This is not a probable account, Giddens
says, as our experience to the contrary tells us that humans in general are able
to produce some kind of verbal account of why they do what they do --
demonstrating a discursive knowledgeability about the reasons for the actions.
Structural models fafl to “adequately grasp the level of control which agents are
characteristically able to sustain reflexively over their conduct“ (Giddens, p.5).
Knowledgeability is thus central in Giddens’ stratification model, but in a
precise and limited way. He identifies three layers of cognition/ motivation
characteristic of the human actor: “discursive consciousness, practical
consciousness and the unconsciousness“ (Giddens, p.376). Self-reflexivity
corresponds to the first layer in the stratification model: “To be a human being
is to be a purposive agent, who both has reasons for his or her activities and is
able, if asked, to elaborate discursively upon these reasons“ (Giddens, p.3).
But he is careful when referring to terms like ’purpose’ and ‘reason’ -- he does
not want to ascribe to them the sort of ’hermeneutical voluntarism’ they have

ofien been associated with in philosophical literature. Whereas the first level is

17

one of discursive knowledgeabflity, the second is one of practical
knowledgeability (which Giddens sees as a factor of greater importance than
the discursive consciousness in day-to-day activities). Both this second layer
and the third (the unconscious) represent grounds for action which differ from
rationalizable, goal-oriented reasons for behavior, in that they cannot be
immediately represented in discursive form, indeed, often individuals would be
incapable of producing a verbal account of these grounds for their action. Even
if a person can present a coherent and probable explanation for her behavior,
this does not mean that this account represents all relevant information of the

causes for that behavior. Reasons are simultaneously found at several levels,

 

not only that of discursive consciousness. Giddens reminds us that the
tendency to “equate reasons with ’normative commitments’ should be resisted;
such commitments comprise only one sector of the rationalization of action. If
this is not understood, we fail to understand that norms figure as ’factual’
boundaries of social life...“ (Giddens. p.4 ). The stratification model assumes
that all these three levels must be included in a probable account of human
social behavior, that they together form an account of social action (practice) as
“embedded sets of processes” (Giddens, p.3).

Contextuality: “The situated character of interaction in time-space,
involving the setting of interaction, actors co-present and communication
between them” (Giddens, p.371). We may say that contextuality is a concept
that describes one particular instance of social interaction between agents,
whereas practice is used as the repetitive, structured, and reoccurring forms of
social interaction. According to the stratification model, individuals who engage
in social practice, are not only constantly self-reflexively monitoring their

activities. they are also monitoring the conduct of the other actors, and they

18

are also being monitored by the others. This happens at the level of discursive
consciousness, which explains part of the interactive aspects of social
behavior, but not all of it. More important to the ongoing flow of social
interaction is the practical consciousness, a type of knowledgeability which

cannot be expressed discursively (Giddens, p.375):

The vast bulk of... mutual knowledge incorporated in encounters, is not
directly accessible to the consciousness of the actors. Most such
knowledge is practical in character: it is inherent in the capability to ’go

on’ within the routines of social life. (Giddens, p.4)

So far we have seen that Giddens wants to describe social behavior in a
multi-faceted manner, of which one level (the discursive consciousness)
permits for explanations similar to that of the agency models; the two others
are influenced by structure (Giddens’ version of structure, that is). None of
these three levels is in and of itself sufficient to describe social behavior: this
is the reason why we must both consider agency and structure to give a
probable account of any contextual interaction. Since we are all familiar with
the experience of planning and deciding upon a course of action, I will assume
that the agency component of Giddens’ model is the one which is most readily
accepted. The structurally determined components of practice, on the contrary.
demands some more explanation. What are the mechanisms by which our
action can be influenced ’from the outside“? One reply to this is formulated in
Giddens’ concept of routinization:

Routinization is the habitual, taken-for-granted forms of daily activities.
all the standardized forms of action and interaction that makes the flow of
social life smooth and predictable. Routinization is expressed in the familiar

habits and standards of behavior which are central for our ability to participate

19

in social interaction. Not only does it serve to organize and structure goups of
human individuals, it is also important for each person’s sense of order and
predictability: “Routinization is vital to the psychological mechanisms whereby a
sense of trust or ontological security is sustained in the daily activities of social
life“ (Giddens, p.xxiii).

The probability of Giddens’ structuration theory partly rests on the
plausibility of his account of routinization. The duality of structure is formulated
as a claim that structure is both a medium and an outcome of the ongoing flow
of social interaction in day-to-day activities. We may say that structure is
reflected in routines, which pose both as a medium and as an outcome of
practices: “Routine is integral both to the continuity of the personality of the
agent, as he or she moves along the paths of daily activities, and to the
institutions of society, which are such only through their continued
reproduction“ (Giddens, p.60). The concept of routinization is grounded in the
second level of human cognition/ motivation in the stratification model: That of
practical knowledgeability. Routines thus cannot be described merely in terms
of structure (as if they were an external ’ruler’ of behavior), neither can they be
understood in terms of agency (as if they all represent the product of
discursive consciousness). Of course, routines may be challenged, interrupted,
questioned by those who engage in them. But the ’bulk’ of everyday activities is
so complexer composed of routinized behavior, to question them all would be
perplexing for the ongoing flow of action and interaction.

Another mechanism by which the structure influences our behavior
’from the outside’ is described in Bourdieu’s account of ’strategies.’ As social
beings we are more than biological organisms -- we occupy a ’position.’ Two

such positions are those of ’physician’ and ’patient.’ Through rules (codes of

 

20

behavior), each of those positions grants the biological organism who occupies
that social position access to a set of structural resources. To ‘get one’s will’
one may try to mobilize those resources, and with skill and cunning, one may
get a lot done -- be powerful. This power does not come from within, it resides
in the structure, who in this sense in not only restricting, but also an enabling
factor (Bourdieu, p.8-9). Again, the practices approach reminds us that we
must consider both agency and structure: our ’free will’ is exercised as well as
modulated within a structurated reality.

The extent to which the different layers of Giddens' stratification model
form an embedded, unified whole can be illustrated by the presence of
routinized, practical knowledgeability at the level of discursive consciousness
expressed as self-reflexibility . To illustrate this point, I will consider a face-to-
face interaction between two individuals: “Focused interaction occurs where
two or more individuals co-ordinate their activities through a continued
intersection of facial expressions and voice“ (Giddens, p.72). The ongoing self-
monitoring, monitoring of the other, as well as the sense of being monitored is
characteristic for interactional practice. This monitoring and awareness is
grounded in the first layer of the stratification model -- that of discursive
consciousness -- in the sense that each individual upon request would be able
to give a verbalized account of the meaning of their behavior and that of their
interactional partner. But to say that this monitoring of one’s self and of
another purely arises from the level of conscious knowledgeability, would be to
miss the vast impact on social interaction by rules and principles, as well as
the impact of the subconscious. At this level, not immediately verbalizable, '
often not verbalizable at all, are all the principles that preclude social behavior

which most people would find bizarre, incomprehensible, threatening to their

 

21

sense of ontological order. There is no need to designate a list of what such
behavior could amount to, indeed, most of it is also barred from my own
imagination, let us just agree that random facial grimaces is not the most
outrageous expression of what is routinely excluded from acceptable social
behavior.

How then does Giddens account for the role of power in social
interaction? He agrees with the commonly held view that power is a capability
to bring about consequences, “to ‘make a difference’ to a pre-existing state of
affairs or course of events“ (Giddens, p. 14). But the capability to ‘make a
difference’ within the structuration model does not arise from some qualities or
capacities intemal to individual actors. Rather, the exercise of power depends
on mobilization of properties found in the structure. The media through which

power is exercised is that of resources:

Resources (focused via signification and legitimation) are structured

properties of social systems, drawn upon and reproduced by

knowledgeable agents in the course of interaction. (Giddens, p.16)
Giddens has defined structure as a set of rules and resources, so we see that
power then is one mechanism by which the components of structure may
influence the course of social interaction; the other is the normative and
significative impact of rules. Resources have been described as either
authoritative or allocative, and from this it follows that power may either take
the form of the capacity to influence and control other persons founded on the
signification or legitimation allocated to the person who exercises authority; or
of the capacity to influence or control others founded on the allocated

command over material goods, or over other material phenomena. To ‘be

22

powerful’ should be understood as occupying a position that through ongoing

practices has acquired a legitimate control over structural resources:

Power within social systems which enjoy some continuity over time and
space presumes regularized relations of autonomy and dependence
between actors or collectivities in contexts of social interaction. But all
forms of dependence offer some resources whereby those who are
subordinate can influence the activities of their superiors. This is what I

call the dialectic of control in social systems. (Giddens, p. 16)

Power, in Giddens view, is not in itself a resource. Nor is it a quality of
individual agents. Power is the capacity of bringing about consequences. This
capacity arises as a product of a, social process which involves individuals
engaging in structured activities.

Russell and Weber focused on situations of conflict in their analysis of
power, the same is true of Poulantzas and Marx, as well as a number of other
authors adhering to both the structural and the agency theories of power. The
authors within practice approaches, on the contrary, have been concerned with
the whole range of social practiCes. Bourdieu typically uses trivial examples to
illustrate his points: a handshake, a friendly greeting, a threat or a favor, or a
gift. The importance of all these activities is that they are structured (and
structuring) social practices, Bourdieu claims (Bourdieu, p. 194-5). Giddens
states that instances of power are not particular to specific types of social

conduct, rather power is a capacity to activate structural resources which

function as a routine element of all social interaction (Giddens, p.16).

Difl'erent ways of combining structure and agency
The agency view of power most closely resembles the trivial, ad hoc

explanations people tend to form when they ascribe a causal role for social

23

effects to individuals or groups of individuals: ’He caused me to feel like an
idiot’; ‘They persuaded us to stop working,’ etc. The structural view of power
holds that the reason for social behavior is not found in the realities
immediately available of the situation we want to analyze, and shifts the focus
to some extra-individual reality: “The ultimate reason why she acted like she did
is found in some religious ideas instilled in her’; ’Society causes people to
commit suicide,’ etc. The practice approaches shift the focus back to the
realities of the situation we are evaluating, but ascribes to power a more
complex nature, one which has to include considerations about the structural
reality: ‘As an effect of the role ascribed to her by the organization she
represents, the executive adopted an intimidating stature toward her inferiors;
in that sense her behavior is not only a function of the organization, but also
her particular contribution to the construction and reproduction of the
organization.’

If we agree that the agency view of power alone is uncompelling as a
theory of all types of human social interaction, and likewise the structural
model, we are left with the task of finding a way of talking coherently about
power which permits consideration of both structure and agency. I will
describe four different ways this could be done. The first two are less
compelling than the two that follow, but nevertheless described, since they
reflect assumptions about power present both in informal and scholarly
discussion.

The first option would be to expand the definition of agents so that it
included all types of structural entities -- our social universe would then merely
consist of different types of agents, some individual, others structural.

Structural agents would be akin to other organized entities (like human beings),

24

striving to obtain its own goals, and also accountable for its actions. This way
of thinking is reflected in the following statement form an article in Time: “How
did it happen, the public wondered angrily, that Elisa’s case was known to the
system, and yet the system so shamefully failed her“ (van Biema, p.34). In this
case, the ‘system’ is held morally accountable for the death of a six year old girl
who was abused and eventually killed by her mother. The ‘system’ is, in this
case, the city authorities, who had been notified at least eight times about the
circumstances of continued violence and terror under which Elisa lived. As an
abstraction, or a metaphor, this way of speaking may provide useful insights,
but if we want to say that a ‘system’ really is an agent, which should be treated
similarly to individual agents, we run into problems. Whereas single actors are
entities spatially limited by the extent of their bodies, social systems have no
such spatial unity. For this reason, and others, the type of agency and moral
and legal accountability we could ascribe to systems would have to be so
different from what we could ascribe to individuals. To treat them as one class
of entities (social agents) would confuse, rather than improve, our thinking.
Indeed, Elisa’s mother, and not the ‘system’ was charged with second-degree
murder; and the critique of the ‘system’ in the Time article bofls down to
critique of single individuals: a judge, awaiting reports from the leader of a
federally run parenting program, who in turn awaits for reports from the Child
Welfare Administration caseworker, who in the article is reported to have said
that he was “too busy“ to stop by Elisa‘s home. Both for practical and
theoretical reasons, this way of reconciling the opposition between structure
and agency is uncompelling.

The second approach is to include the terms of structural thinking

(institutional constraints on physicians’ behavior, the impact of culture on the

25

patients’ perception of illness, etc.), and then conduct a discussion which in
essence is purely based on the agency approach to power. The structural
elements are then merely used to represent a scenario, useful as background
information, in the same way as genetic components (inherited low IQ);
physiologic components (aggressiveness caused by a hormonal imbalance), etc.,
might provide relevant information. Once all the relevant background
information is gathered, we can get down to the real analysis, which still deals
with how the doctor and patient interact as single individual agents. Structural
elements are in this view regarded as ‘analytical noise,’ which should be filtered
away to better understand how the true selves (the knowledgeable agents)
interact as they do. The central methodologi and assumptions of this approach
are similar to a pure agency view.

A third model is offered by the practice approaches. Giddens’
structuration model has illustrated this view. According to the practices
approach, power is not an entity possessed either by agents or the structure,
rather it is the product of a social process which involves both agency and
structure. In this model, the sources of power are located in a structured
reality, but activated through agency.

A fourth approach, which I will refer to as theoretical pragmatism, holds
that our models about the world should not be based on any definite location of
power. This could be argued either by holding that there is no such thing as
power in the real world -- power is only a theoretical construct designed to
make sense of the human conditions; or by holding that theoretically power
might well have a fixed location, but since we can never know for sure or agree
on what its location is, we would be better off basing our theories on other

grounds than ideas about a definite power locus. According to the pragmatic

26

approach, if we want to discuss the impact of social institutions, it would be
permissible to view power as a purely structural entity. If, on the other hand,
we want to describe single individuals, we might view power as a resource or
capacity possessed by that person, independently of structural consideration.
For certain types of social interaction, again, the best option might be to view
power as the product of a process which involves both structure and agency.
For the pragrnatist, statements like ’he is powerful in himself,’ and ‘the
structure by its own virtue exercises an influence over people,’ may both be
valid and sensible statements, depending on the context in which they are
uttered. The proponent of the practices approaches would object to such
statements, claiming that it is misleading to suggest that a person by virtue of
her own characteristics and capacities could be deemed powerful, regardless
of the structure in which she is situated.

Theoretical pragmatism, like all eclectic approaches, has the advantage
of providing a set of different theoretical tools, which makes it a dynamic.
adaptive, multi-faceted approach to theoretical problems. There is no
conclusion reached from more stringent models that cannot simultaneously be
reached from the pragmatic position, since the pragmatist may merely add to
his toolbox the methodology of the more stringent tradition. But theoretical
pragmatism also shares a problem with all other eclectic approaches: How can
we decide which tool to use if two of the available tools would yield different
types of conclusions in a given situation? This illustrates the strength of the
practices approaches: it provides a unifying theory which describes within one

coherent theoretical framework the interaction between structure and agency.

3: POWER IN THE CLINICAL SETTING

He who needs something may hope that somebody else has what he
needs. The object of his hopes, the provider, has power over him. First, she
has the power to shatter his hopes, to say that she does not have what he
needs. Second, if the provider has what he needs, she has the power to give or
withhold it.

The patient comes to the physician with a demand, assuming the
physician has something that the patient needs. That ‘something' is one obvious
source of the physician‘s power. Sometimes this source is theoretical
knowledge about symptoms, treatment, and prognosis. It may also consists of
practical skills: diagnosing a fractured arm, putting on a cast, suturing a
wound. When the patient wants a medical certificate or a drug prescription.
the source of the physician's power is the formal authority to issue such
documents. The physician’s power can also arise from a combination of medical
and psychological knowledge, and personal charisma. This may be the case in
the treatment of a depressed patient, one that wants help with a drinking
problem, or one that wants to lose weight.

Having something the patient needs is only one among many sources of
the physician’s power. She decides how much time -to spend with the patient.
She controls the content of the conversation: she asks the questions, she
interrupts, she changes the topics (Meetuwesen et al.). She decides where the
patient should sit or lie down. and whether or not he keeps his clothes on. If
she finds it necessary, she touches the patient’s body, sometimes even the
intimate areas. Based on his own experiences in medical training, Pappas

writes:

27

 

28

Perhaps the most profound manifestation of power in doctor-patient
interaction is its disposition of the human body. The intimate nature of
the physical exam involves the physician taking control over the body,
even its very motion... Descriptions of the simple techniques of
percussion, auscultation, and palpation do not convey the intimacy of
the physical exam, which unlike even sex, requires near total surrender
of the body. (Pappas, p.202)
To be powerful
Power is thus typically unevenly distributed between the physician and
the patient, and for a number of various reasons. Still, it is not the case that
the physician can manipulate the patient to do whatever she wants him to do.
If the patient refuses to submit himself to a clinical examination, none of the
physician’s sources of power give her capacity to force her will through. She
might of course try to use her authority, for example by pointing out that the
patient might be suffering from a dangerous condition, and that the quickest
way to exclude this possibility would be to undergo a physical examination. But
if the patient distrusts the physician, or for other reasons refuses to succumb
to her authority, she has no power over him. Rather than the physician being
powerful by the virtue of her training, her skills, or her position alone, we see
that her power arises as a product of the interaction between her and the
patients -- through this process the legitimacy of her power is established,
reproduced (and sometimes, challenged). Skills and insights are not in
themselves sources of power, they become so only when the patient ‘invests’
his trust in them. The physician’s skills and her insights are hers, the patient
cannot take them away from her. But her power is not hers alone, and can, at

any moment, be taken away from her (although this does not happen very

often, we should not think it cannot not happen).

29

In other words, the physician is not powerful by herself. Her power is a
product of a social process. If the physician's power appears as a static entity,
rather than a dynamic and alterable product, it is because the contextuality of
each clinical encounter follows established canons of appropriate behavior for
the practice of medicine: the patient adapts to a set of constraints and
expectations embedded in the ’patient role,’ and the physician acts like
physicians are expected to act. To say this is not to suggest that the patient's
and the physician’s behavior are mechanical responses to the force of tradition.
The two roles neither embed universal or necessary sets of behavior, rather
they represent the effect of well established structural rules, which dictate
both the norms and meaning of behavior in the clinical setting. Giddens refers
to such a situation as the arising from routinization. Routinization helps to
coordinate behavior without inventing the set of social rules form scratch for
each new encounter, as well as it serves to produce a sense of ontological
predictability, crucial for the feeling of trust, as well as control. If the
physician’s phone rings during the encounter, it will not profoundly challenge
her patient’s feeling of trust or control if the physician spends a couple of
minutes on the phone. If, on the contrary, the patient’s wire -less phone rings,
and the patient engages in a conversation, the physician's sense of control and

order would be threatened:

We can probe the psychological nature of the routine by considering the
results of situations where the established modes of accustomed daily
life are drastically undermined or shattered -- by studying what may be
called ’critical situations'. (Giddens, p.60)

The principle of who may answer phones and who may not in the physician’s

office is not a universal law, dictating the content of the clinical encounter from

30

the ‘outside,’ and it is not a rule which could not, in principle, be re-negotiated.
But neither is it the product of a conscious reflective process, by which the
physician and the patient come to agree that this rule is the best way to
organize their encounter (at least, for most patients and physicians it is not).
Rather, the phone -rule is an embedded part of the practical knowledgeability
which facilitates the ongoing flow of interaction between physician and patient:
a routinized form of knowledge.

Throughout the clinical encounter, both physician and patient engage in
a constant reflexive monitoring of their own behavior, and of each other's
behavior. The experienced physician can also rely on an embodied practical
consciousneSs. represented in automatized sets of behavior which she
performs without having to reflect consciously on them. The patient also has
his practical consciousness, but since the setting is unfamiliar. and since he is
alert, maybe stressed, his level of knowledgeability is of relatively less help to
him than the physician’s practical knowledgeability is for her. To guide his
actions, he will have to rely more strongly on his discursive consciousness, and
most probably, the setting will make him feel somehow stressed, incapable,
and powerless.

Even if we agree that the physician’s power is a product of the mutual
interaction between her and the patient, it would be unwise to suggest that we
should refrain altogether from speaking and thinking of the physician as a
‘powerful person.’ This is because phenomenologically, we have all experienced
situations where we felt ‘powerful' or ‘powerless‘ in and of ourselves, and we
have all been exposed to people who we perceived as ‘powerful,’ in that they
could get us to do things we did not want to do. Not only does the statement,

‘she has power over me,’ correspond to a universally known experience, that

31

type of statement is also so widely and commonly used. We are not likely to do
away with such phraseology. What we should do though, is to get clear on
what exactly that statement signifies, to avoid being confused about the nature
of power relationships. Thus, I suggest that to ‘have power’ and to ‘be powerful’
should be taken to mean the same thing. They both arise from having access to
sources (skills, capacities, social position, material resources, etc.) which
through a contextual process is validated as structural resources (authoritative
or allocative). These structural resources may be activated by the powerful
person to bring about social consequences. To ‘have power' or to ‘be powerful’
is a relative concept, one is only powerful within a certain context, and in
comparison to another agent: “Power within social systems which enjoy some
continuity over time and space presumes regularized relations of autonomy and
dependence between actors or collectivities in contexts of social interaction”
(Giddens, p.16).

It may confuse the reader when I talk about the physician ‘being
powerful’ and ‘using power,’ since the point of this thesis is to argue that power
is not something you ‘have‘ or ‘use‘ in the same fashion as you may have
money. or use your physical strength. To be in line with Giddens' theory it
would indeed have been better to talk about ‘having capacities’ and ‘activating
power’ -- but again, I can hardly expect to change everyday language, so [will
instead accept it, but retain a sensitivity to its potential for creating confusion
about the reasons why some people are conceived as powerful.

Even in well established unequal power relationships, the control
exercised by the powerful is never absolute. Giddens reminds us that the
subordinate always has access to some resources, which, if skillfully

employed, may be used to influence the behavior of the more powerful. From

32

this, it follows that the mere fact of getting one’s will with regard to a particular
desire or objective, does not signify that one is in a position of power: To be in
power entails a temporally relatively stable situation of authoritative or

allocative superiority.

To use power

In the remainder of this chapter, I will be particularly interested in the
expression ‘to use power.’ I will hold ‘uses of power’ to be situations where A is
more powerful than B; and A consciously uses this power advantage to bring
about particular changes in the conditions of B. When I say consciously, I want
to include both the discursive and practical consciousness from Giddens’
stratification model. This means that if a physician is self-reflectively aware of
the fact that she uses her position of authority to influence the patient in a
certain way, she is then ‘using power.’ But also instance where she influences
the patient through practicing her skills, without being self-reflectively aware of
all the consequences of her behavior, should count as uses of power. Whereas
a discussion of the mechanisms by which the physician is constituted as
powerful must focus on the process and mechanisms of interaction in the
clinical encounter; a discussion of when it makes sense to say that the
physician has exercised power takes an unequal power relationship for
granted, and focus on the extent to which the power imbalance can account for
specific outcomes of the interaction.

I will compare the concept of ‘use of power‘ With the so-called placebo
effect in three clinical cases. The first one is a commonplace clinical encounter,
presented in an article by Brody. The other two cases are more spectacular:

one told to me by a lecturer in psychosomatic medicine; the other a piece of

33

fiction. In discussing the cases I will focus on the impact of the physician’s
power on the outcome of the encounter, and in particular on the placebo effect.
I will argue that the placebo efi"ect in the first two cases represents instances
of a physician using power. As for the third case, even though spectacular, I
will argue that the physician did not exercise power in that case.

A patient presents to his primary-care physician with rhinitis and other
signs of an upper respiratory viral infection. He expresses fears that he may have
pneumonia, since a fellow worker with similar symptoms had to be hospitalized
for this illness a week before.

After taking a careful history and auscultating the chest, the physician
announces that pneumonia is quite unlikely and that the worrisome cough is
probably due to nasal secretions. The doctor recommends a decongestant along
with rest and fluids. The patient immediately starts to feel better and. two days
later; is almost completely recovered (Brody 1 986, p. l 06- 7 ).

To assess whether or not the physician ‘uses power‘ in this case, we
must first identify an outcome; and then discuss whether or not the physician
was central in causing this outcome. Identifying the outcome is a task that
poses it’s own theoretical problems. I do not want to engage in such discussion
here. Rather, I will assume that it is plausible to view the outcome of this
encounter as consisting of four components: 1) Agreement on a diagnosis (also
excluding the probability of differential diagnosis); 2) Agreement on an
appropriate treatment; 3) The patient instantly feeling better; 4) Recovery some

time later.

34

The power to determine diagnosis: The diagnosis is reached through a
process where the patient presents his complaint and the physician gathers
supplementary information by asking questions and conducting a clinical exam.
We can assume that the physician reasons by comparing this particular case
with paradigmatic cases that come to her mind. and with her body of general
medical knowledge. This reasoning leads to a conclusion that takes the form of
a probable diagnosis. and a statement that the patient is most likely not
suffering from what he had feared. We can only guess what line of reasoning
lead to the physician’s conclusion, and how sure she is about the diagnosis
being correct. Maybe she feels convinced that the case is unproblematic, and
she adds the modifiers (probable, quite) only because she knows that in
medicine even the surest thing is never one-hundred percent sure. The use of
modifiers is also encouraged by the constant threat of being sued for
malpractice: most expressions of certainty will look bad if quoted in a court
case.

Having analyzed audiovisual recordings of physician-patient encounters
in family practice, Richard Street concludes that “through their communicative
styles, patients can exert considerable control of the amount of information
they receive from doctors” (Street, p.546). Physicians may also encourage
patient participation and input in the clinical reasoning. In the case discussed
here, the patient might ask: What do you mean by probably; What- other
diagnosis had you thought of; Why did you rule out those others? A physician
who is motivated to encourage patient participation will give open and truthful
answers to such questions. But no matter how strong the patient’s desire for
being in control, and how strong the physician's sympathy for this desire - the

ultimate power to set the diagnosis lies with the physician. This is so, not so

35

much for traditional and legal reasons, as it follows fi'om the nature of the
doctor-patient encounter. It would be impossible, as well as bad medicine, for
the physician to give a complete disclosure of all the reflections and whims that
passed her mind in the process of reaching a diagnosis. Maybe a medical
rarity, like a primary tuberculosis of the nasal septum, had come to her mind
while examining a patient with a probable common cold. Though conceivable,
such a diagnosis is exceedingly rare. If she felt obliged to disclose this
differential diagnosis to an interested patient, her explanation would generate
more questions, and the encounter would have no end.

The power to determine treatment: We do not know exactly what happens
where our summary case description states that the doctor “recommends a
decongestant along with rest and fluids.” If the physician bluntly states what
treatment is recommended, and this is accepted without questioning, we have
a clear-cut example of physician using her power to influence the patient. For,
obviously there are other treatment alternatives available, both more and less
aggressive. If the physician presented this treatment as the only possible
response to the condition at hand, the advocates of patient autonomy and
informed consent would probably hold this to be an example of physician
paternalism, or translated to power terms: unjustified use of physician power.
According to the model of patient autonomy developed in the 1970’s, the
physician‘s role should be strictly confined to presenting facts to the patient,
leaving to the patient to make his own decision, fi'ee of coercion, and purely
based on his own values. In his critique of this model and later more refined
versions, Brody argues for a conversation model, where “the goal is to help the
patient make the best decision, not to provide the patient with maximal

knowledge” (Brody 1992, p.99). Let us assume that the physician in our case

36

had presented to the patient her three most plausible responses to the given
conditions (no treatment; decongestants; antibiotics), and the arguments for
and against each of those. This could possibly enhance the patient’s ability to

make the choice that best fits his personal preferences. Brody points out that:

At any rate, the general statement. ‘In your situation, some people would
do X, whfle others would do Y.‘ is often suggested by skilled
interviewers to give the patient the widest possible permission to make a
free choice. (Brody 1992, p.98)
Some studies suggest that patients are well equipped to choose between
alternatives even when the alternatives entail considerable risks. Other
research on the contrary indicates that chances of choosing the optimal
treatment diminishes if extra options are added to a list of possible treatments:
“adding new options can increase the probability of choosing a previously
available alternative or, in particular, of maintaining the status quo” (Redelmeier

and Shaflr, 304):

The economist Schelling, for example, tells an anecdote about going to a

bookstore to buy an encyclopedia, discovering that two different

encyclopedias were on sale that day, and purchasing neither because he

lacked a clear reason for choosing one over the other.2

For the purpose of our discussion, namely to determine whether an
asymmetrical power relationship is likely to have caused the patient’s decision.
the point is not to determine how many different alternatives the physician
ought to present in order to enhance patient participation. For, ultimately it is

only the physician who can judge what alternatives to present, and how to

 

2 ’l‘versky A. Shafir E. Choice under conflict: the dynamics of deferred decision. Psychoi
Sci. 1992;3358-61 (Quoted by: Redelmeier and Shafir. p.302).

37

present them. Paraphrasing two social scientists. we can say that power may
be, and often is. exercised by confining the scope of decision-making to issues
that the physician finds acceptable.3 This sort of influence is not objectionable
per se. On the contrary, we might say that it amounts to what Maseide refers
to as a necessary form of influence in competent clinical work (Maseide.
p.553). The list of treatments a skilled physician finds potentially acceptable in
the case of one particular patient, is exactly what the patient is looking for.
One could say that by choosing to visit a physician, the patient is implicitly
asking to be the object of this sort of influence.

What causes the patient instantly to feel better, and to recover some time

later? In the same article as our case is taken from, Brody writes:

Throughout the history of Western medicine, three basic types of
explanations have been offered to explain ill patients’ improvement.
Explanation 1: The patient improve because of the specific
pharmacologic or physiologic potency of the treatment prescribed.
Explanation 2: The patient improved because of the natural history of
the disorder and the body’s inherent recuperative powers. Explanation
3: The patient improved because of the symbolic dimensions of the
healing encounter (in this case. the patient’s hOpe and expectation that
the physician could help, the reassurance that the patient did not have
pneumonia, and the comfort promised by the decongestant).

(Brody 1986. p.115)

The placebo effect is the change in a patient’s condition arising from the
symbolic meaning of the clinical encounter, independent of the pharmacological
properties of drugs. and other medical interventions. It corresponds to the

third explanation in the scheme above. Explanations for the placebo effect are

speculative: endorphins have been suggested. as well as the catecholamine

 

3 Barach P. Baratz MS. Power and Poverty. Theory and Practice. New York: Oxford
University Press, 1970 (Quoted by: Lukes 1974. p. 18).

38

system, and psychoneuroimmunological responses (Brody 1988, p. 151). It is
often claimed that the placebo efiect has an impact in virtually all medical
encounters, sometimes alone, sometimes in addition to the other two
explanations suggested. The placebo effect can be demonstrated in clinical
research: K.B. Thomas reported a significant difference in self-reported
recovery between patients that were given “a firm diagnosis and told
confidently that they would get better in a few days” and patients told “I cannot
be certain what is the matter with you” (Thomas, p. 1200).

As for our case. there is no reason to think that the natural history of
the patient’s condition would entail an abrupt amelioration at the end of the
encounter; and since the patient had not yet started the proposed treatment
regime. we must conclude that the ‘immediately starting to feel better’ was
probably caused by the placebo affect.

Theoretically, each of Brody’s three explanations could by itself provide a
sufficient explanation for the complete recovery after some days. As we could
see the placebo efiect demonstrated immediately, we must believe that this
effect also played a part in the process of further recovery. In cases where
treatment plays a central role in recovery, the physician’s skill is typically the
major influential component. The physician’s influence is either direct, as in
surgery; or indirect. as in the choice of drugs and other treatment the patient
administrates himself. In this case we can only speculate about the impact of
the physician’s impact on recovery via choice of treatment. The natural history
of the disorder is by its definition immune to influence fi'om the physician.

That leaves us to assess the impact of the physician’s power on the
placebo response. It might be tempting to see the placebo effect as a prime

example of the physician’s power. But, as I will argue later. all instances of

39

placebo are not necessarily instances of power. Dr. Thomas is clearly
exercising control over the patients that get better when he assures them that
they will do so. His source of power is an explicit knowledge of his ability to
influence the patients: “The doctor himself is a powerful therapeutic agent; he
is the placebo and his influence is felt to a greater or lesser extent at every
consultation” (Thomas, p. 1200). But does the physician in our case possess a
similar source of power? The case is presented in a dry and sober fashion; we
cannot know whether this physician shares Dr. Thomas' insight in her potential
as a ‘human placebo.’ Still, her actions convey at least an implicit
understanding of the symbolic dimensions of the clinical encounter. Even
though the symptoms are relatively sparse. the physician takes a “careful
history.” she auscultates the chest, and she specifically addresses the patient’s
fears in her conclusion. She also suggests a treatment ritual that would remind
both the patient and his surroundings that he is legitimately ill. and should be
excused from his normal duties. The practical skills of this physician embrace
an insight in the placebo efi‘ect: these skills mobilize power which rests in the
structure of the clinical encounter. More specifically. the structural resource of
power in this case is a type of authoritative power, founded in the patients’
beliefs in the physician’s healing powers. Through her practice (what Giddens
calls practical knowledgeability) the physician mobilizes this power. We will
now leave this case, and consider the first case of mysterious healing:

A seven year old boy comes to the physician accompanied by his mother.
Part of his hands are covered with persistent warts. Several treatments have
been tried. with no effect. The boy is unhappy, his mother concerned, and the
physician tired of being of no use in the war against the warts. A strange idea

suddenly strikes the physician, a plan he cannot resist. He takes the boy ’5 hand

40

and studies it carefully. Then: ‘There! There, do you see it? The mother wart. All
the other warts are her children. ’ He proceeds to dramatically sticking a syringe in
the center of the big wart, announcing ‘Now that the mother wart is dead, all her
children must die too. ’ Within two weeks the warts are gone (This story was
passed on to me by Professor M. Patris, at the psychiatric clinic of Strasbourg
University Hospital).

This is an example of pure placebo effect, and also a striking example of
the physician’s capacities. It permits us to more clearly analyze the impact of
power in cases of placebo. What is the source of power in this case: the
physician’s cunning idea; the magic role he assumes in the interaction; or is
the source of power within the boy himself? With an imprecise definition of
power. we could say that all three suggestion are valid evaluations of power --
indeed, in everyday language, .we would both speak of a powerful idea, a
powerful magician. and the patient’s self-healing powers. The physician’s
success here arises from a special type of knowledge. He has at least some
instinctive ideas about the patient’s transference mechanisms: “the process of
transference leads patients to project onto the [physician] all kinds of magic
expectations, hopes, and fears that are intrinsically irrational because they
emerge out of confusion not only of past and present but also of fantasy and
reality” (Katz, p.143). In this case the physician not only takes advantage of
unconscious beliefs, he actively leads the patient into a magical world. He lies
to the patient, this type of influence we might refer to as a form of benevolent
manipulation.

It is true in a sense that the boy must be seen as the possessor of his
own fantasy world, and consequently as possessing the source that causes the

warts disappearance. But his imagination by itself had been no source of

41

healing. The boy‘s magical beliefs are consciously being used as the physicians
tool for doing away with the warts. The physician knows something about his
patient that the patient himself does not know: that if he believes strongly
enough that the warts will disappear, that will actually happen.

Respecting Giddens’ structuration theory, we should say that the
physician’s insight, as well as the patient’s fantasy world represent aspects of
the resources each of them may mobilize. but not in and of themselves power,
or even sources of power. The physician’s insight becomes a source of power
in the process where the patient and the physician engage in interaction, and it
does so because his patient trusts him. In this connection it would be wrong to
think that boy’s imagination functions as a source of power: it is nothing he
uses to bring about changes in his environment. From his powerful position,
the physician is able to mobilize the potential of self-healing in his patient, and
this he does by employing his knowledgeability as a physician. The
knowledgeability he uses in this case is partly conscious, partly practical: the
innovative idea which “suddenly struck him” partly arising from a level of
discursive insights, partly from a level of intuitive knowledge embedded in the
practical skills of physicians. Lastly. we will look at one more case of
miraculous healing:

A twenty-six year old female student comes to a general practitioner. She
is experiencing an intense stomach pain. On examination, the physician finds
yellowish scleras, and a mildly enlarged liver. Suspecting liver disease, he takes
blood sample for liver enzymes (they later turn out to be normal). As the
physician is about to end the encounter, the patient sees a crucifix he carries
around his neck. She exclaims that this is the sign she had dreamed Qf, and that

the pain is now gone. Before the physician even has time to respond. the patient

42

is out of his office. The physician happens to encounter the patient a week later,
and notices that the scleras now are perfectly white.

Again an instance of placebo - at least for those of us who normally
prefer to explain healing in other terms than divine intervention. No treatment
has been suggested, and we have no reason to think that the natural history of
a suspected liver affection would be to heal spontaneously while the physician
watches. The symptoms are gone. and some objective signs vanish within
days. Maybe the patient attributes magical forces to the physician, we cannot
know. In spite of the similarities with the previous case, there is one
fundamental difference here: the physician is totally without control over the
outcome. Healing took place in front of his eyes, he did not initiate it, he did
not suspect it. he did not understand it. We could say that the episode is
powerful (meaning striking). We could refer to the fact that physicians, in
general, are in a position of power. But it would be wrong to claim that this is
an instance where the physician exercises power. The physician has become
the object of his patient’s mystical fantasies, he has been used as an
instrument in her auto-healing. The patient’s fantasies about the physicians
healing capacities do not by themselves make the physician powerful. Only
when the physician has some understanding (discursive or practical) of these
fantasies, and employs these from a position of power stemming from
authoritative or allocative resources, legitimated through the interactional
process between physician and patient, we could say that the placebo

response is mobilized through the physician’s use of power.

4: POWER AND ETHICS

In order to meaningquy discuss what we ought to do, one must make
some assumptions about what kind of beings we are, and about the natural
conditions in which we live, act, and interact. A particular paradigm is explicitly
or implicitly assumed to represent a plausible starting point for a moral
discussion: the validity of the moral conclusions ultimately turns on the
plausibility of the paradigm within which they are formulated. By paradigm I
mean a set of fundamental assumptions about the natural world and the
human mind. The focus on power makes it possible to understand one aspect
which ties insights from moral philosophy to insights from social sciences.
Some theoretical moves must be made to establish this connection:

Among the most fundamental problems to be dealt with at the level of
paradigmatic beliefs about the social world, are questions of why people do
what they do around me, and what causes me to act and feel the way I do.
Such questions could be formulated in terms of power. Within social sciences
power is, in its most general sense, seen as the capacity of causing socially
relevant effects. A locus of power is an entity that has the capacity to generate
socially significant effects. A moral agent is in ethical theory understood as a
being who can be held morally accountable for its actions (or non-actions). One
of the prerequisites for being a moral agent, thus, is the capacity to cause
morally significant effects (or by refraining from acting, not causing such

effects).4

 

4 For the purpose of this argument, we need not be concerned with the question of how
to define socially relevant and morally significant effects. The important thing is how the
two classes of effects relate to each other. I will hold that all morally significant effects
are necessarily also socially relevant -- since they directly or indirectly involve at least
one member of the social community. But not all socially relevant effects are necessarily

43

44

Moral philosophy might not hold all socially relevant effects to be morally
significant. Being a locus of power cannot then be a sufficient condition for
being a moral agent. But. an entity that possesses no power (that cannot cause
any socially relevant effects), can neither cause any morally signficant effects.
Being a locus of power is thus a necessary condition for being a moral agent.

If, for example, the locus of power is placed wholly outside individual actors (as
done in structuralism), then there is no agent that can act morally right or
wrong, and no-one to hold morally accountable. A radical structuralist, who
theoretically denies individuals all form of agency, cannot coherently talk about
moral obligation, at least not with the meaning most often ascribed to obligation
(perceiving a impetus to act in a certain way, and then voluntarily choosing to
do so).

Since an ethical theory is founded on a paradigm, and a view of power
forms a central part of a paradigm. we must assume that there is some form of
coherence between how people think about power, and what type of moral
theory they adhere to. I will not suggest that this coherence takes the form of a
mechanical, rigid set of interdependencies. Neither will I suggest that a view of
power is developed prior to (or after) a moral theory, or moral convictions. It
seems to me more plausible that there is some sort of interaction, often
unconscious, between a set of ideas about power, causes and effects (beliefs
about the empirical world) on one hand. and a set of ideas about moral
tightness and wrongness (moral convictions) on the other. The idea that people
actually think this way has its counterpart in a normative model, wide reflective

equilibrium. which says that this is a good way of thinking -- that. morally

 

morally significant. How dress codes change between social classes may for example be
interesting from a sociological point of view. but not necessarily so from an ethical

perspective.

45

speaking, we ought to reflect in this fashion when we are faced with moral

questions:

The method of wide reflective equilibrium is an attempt to produce

coherence in an ordered triple of set of beliefs held by a particular

person, (a) a set of considered moral judgments, (b) a set of moral

principles, and (c) a set of relevant background theories.5

In certain cases then. a debate over a moral question may be founded in
a difference between the paradigms upon which the moral assumptions are
made. In such cases the key to a moral resolution may lie in pin-pointing and

evaluating specific parts of the paradigms, for example the views about power.

Views of power in schools of moral philosophy

I want to be a little more specific in suggesting what coherence there
might be between views of power and moral theories. I think that moral
philosophy is limited by the view of power it assumes. The connections I make
are based on what I take to be resemblance of central features. I want to
suggest, rather than prove, that such relationships exist. My conclusions here
will have to be modest, because of the ‘looseness’ in the way I have described
paradigm; and because there is no space here to discuss the particular ethical
theories in any detail.

The agency view locates power within individual agents. One way of
locating moral responsibility is to trace backwards through time, in search of
the true origin of a particular effect. Once we find the causative locus of the

effect we are interested in, we have simultaneously found the morally

 

5 Daniels. N. “Wide reflective equilibrium and Theory Acceptance in Ethics.” The Journal
ofPhilosophy 76: 1979 (Quoted by Clarke. p.241).

46

responsible locus for that same effect (if the locus is an entity that can be held
morally accountable). This line of thought matches fairly well with
foundationalist theories of ethics. They will typically formulate some general
rules or principles that apply to all moral agents, moral agents could here be
described as self-reflective loci of power. We can then say that the agency view
of power fits nicely with, and permits for foundational styles of moral reasoning
(like deontological and teleological models).6 Moral cynicism. the view that each
and all agents act according to what fits their personal interests best, and that
this is OK, is also compatible with the agency view of power.

The structural view gives power a definite location outside the individual.
This is bound to somehow ease the moral burden that rests on the shoulders
of rational agents within ethical theories based on the agency view. The extent
to which single individuals are released of individual moral responsibility will
depend on how consistently and deeply held the structural view of power is. A
view that places all power in the structure would indeed hold that there can be
no such thing as morally good or bad actions, since we are all acting exactly
like the structure predetermined us to act, neither better nor worse. If anyone
is to be held morally responsible, it would have to be the structure itself. We
see that the idea of formulating a valid moral theory collapses within a stringent
structuralist view. We could call such a position moral apathy.

A less extreme version of structuralist thinking is to claim that
fundamental moral convictions are instilled in us by the culture in which we

are raised, morality is therefore wholly context -dependent, and it is only

 

6 Kant’s maxim ‘never treat a person as a means. but always as a goal in itself is the
base of the most prominent deontological models; Bentham/ Mill’s maxim ‘act so that
you maximize the sum of utility’ has been the common starting point of teleological
models.

47

according to the local code a person can be held morally responsible. But --
this position grants at least an element of agency to single individuals. and
should not be classified as coming from a pure structuralist view of power. It
would better fit with what I will describe as ethical pragmatism.

The practices approach holds that power arises from a process that
takes place when individuals engage in social interaction. Rather than being a
static entity, residing in agents or in structure as a quality characteristic of
these. power is seen as a dynamic entity, produced and reproduced through
practice. Individuals exercise agency, but within structural constraints.
Through routinization. the structure also has a modeling impact on the agency
of individual actors. This account of power shares some characteristics with
the Aristotelian concept of virtue. Virtue is a durable, internalized disposition to
act morally rightly, acquired by engaging in practical activities7. One view of
virtue is to see it as an entity by which human qualities deemed cherishable by
a community are instilled in a person.

Theoretical pragmatism accepts structure, agency, and practices
approaches as plausible models of power. This happens by a completely
different approach fi'om the one by which the practice approaches assimilate
structure and agency. For, whereas practice approaches still hold that there is
something definite to be said about power, and indeed propose a specific
mechanism by which power functions, theoretical pragmatism is skeptical, or
rather indifferent, to the question of where power really is located. The concept

of power, like the concept of God, can only be evaluated in terms of what

 

7 In his rewriting of Aristotelian ethics, Macintyre defines a virtue as “an acquired
human quality the possession and exercise of which tends to enable us to achieve those
goods which are internal to practices...“ (Maclntyre. p. 191).

48

capacity it has for forming coherent and probable models of social life, the
pragmatist holds. For certain purposes one power model may serve, for other
purposes we would be better off using another model. There is no truth about
power, or--if there is. we can still not quite grasp it, and surely never agree on
it, so it is strategically wiser to leave such questions aside, and see models of
power as conceptual tools and nothing more. The person whose mindset makes
him likely to find such reasoning attractive, will probably also be attracted to
ethical pragmatism -- a method of moral reasoning that takes nothing for
granted, no God, no eternal moral imperative. no universal moral truth -- and
who therefore invests his moral reflection in a purely reflective, rational.
coherentist theory. Central theoretical ‘tools’ of ethical pragmatism are
integrity-preserving compromise‘ (Benjamin, p.7) and ‘wide reflective
equilibrium’ (Beauchamp and Childress, p.20-8; Benjamin and Curtis, p.39-44:
Clarke, p.241-4; Nielsen, 316-32; Rawls, p.46-54). Theories of ethics whose
conclusions are based on rational reflection are exposed to a radical critique
from social sciences, one which questions the solidity of ideas about rationality

and reason:

A deconstructivist critique of power

So far I have avoided including any views of power arising from
postmodern thinking. This is partly because of my own lack of famfliarity with
such thinking, and partly because nothing seems to have been formulated
within the postmodern tradition which aspires to do the job that moral
philosophy does today (help people deal with what they perceive as moral
dilemmas). It seems that at least one version of postmodern thinking, one that

radically rejects any authoritative statements, will lead to the same point as

49

radical structuralism: moral apathy. Such a position cannot contribute in any
constructive way to a discussion about the content of moral philosophy.

I will briefly indicate what a post-structural approach to the role of
power in medical ethics and medical practice could look like. Central in Michel
Foucault’s writings is a critical assessment of power and knowledge,

formulated in terms of their disciplining potential.

Power must be analyzed as something which circulates, or rather as
something which only functions in the form of a chain. It is never
located here or there, never in anybody’s hands, never appropriated as a
commodity or a piece of wealth. Power is employed and exercised
through a net-like organization. And not only do individuals circulate
between its threads; they are always in the position of simultaneously
undergoing and exercising power... ‘Truth’ is linked in a circular relation
with systems of power which produce and sustain it. and to effects of
power which it induces and which extend it. (Foucault. p.988: 133)
Accepted as a methodological starting point, Foucault provides one perspective
from which to evaluate moral philosophy. without necessarily having to reject
the practical purpose of moral phflosophy. The method would consist of seeing
moral philosophy as a type of expert knowledge. Among the names
philosophers use to describe views they are strongly opposed to are “dogmatic,
biased. arbitrary. or irrational views“ (Beauchamp and Childress, p.31). If the
dogmatist wants to engage in a dialogue with the philosopher, he will be
expected to respect some basic rules, often formulated in terms of reason,
rationality, and coherence. But the very question of what amounts to a
reasonable, rational or coherent argument is by philosophers treated as a
philosophical question, a question to be dealt with by philosophical expert
knowledge, briefly, the knowledge they themselves possess. From a

Foucaultian perspective, we could exchange the capital “1" in the Truth about

50

rationality with a small ‘t’ -- regarding the current truth of reason and rationality
as a constructed truth. This gives us the possibility of evaluating the historical
process which lead to what is today deemed phflosophically convincing and
coherent thinking (what Foucault might have called the genealogy of moral
philosophy). This perspective would also permit for a self-critical appraisal of
the practice of moral philosophy by posing the question: what are the
mechanisms by which the experts (the possessors of knowledge) as well as the
novices (the students of knowledge) are disciplined by the power of moral
knowledge.

Applied to the field of medical ethics, this perspective would lead us to
focus on forces (for example economical and political) that influence the theory
as well as the practice of medical ethics from the outside. Applied to medical
practice, it would lead us to see physicians as the possessors of medical ‘truth’
-- a knowledge which generates power, and is generated by power (the
circulating relationship between power, knowledge, and truth). Through their
training, physicians have subjected themselves to the power of medical
knowledge, and through their practice, they function as the disciplined and
disciplining instruments of this knowledge. I will not pursue this line of
thinking in this paper.

5: POWER AND ETHICS IN THE CLINICAL SETTING

The questions asked about power in social sciences are descriptive:
What is the role of power in doctor-patient interaction?; How does the
physician exercise power? The questions we could ask about power from the
perspective of moral philosophy are normative: What role should power play in
doctor-patient interaction?; When is it morally justifiable for the physician to
use power?

Brody argues that the term power was excluded fi'om the field of
medical ethics in the late sixties, as a strategic decision by philosophers and
theologians. By insisting on a strictly normative terminology, avoiding concepts
connected with behavioral and social scientists, the pioneers of a new
movement of medical ethics sought acceptance within clinical medicine (Brody

1992, p.36-9).

My central argument is that medical ethics is about power and its
responsible uses... I wish to show that with power put back into the
vocabulary, a few puzzling issues in medical ethics begin to make more
sense and the enterprise as a whole starts to hang together in a more
enlightening way. (Brody 1992, p.12)

In chapter two we saw that hidden behind the word power are several
concepts and models, many of which are mutually exclusive. The structural
view of power, for example, is not something we could introduce to the
vocabulary of medical ethics, as it denies the very ground upon which moral
phiIOSOphy is based. the concept of individual agency. In this chapter I will
argue that the concept of power which Brody brings back to medical ethics is

based on what I have referred to as theoretical pragmatism, with an emphasis

51

52

on reasoning founded in the agency view of power. As Brody claims, this helps
to resolve several puzzling issues -- but not all. We come to suspect there are
problems with his concept of power when it leads to what he calls “an apparent
contradiction,” and that the physician’s healing power forms a synthesis of
power and powerlessness. the “possibility that the physician’s power to heal in
some way depends on his vulnerability - that. ironically, powerlessness can
empower...” (Brody 1992, p.260-5). I will demonstrate how the practice
approaches view of power can resolve these theoretical problems. But first I
want to discuss the role of power for an ethical appraisal of medical practice in

some more general terms:

Use of power and moral accountability

It is obvious, but worthy of noting, that a theory of power alone tells us
nothing about what physicians and patients ought to do in their interaction.
Our insights about power will not serve as an alternative to ethical theory, or
as an affix to a list of moral guidelines in doctor-patient interaction. Rather, the
power analysis may give us a richer understanding of doctor-patient
encounters, and thereby a capacity to evaluate ethical aspects of clinical
encounters in greater detail.

When asking normative questions about power, we may either focus on
‘power relationships,’ in which case we would be interested in the dynamic,
interactive process of medical encounters: or we could focus on the ‘uses of
power.’ in which case we would be more interested in causation and effects:
the outcomes of medical encounters and what causes them. Obviously, these
two approaches are interrelated. In chapter three I dedicated most space to

discussing the ‘uses of power,’ and that is the question I will start with here.

53

To use power means to activate resources located in a structurated relationship
in order to bring about an outcome.

If we want to evaluate doctor-patient encounters in terms of outcome.
we need to decide what should count as a relevant outcome. A number of
studies within linguistics and anthropology measure the quality of doctor-
patient communication in terms of patient satisfaction (Greene et a1). One
illustrative conclusion follows these lines: Female physicians are more
egalitarian in their approach to patients, thus allowing patients more control in
the encounter; this gives higher patient satisfaction and. it is suggested, higher
patient compliance and better patient health (Meetuwesen et al. p. l 148).
According to such reasoning. the egalitarian doctor is the good doctor; the
physician's use of power and control should be minimized. This reasoning has
several flaws. First, like Maseide has already pointed out. use of power is a
necessary aspect of competent medical practice. A physician who exercised no
power would be seen as (and indeed be) incompetent -- if she did not use her
authority to influence the patient in any way, there would be no reason for
patients to see her.

Some patients come to the physician “armed with consumer reports and
prescription-drug ads, demanding inappropriate treatment” (Wyman, p. 30).
The physician might try to reeducate such a patient, by presenting the reasons
why taking these drugs is not a good idea. If the patient insists on getting his
will, the physician ought to exercise her power, the power to withhold a
treatment that is being asked for. The patient will not be content. This point
illustrates the second flaw in reasoning which holds patient satisfaction in itself
to be an outcome which indicates the quality of clinical encounters: a satisfied

patient is not always a proof of good medicine. Dr. Knock, the main character

54

in Jules Romain’s comedy, provides an example of this: rarely have we seen a
physician more successful in manipulating patients to benefit his personal ends
(mainly, economic profit); and rarely have we heard of patients more content
than those in the village of St. Maurice, where Dr. Knock has his practice.
Sometimes. then, the use of power is morally permissible, as well as
demanded by the standards of competent medical practice. Other times it is
morally wrong for the physician to exercise power. Instances of sexual
exploitation are among the most blatant examples of the latter. When we hear
of such an instance, we naturally suspect that this is a case where a powerful
person exploits the thrust of a relatively powerless person. Still, physicians
who engage in such behavior will sometimes refer to the episode as them
having been seduced, that it was an instance of true love, or that the sex was
something they offered to the patient. as part of the therapy (Kluft. p.473-4).
Sex in the physician-patient relationship provides an illustration of the
role a power analysis may play for the ethical evaluation of medical practice.
The example is not far-fetched: up to 20% of practicing physicians have been
reported to condone sexual contact between doctors and patients (Fisher, p.5).
The intuitionist may state that sex between doctor and patient is wrong in and
of itself. and that any sensible and moral person would feel likewise. The
proponent of rule-ethics may say that sex in the clinical situation is wrong
because this is stated in the ethical guidelines of the physicians’ professional
code. None of these replies are very convincing. Intuitions vary, between
different persons, and also within a person over time. The goal of moral
philosophy is to provide a more solid ground for our moral judgments than
mere intuition. Referring to guidelines is also a shallow form of argumentation.

as guidelines may, and often are. changed over time -- and often because a

55

new consensus is reached that part of the old guidelines were wrong.

If pushed to give a more informed reason why sex between physicians
and patients is unethical, many will answer that this is so. because the
physician is taking advantage of his position. He is exploiting the patient, who
is in a vulnerable position. We see that there are some underlying ideas about
an unequal power relationship. and unethical uses of power. Maybe
assumptions about the role of power, then, are probable and satisfactory
underlying reasons for the moral intuition about the wrongness of sex between
physician and patient, as well as for the ethical guidelines? The general
features of this situation are as follows: A and B are in a professional
relationship which gives A power over B: A then uses this professional power
to derive a personal benefit from B, which B would not have given A were they
not in this unequal power relationship.

If this is at the core of what makes sexual relationships between
physicians and patients wrong. do we then have a good reason to deem all
such instances morally objectionable? Maybe not. Consider the following: The
physician is a single female who recently graduated from medical school. The
director of the hospital where she works as a resident, a middle -aged married
man, comes to her office one day for a minor complaint, and the two
subsequently develop a sexual relationship. The physician falls in love with her
patient, and thinks that he will leave his wife and two daughters for her sake.
She does not yet know that her patient is famous at the hospital for using his
position as a director to seduCe female nurses and physicians.

This example is chosen to challenge the standard view that sexual
relationships between doctors and patients can always be described as

instances where the physician exploits his or her patients. It may not represent

56

the typical situation. but it reminds us that it is not by the virtue of being a
physician alone that most physicians are perceived as powerful. We are back
to Giddens’ point: that a power relationship arises as the product of an
interactive process where the two partners negotiate positions. Their power
relationship depends on how much legitimacy each of the two grants to the
structural resources possessed by the other. Structural resources are of two
kinds: allocative or authoritative. In the case where the director has sex with a
female resident who works at his hospital, the physician has virtually no
structural resources on which her patient depends -- she has no authority over
him, she has access to no material goods which he needs. and consequently,
she has no power over him. The physician is relatively powerless. and the
episode cannot be described as arising from the physician's use of power.
Rather it seems that it may be the other way around, that the patient in this
instance possessed structural resources which gave him a power advantage
over the physician. and that he used this power to manipulate the physician to
get what he wanted.

We have been reminded that we cannot take for granted a power
relationship which gives the physician an advantage over her patients, rather
we must evaluate each case separately if we want to make definite conclusions
about when the outcome of the clinical encounter can be described as the
efiect of the physician's use of power. Such an evaluation is relevant for an
ethical analysis: If the physician has little power in a specific relationship, like
in the case described above, we are less willing to hold her morally accountable
for the outcome of that relationship. If she is in a powerful position, she may
use her power to bring about effects -- she will then be morally accountable for

these effects. If she is in a powerful position and does not bring about effects.

57

she may still be held morally accountable. This is the case if a psychiatrist
refrains from using her power to retain a clearly psychotic patient in the
hospital. If the patient leaves the hospital and then kills somebody, the
physician may be held morally accountable, for in this case she ought to have
used her power.

Still, an analysis of the power relationship does not tell us all we need to
know about whom may cause effects, and whom may not. Giddens has
reminded us that there are always some resources by which the powerless
may influence the activities of the powerful in a relationship. The very powerful
physician may be manipulated by a very powerless, but cunning, patient. We
would then say that the patient, not the physician, is morally responsible for
the outcome of that interaction whereby he manipulated the physician to create
a certain effect.

Where did this take us? I have not tried to demonstrate that any classes
of use of power, in themselves. are ethical or unethical. Neither have I
described outcomes of uses of power as ethical or unethical. What I have
demonstrated is the complexity of the power relationship in doctor-patient
interaction. In particular. I have shown that we cannot take for granted that
the physician in and of herself is the powerful part in this relationship. This
observation is missed by traditional medical ethics, which starts with the
assumption that the physician is powerful -- the power of the physician is
presented as a static entity. The literature on paternalism vs. autonomy has.
for example, been formulated to make it seem as if patient autonomy is
something the physician ought to ‘give’ to the patient. The physician ‘has’
power; the paternalistic physician uses this power based on his best

judgments; the less paternalistic physician ‘gives’ some of this power or

58

authority to the patient, who thereby becomes more autonomous. Such
thinking sees power as a resource in itself, like a commodity which can be
used by individuals in the same fashion as technical instruments. and also as
something which can be ‘given’ from the one who has a lot of it to the one who
has less of it.

Following Giddens’ approach, we would rather say that the physician’s
power is ‘given[ to her, by the patient who invests trust and legitimacy in the
physician’s resources, which thereby becomes a source of power. hi this way,
the physician does not have any power at her disposal which she might ‘give’ to
the patient, like a present. Still. to say that power is ‘given’ from the patient to
the physician is imprecise, because again, it gives the impression that power is
a commodity, which can be passed between individuals at their will, like money
or pieces of information. To be stringent, and keep our reasoning precise. it
would be better to refrain fiom using terms like ‘giving’ and ‘using’ power, and
instead say ‘activating' power. If we continue to talk about ‘the uses of power,’
we should remember that power arises from structural resources. which
always depend on the legitimacy granted to them, and that ‘to use power’ in
this way is fundamentally different from, for example, ‘the use of physical
force.’ The traditional approach in medical ethics, which takes the physician’s
power as a static entity, also fails to see that in certain cases the patient may
be exercising power over the physician.

If this point is missed, we also miss the possibility to understand why it
would be wrong to say that the young resident physician who engaged in a
sexual relationship with her patient used power over him. A power analysis is
thus central for an evaluation of the mechanisms of causation in interpersonal

behavior. Empirically, as well as morally speaking, the outcome of this

59

encounter should not be causally attributed to the actions of the physician, in
the same way it would have been in a case where a physician, from her
position of power, used her authority to manipulate the patient for her own
end. I will not enter into a discussion about the extent to which the young
resident should be relieved of moral accountability in this specific case, for
there are surely a number of other moral concerns one might want to consider.
The point 1 have established is that a correct power analysis here yields
morally relevant information.

This is a small point. For, in most cases the physician is in fact the
powerful part. Giddens’ approach has implications for those ‘normal’ situations
as well, mainly for the way physicians should think about their professional

role. I will return to this point in the next and concluding paragraph.

Practice approach view of power: a contradiction resolved

Brody's approach in The Healer’s Power is to start out with what he
takes to be the accumulated wisdom from the field of medical ethics, and then
refine and reformulate this in terms of power. He uses a reformulated version
of Charles Fried’s list of the basic conditions for an ethically sound doctor-
patient relationship as a starting point for his discussion of the physician’s
power. Whereas Fried had stated these principles in terms of patient rights.

Brody formulates them in terms of physician obligations:

Lucidity: The physician has the obligation to disclose the nature of the
patient’s illness as well as the nature of any proposed treatment, along
with any viable alternative treatments.

Autonomy: The physician has the obligation to allow the patient to make
important medical-care decisions on his own behalf and to defer to the
patient’s competent wishes even if he feels that the course chosen may
not be in the patient’s best interest.

60

Fidelity: The physician has an obligation to act in a trustworthy fiduciary
manner and to view himself as the patient’s agent in health-care matters.
Humanity: The physician has an obligation to treat the patient with
compassion and sensitivity, especially bearing in mind the increased
emotional vulnerability brought about by illness and fear of death.
(Brody 1992, p.45)

Brody uses these principles to develop guidelines for the responsible
uses of power: Fidelity, he claims, requires that the physician use her power in
the best interests of the patient (aimed power): humanity urges the physician
to consider the relative powerlessness of the suffering patient; autonomy
requires that the physician allow the patient to have an impact on the decision-
making process (shared power): lucidity requires the physician to be
accountable for the ways she uses her power (owned power): “We can have the
highest degree of confidence that the healer’s power is being used ethically and
responsibly when that power can be described as owned power, shared power,
and aimed power” (Brody 1992. p.43).

Brody classifies the physician’s power in three types: aesculapian,
charismatic, and social (Brody 1992. p. 16-7). ‘Aescuiapian power‘ is founded in
the theoretical and practical knowledge of the physician: she possesses
insights in medical knowledge, she masters the skills for examining and
manipulating the body, and through her experience she has insights in the
psychological processes which are involved in the patient’s experience of
sufi’ering. Aesculapian power is an impersonal power, transferable from one
physician to another, and also a type of power that the physician ought to
share with her patients, according to Brody. ‘Charismatic power' arises from

personal characteristics of the physician (courage, decisiveness, firmness, and

so forth), and cannot be transferred form one person to another. ‘Social power'

61

arises from the social status and authority that society generally accords to
physicians.

With this theoretical starting point, Brody evaluates and discusses a
number of clinical situations. The terminology he uses represents what I have
referred to as a pragmatic view of power, with an emphasis on the agency
view. He talks about the “ability to heal; the power to relieve suffering; and
healing. . through the exercise of social and charismatic power“ (Brody 1992,
p. 19:20:35, my emphasis). Still, it would be unfair to suggest that Brody wants
to advocate a pompous image of the physician’s capacities. When he talks
about the physician’s vulnerability, and need for ‘sharing power,’ he is. on the
contrary, urging physicians not to ‘play God,’ and reminding them that they are
working within a close relationship, rather than as lonely soldiers combatting
disease and decay. Certain passages in his book also explicitly convey an
appreciation of the insights contained in the practice approaches: “A skilled
healer can facilitate all these changes...“ (Brody 1992, p.34). In my critique of
Brody I will discuss passages from his book which rely on the agency
approach, in order to demonstrate the advantages of the practice approaches
for a discussion of the physician’s power.

Let us focus on the aesculapian power. In Brody‘s view, power is
founded on theoretical knowledge and mastering of practical skills. The one
who has the insights and masters the skills of a physician, is by virtue of this a
powerful person, in and of herself. She ‘has’ aesculapian power. To use this
power responsibly, she needs to recognize that she possesses it (owned
power); to direct it so that it benefits her patients (aimed power); and allow the
patient insight and understanding in how she uses her aesculapian power

(shared power).

62

How would this be regarded differently from a practice approach view of
power? Giddens introduces one extra step in the process by which medical
knowledge and practical skills are transformed into power. For Brody. there is
a direct connection between the sources of power and the reality of power; for
Giddens, the sources of power turns into a power reality through a legitimizing
process, by which they are ‘upgraded’ to structural resources (allocative or
authoritative). This extra step explains why possessing medical insights makes
the physician powerful in certain contextualities (like in the typical clinical
encounter): and not in other settings (for example a clinical setting where the
patient does not need any of the material goods over which the physician has
control, and does not experience any structural pressure to succumb to the
authority of the physician -- like in the case of a sexual relationship between
the physician and patient described above).

I do not wish to challenge the wisdom contained in Brody’s advice (that
power ought to be owned, aimed and shared), for if physicians strived to
respect them, they would surely then be striving to act in a way that could
improve their practical interaction with patients. What I want to challenge is
the way this wisdom is formulated, because of the message about the nature of
the physicians’ power which comes along with such formulations. To say that
the physician ought to be aware of the power she possesses by virtue of being
a physician, and that she. for ethical reasons, ought not to misuse this power,
is similar to telling a regent that surely, as a regent he is powerful, but ethical
considerations should compel him not to misuse the power he has over his
people. If the message moves the regent to act responsibly and caringly, it is
good in that restricted sense: but if the regent takes the message to mean that

he is powerful in and of himself. as some kind of natural law, God-given fact.

63

or by virtue of the special qualities he possesses, then the message might not
be all that good. For, it would lead the regent to think that his goodness
towards the people is some kind of charity, some generous act, for which his
people ought to be thankful; and it would lead him to forget that if he is a
regent, he is so because the people allow him to be a regent.

There are few professions which surpass medicine when it comes to
communicating to its students as well as practitioners a message of slight
superiority over the average person. Consider for example the following
paragraph in a letter to students graduating from a medical school: “May your
tremendous knowledge. your strength of character, and your care and
compassion serve you ~- and your patients -- well.”8 It is good for patients as
well as physicians that physicians feel competent about their professional
skills. But when physicians congratulate each other about their outstanding
human qualities, they may reinforce personal traits that lead patients to feel
intimidated in their presence. Talk about the physician’s power may serve the
purpose of unwarrantly boosting physicians’ ideas about what role they play in
the healing process. This is, the way I see it, the most important reason why
we need to find a way to talk about the physician’s power without
communicating a wrong message about what the nature of this power is.
Referring to an article by Hannah Arendt, Ji‘irgen Habermas puts it this way:
“This is the impotence of the powerful -- they have to borrow their power fi'om
the producers of power” (Habermas, p.87).

 

8 Letter to the graduating medical students from the Dean of the College of Human
Medicine. Michigan State University, 29 April 1996.

64

Power corresponds to the human ability not just to act. but to act in
concert. Power is never the property of an individual; it belongs to a
group and remains in existence only so long as the group keeps
together. When we say of somebody that he is ‘in power’ we actually
refer to his being empowered by a certain number of people to act in

their name. (Arendt, p.64)

Finally, I will turn to a theoretically confusing conclusion Brody reaches
based on his pragmatic use of the term power: that the healer’s power consists
in the apparently contradicting synthesis of power and powerlessness.
According to the practice approach, the physician's capacity for initiating
healing, and his feeling of ‘powerlessness’ are different entities, and neither of
them is ‘power’ in the strict sense. Thus, the apparent contradiction is not a
real one, but merely a product of an imprecise use of the concept of power.
Brody discusses which of the physician’s characteristics are important for
engaging in a healing relationship with her patients, and concludes that her
humility might be one of her most “empowering” attributes (Brody 1992,
p.260). Healing takes place when the patient tells his story of confusion, pain,
suffering, and fear, and sees that the physician recognizes the patient as the
suffering being he is. For this to take place, the physician must enter into an
intimate relationship with her patient, one where she experiences, as well as
conveys to the patient “a genuine openness and vulnerability" (Brody 1992,

p.258).

To be compassionate in response to the suffering of the patient is
therefore one of the most powerful things a physician can do; but this is
possible only to the extent that the physician is willing to adopt a
position of relative powerlessness, to acknowledge that the patient’s
suffering has incredible power over him and that he cannot remain
unchanged in the face of it. This is a major irony of the physician-patient
relationship, in which a sense both of one’s healing power and of one‘s
necessary humility forms a synthesis of the apparent contradiction of
power and powerlessness. (Brody 1992, p.260)

65

With the practice approaches view of power, the apparent contradiction
as well as the irony of Brody’s description disappears. In this paragraph,
power is described both as a characteristic of a particular action (of being
compassionate); the characteristics of positions in a social relationship one may
chose to occupy or to not occupy (a position of less control); the capacity a
certain reality has of inducing emotional responses; and a capacity of
facilitating healing (the healer’s power). In line with Giddens’ structuration
theory, I would not like to refer to any of these as ‘power‘ in the strict sense. I
do not think the healer is powerful by herself, rather I think that healing is the
fortunate outcome of certain relationships that physicians and patients may
engage in. Healing is the product of a process where the physician invests
some of her resources, and the patient invests some of his resources. Neither
of the two have the power within themselves to heal, and neither of the two
should think of themselves as a healer, in the sense of one who heals, simply
by using her own skills and capacities. The last quoted paragraph from Brody‘s
book could be rephrased in the following manner, which makes it clear what
role the power plays in a healing relationship:

The physician who allows the patient's suffering to have an impact on
her becomes vulnerable. In a sense, she will experience some of the patient's
own fear and suffering. By suffering with the patient in this way, she engages
in an intimate relationship with him. In stead of retaining a ‘professional
distance,’ she might decide to expose some of her emotional reactions to her
patient. This does not necessarily mean to act unprofessionally, to be in an
intimidating position, or to feel intimidated. Rather, it may be an efficient and

professional approach of establishing a healing relationship with suffering

66

patients. This approach will induce a feeling of trust and security within the
patient. A physician who successfully engages in such relationships activates
the potential for a healing power which lies in the structure of the doctor-

patient relationship.

REFERENCES

Arendt, H. "Communicative Power." In: Power. Lukes, S. ed. New York: New
York University Press, 1986:59-74.

Beauchamp, T.L., and Childress, J .F. "Morality and Moral Justification." In:
Principles of Biomedical Ethics . New York, Oxford: Oxford University
Press, 1994:3-43.

Benjamin, M., and Curtis, J. In: Ethics in Nursing . New York: Oxford
University Press, 1986:39-44.

Benjamin, M. Splitting the Difference. University Press of Kansas, 1990.

Best, 8., and Kellner, D. "Foucault and the Critique of Modernity." In:
Postmodern Theory. New York: Guilford Press, 1991:34-75.

Bourdieu, P. Outline of a Theory of Practice. Cambridge University Press, 1977.

Brody, H. The Healer's Power. New Haven: Yale University Press. 1992.

Brody, H. "The Placebo Response," Drug Therapy. 1986: 106-131.

Brody, H. "The Symbolic Power of the Modern Personal Physician: the Placebo
Response under Challenge." Journal of Drig Issues. 18(2) 1988:
149-161.

Clarke, S.G. "Anti-theory in Ethics," American Philosophical Quarterly. 24(3)
1987: 237-244.

Dahl, R.A. "Power." In: International Encyclopedia of the Social Sciences. Sills,
D.L. ed. The Macmillan Company 8: THe Free Press, 1968:405-415.

Fisher, N., and Fahy, T. ”The Eroticised Consultation: Practical and Ethical
Considerations." Comprehensive Therapy. 18(5) 1992: 2-5.

Foucault, M. In: Power/ Knowledge. Selected Interviews 82. Other Writings
1972-1977. Collin G, ed. New York: Pantheon Books, 1980:

Giddens, A. The Constitution of Society. Berkeley and Los Angeles: University
of California Press, 1984.

Greene, M.G.. Adelman, R.D., Friedman, E., and Charon, R. "Older Patient
Satisfaction with Communication During an Initial Medical Encounter."
Social Science and Medicine. 38(9) 1994: 1279-1288.

Habermas, J. "Hannah Arendt's Communications Concept of Power." In: Power.
Lukes, S. ed. New York: New York University Press, 1986:75-93.

67

68

Hall, S. "The Toad in the Garden." In: Marxism and the Interpretation of Culture.
Grossberg, L., and Nelson, C. eds. Urbana: University of Illinois,
1988:35-73.

Katz, J. The Silent World of Doctor and Patient. New York: Three Free Press,
1984.

Kleinman, A. The Illness Narratives. Basic Books. 1988.

Kluft, R.P. "The Physician as Perperator of Abuse," Primary Care. 20(2) 1993:
459-480.

Levi-Strauss, C. "Social Structures; The Story of Asdiwal." In: High Points in
Anthropology. Bohannan. P., and Glazer, M. eds. New York:
McGraw-Hill, inc., 1988:422-500.

Lukes, S. Power. A Radical View. London and Basingstoke: Macmillan Press,
1974.

Lukes, S. "Introduction." In: Power. Lukes, S. ed. New York: New York
University Press, 1986:1-18.

Maclntyre, A. After Virtue. Notre Dame, Indiana: University of Notre Dame
Press, 1981.

Maseide, P. "Possibly abusive, often benign, and always necessary. On power
and domination in medical practice." Sociology of Health 82. Illness. 13(4)
1991: 545-561.

Meetuwesen, L., Schaap, C., and van der Staak, C. "Verbal Analysis of
Doctor-Patient Communication," Social Science and Medicine. 32(10)
1991: 1143-1150.

Nielsen, K. "Relativism and wide reflective equilibrium." The Monist. 1993:

3 16-332. ,

Nyberg, D. Power over Power. Ithaca and London: Cornell University Press,
1981.

Ortner, S. "ls Female to Male as Nature Is to Culture?" In: Woman, Culture, and
Society. Rosaldo, M., and Lamphere, L. eds. Stanford: Stanford
University Press, 1974:67-88.

Ortner, S.B. "Theory in Anthropology since the Sixties." In:
Culture/Power/Htstory. Dirks, Eley, and Ortner. eds. 1994:372-411.

69

Pappas, G. "Some Implications for the Study of the Doctor-Patient hiteraction:
Power. Structure, and Agency in the Works of Howard Waitzkin and
Arthur Kleinman." Social Science and Medicine. 30(2) 1990: 199-204.

Philip, M. "Power." In: The Social Science Encyclopedia. Kuper, A., and Kuper.
J. eds. London, Boston and Henley: Routhledge & Kegan Paul,
1985:635-639.

Poulantzas, N. "Class Power." In: Power. Lukes, S. ed. New York: New York
University Press, 1986:144-155.

Rawls, J. A Theory of Justice. Cambridge, Massachusetts: The Belknap Press
of Harvard University Press, 1971.

Redelmeier, DA, and Shafir, E. "Medical Decision Making in Situations That
Offer Multiple Alternatives," JAMA. 273(4) 1995: 302-305.

Romains, J. Docktor Knock. A Comedy in Three Acts. New York: Samuel French,
inc, 1925.

Russell, B. "The Forms of Power." In: Power. Lukes, S. ed. New York: New
York University Press. 1986:19-27.

Street, R.L. "Information-Giving in Medical Consultations: The influence of
Patients' Communicative Styles and Personal Characteristics." Social
Science and Medicine. 32(5) 1991: 541-548.

Thomas, K.B. "General Practice Consultations: Is There Any Point in Being
Positive?," British Medical Journal 294 1987: 1200-1202.

van Biema, D. "A Shameful Death," Time. 146(24) 11 Dec 1995: 32-36.

Waitzkin, H. The Politics of Medical Encounters. How patients and doctors deal
with social problems. New Haven: Yale University Press, 1991.

Weber. M. "Domination by Economic Power and by Authority." In: Power.
Lukes, S. ed. New York: New York University Press, 1986:28-36.

Wyman, I. "Doctors Under the Knife," Newsweek. 5 Apr 1993: 28-41.

"Illlll'lllllilillllf