 

 

s1,

. .1
, ”.m.
7" .

~».,-”hu
.4.-

,.
394m!“

Lu:
‘ n

m-

‘" "’2‘ 447.14»;
{78:33 < m

.on~

 

 

 

 

 

my“ ' ,LIBRARY
ZJ ((7 Michigan State
° University

 

 

This is to certify that the
dissertation entitled

RESPECTING AUTONOMY IN CASES OF AMBIVALENCE
REGARDING END OF LIFE DECISIONS

presented by

JENNIFER SPROUL SWINDELL

has been accepted towards fulﬁllment
of the requirements for the

DOCTORAL degree in PHILOSOPHY

 

 

Major Professors ignature

7/0/53

Date

MSU is an afﬁrmative-action, equal-opportunity employer

_ _<—-.-.—---¢------o-—

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5(08 K:/Proj/Acc&Pres/ClRC/Date0ue indd

RESPECTING AUTONOMY IN CASES OF AMBIVALENCE REGARDING END OF
LIFE DECISIONS

BY

Jennifer Sproul Swindell

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Philosophy

2008

ABSTRACT

RESPECTING AUTONOMY IN CASES OF AMBIVALENCE REGARDING END OF
LIFE DECISIONS

By
Jennifer Sproul Swindell

Although personal autonomy is now discussed frequently in both theoretical and
applied philosophy, the meaning of the term often shifts; and this has gone largely un-
noted, creating much confusion. I draw a distinction between the way that the term is
used in the context of theoretical philosophy and in the context of bioethics. In the
context of theoretical philosophy, the trend is to focus is on the agent’s internal states.
This is largely due to the inﬂuence of Harry F rankurt, who has changed the direction of
the ﬁ'ee will/determinism debate by arguing that a person is autonomous when the desires
that move her to act are ones that she wants to move her to act. I label this notion
“metaphysical autonomy.”

The overall trend in discussions about autonomy in the bioethical context is to
focus on whether the agent has an adequate understanding of the world; if she does, then
she is permitted to make decisions about her medical care. I label this notion “bioethical
autonomy.” There is, however, interesting overlap between metaphysical and bioethical
autonomy. Young children, for example, are not permitted to make their own medical
decisions because they are not able to engage in the sort of reﬂection required for

metaphysical autonomy. Certain psychiatric patients, for example, are not permitted to

make their own medical decisions because the reﬂection required for metaphysical
autonomy is distorted, or the desires that are moving them to act are not ones that they
want to move them to act. But often, patients are ambivalent and cannot decide, or are
deeply conﬂicted about, which of their desires they want to move them to action.

Little to no attention has been paid to the phenomenon of ambivalence in
contemporary autonomy and moral psychology literature. This is surprising because
accounts of autonomy rely on the assumption that there is something that a person really
wants, or really values. But ambivalence calls this into question. I develop an account of
ambivalence, drawing a distinction between paralyzing ambivalence and residual
ambivalence. I argue that paralyzing ambivalence results in a loss of both metaphysical
and bioethical autonomy, whereas residual ambivalence does not, despite the fact that
several major personal autonomy theorists imply that it does. Drawing from existential
philosophy, I propose methods to resolve paralyzing ambivalence and to restore
metaphysical and bioethical autonomy; and I also respond to challenges the proposals
may face. I argue, however, that a certain amount of residual ambivalence is desirable.
With an analysis of autonomy and ambivalence in hand, I apply it to cases of
ambivalence about end of life decisions. The issue of ambivalence at the end of life is
important because it occurs ﬁequently, yet is often responded to with medical
paternalism (deciding for the patient). Understanding ambivalence and how to resolve it

promotes patient autonomy and patient well-being.

COpyrisht by
JENNIFER SPROUL SWINDELL
2008

ACKNOWLEDGMENTS

First, I would like to acknowledge my advisor, Tom Tomlinson. I honestly
couldn’t ask for a better advisor and mentor: philosophically razor sharp, encouraging
and supportive, but always challenging and pushing me philosophically. Thank you so
much, Tom. Second, I would like to acknowledge Michigan State University philosophy
department faculty and graduate students. I would like to particularly thank faculty
members Fred Rauscher and Fred Gifford, who were highly supportive of me during
my graduate career. I would also like to particularly thank my graduate student
colleagues and close friends Lisa Campo Engelstein, Meredith Verrochi, Jason Mask,
Shaman Burns Proctor, Karen Meagher, and Dom Sisti. They helped make my
environment intellectually stimulating, not to mention highly enjoyable. They have also
been my cheering section, which has been invaluable. Third, I would like to
acknowledge Katrina Bramstedt. Thank you for not only mentoring me during my time
at the Cleveland Clinic, but for mentoring me ever since. Your support and
encouragement is deeply appreciated. Fourth, I would like to thank Harry Frankfurt for
writing such engaging and thought provoking philosophy. Your work has inspired me
both professionally and personally, more than the work of any other philosopher.

Finally, I would like to thank my committee members: James Lindemann Nelson,
Christian Lotz, Tamra Frei, and external reader James Stacey Taylor for their many

helpful comments on this dissertation.

PREFACE

Although personal autonomy is now discussed frequently in both theoretical and
applied philosophy, the meaning of the term often shifts; and this has gone largely un-
noted, creating much confusion. I draw a distinction between the way that the term is
used in the context of theoretical philosophy and in the context of bioethics. In the
context of theoretical philosophy, the trend is to focus is on the agent’s internal states.
This is largely due to the inﬂuence of Harry Frankurt, who has changed the direction of
the free will/determinism debate by arguing that a person is autonomous when the desires
that move her to act are ones that she wants to move her to act. I label this notion
“metaphysical autonomy.” Of course, the term ‘want’ begs the question, and this is where
contemporary metaphysical autonomy theorists split to develop several different theories
of metaphysical autonomy. For Frankfurt, to want a desire to move me to action is to
decisively identify with it (for any reason, or for no reason at all). For Michael Bratman,
to want a desire to move me to action is view it as reason-giving in light of the normative
content of my self-governing policies. For John Christman, it is to be satisﬁed with the
process that led to the desire.

The overall trend in discussions about autonomy in the bioethical context is to
focus on whether the agent has an adequate understanding of the world; if she does, then
she is permitted to make decisions about her medical care. I label this notion “bioethical
autonomy.” There is, however, interesting overlap between metaphysical and bioethical
autonomy. Young children, for example, are not permitted to make their own medical

decisions because they are not able to engage in the sort of reﬂection required for

vi

metaphysical autonomy. Certain psychiatric patients, for example, are not permitted to
make their own medical decisions because the reﬂection required for metaphysical
autonomy is distorted, or the desires that are moving them to act are not ones that they
want to move them to act. But often, patients are ambivalent and cannot decide, or are
deeply conﬂicted about, which of their desires they want to move them to action.

Little to no attention has been paid to the phenomenon of ambivalence in
contemporary autonomy and moral psychology literature. This is surprising because
accounts of autonomy rely on the assumption that there is something that a person really
wants, or really values. But ambivalence calls this into question. I develop an account of
ambivalence, drawing a distinction between paralyzing ambivalence and residual
ambivalence. I argue that paralyzing ambivalence results in a loss of both metaphysical
and bioethical autonomy, whereas residual ambivalence does not, despite the fact that
several major personal autonomy theorists imply that it does. Drawing ﬁ'om existential
philosophy, I propose methods to resolve paralyzing ambivalence and to restore
metaphysical and bioethical autonomy; and I also respond to challenges the proposals
may face. I argue, however, that a certain amount of residual ambivalence is desirable.
With an analysis of autonomy and ambivalence in hand, I apply it to cases of
ambivalence about end of life decisions. The issue of ambivalence at the end of life is
important because it occurs ﬁ'equently, yet is often responded to with medical
paternalism (deciding for the patient). Understanding ambivalence and how to resolve it
promotes patient autonomy and patient well-being. In fact, the entire dissertation itself
was motivated by a case of ambivalence about an end of life decision: The Case of Mr.

X.

vii

The Case of Mr. X

During the winter of 2004 I spent most of my time interning at a major US
hospital in the bioethics department. Of all of the cases that I was involved in, one in
particular stands out. Mr. X, a youthful 70 year old man, was involved in a head on
motor vehicle accident. As a result, he lay in a hospital bed unconscious, on a ventilator,
and paralyzed from the neck down. Mr. X’s wife and adult son and daughter all agreed
that there was no way that Mr. X would want to live this way. In fact, they pointed to
Mr. X’s advance directive to afﬁrm this View. Shortly thereafter, Mr. X regained
consciousness. To everyone’s surprise, Mr. X asserted that he wanted to remain on the
ventilator; he wanted to stay alive. The following day, we went in to talk with Mr. X and
he asserted that he did not want to remain on the ventilator; he wanted to die. The family,
the physicians, and the ethicists talked to Mr. X to try to ascertain what he really wanted.
These conversations spanned over weeks, but Mr. X did not know what he really wanted.

He was ambivalent.

viii

. .-

TABLE OF CONTENTS

CHAPTER ONE
METAPHYSICAL AND BIOETHICAL AUTONOMY .................................................... 1
Usage of the term “autonomy” ....................................................................................... 1
Autonomy as a metaphysical concept ............................................................................. 7
Autonomy as identiﬁcation ........................................................................................ 7
Identiﬁcation as decision and satisfaction ............................................................ 7
Identiﬁcation as reason giving ............................................................................ 10
Historical identiﬁcation ...................................................................................... 12
Identiﬁcation as non-alienation .......................................................................... 14
Autonomy as coherence ........................................................................................... 15
Substantive accounts of autonomy .......................................................................... l6
Autonomy as a bioethical concept ................................................................................ 18
Autonomy as decision making capacity .................................................................. 18
The connection between metaphysical autonomy and bioethical autonomy ................ 20
Children ................................................................................................................... 21
Psychiatric patients .................................................................................................. 22
Inﬂuence on the criteria ........................................................................................... 24

CHAPTER TWO
A PHILOSOPHICAL ANALYSIS OF AMBIVALENCE ............................................... 26
First person accounts of ambivalence ........................................................................... 26
A conceptual analysis of ambivalence .......................................................................... 28
Locating ambivalence .............................................................................................. 28
Paralyzing ambivalence ........................................................................................... 33
Residual ambivalence ............................................................................................ '..34
Causes of ambivalence .................................................................................................. 38
Ambivalence and the importance of the decision ......................................................... 42
Ambivalence and information ...................................................................................... 43
Ambivalence and action ................................................................................................ 46
Indifference and temptation .......................................................................................... 48

CHAPTER THREE

THE COMPATIBILITY OF AUTONOMY AND AMBIVALENCE ............................. 51
Paralyzing ambivalence and metaphysical autonomy .................................................. 51
Paralyzing ambivalence and bioethical autonomy ........................................................ 55
Residual ambivalence and metaphysical autonomy ..................................................... 57
Residual ambivalence and bioethical autonomy ........................................................... 62

ix

CHAPTER FOUR

RESOLVING PARALYZING AMBIVALENCE ............................................................ 64
Provide basic information and information about coherence ....................................... 65
Assess coherence and invite a leap ............................................................................... 70
Change the world .......................................................................................................... 80
Resolve for the person .................................................................................................. 81
Not choose .................................................................................................................... 82
Commit suicide ............................................................................................................. 85
First order desires take over .......................................................................................... 86

CHAPTER FIVE

RESOLVING RESIDUAL AMBIVALENCE .................................................................. 88
Drawbacks of residual ambivalence ............................................................................. 88
Beneﬁts of residual ambivalence .................................................................................. 93
Resolving residual ambivalence: types of wholeheartedness ....................................... 97

CHAPTER SIX

AMBIVALENCE AND END OF LIFE DECISIONS .................................................... 103
The context of end of life decisions ............................................................................ 103
A case of ambivalence regarding an end of life decision ........................................... 106
Resolving Mr. X’s ambivalence in ways that promote autonomy .............................. 107
Resolving Mr. X’s ambivalence in ways that do not promote autonomy ................... 111
Choosing not to choose ............................................................................................... 114
Remaining issues ........................................................................................................ 1 15
Dying with paralyzing ambivalence regarding end of life decisions .......................... 117
Dying with residual ambivalence regarding end of life decisions .............................. 119

BIBLIOGRAPHY ............................................................................................................ 122

LIST OF FIGURES

FIGURE ONE
TYPES OF AUTONOMY ................................................................................................... 5

xi

Chapter One: Metaphysical and Bioethical Autonomy

“Last night I shot an elephant in my pajamas. What he was doing in my pajamas I ’11
never know. ”—Groucho Marx

Usage of the Term “Autonomy”

The humor in this joke depends on the ambiguity of language. Unfortunately,
ambiguous language most often results not in humor, but in confusion. Such is the case
with the word ‘autonomy’. Much has been written about autonomy in philosophy, and
especially in bioethics. This is a recent phenomenon. As Tom Beauchamp points out,
the 1967 Encyclopedia of Philosophy had no indexed mention of autonomy.1 The
interest in personal autonomy came into vogue shortly thereaﬁer. Although autonomy is
now written about frequently in both theoretical and applied philosophy, the meaning of
the term often shifts; and this has gone largely un-noted and created much confusion.
The major cause of the shiﬁing meaning of the term is the different contexts/realms of
discourse in which the term is used. In this chapter I will examine the many ways in
which the term is used, focusing on the way that the term is used in the context of

theoretical philosophy (what I will call “metaphysical autonomy theories”) and in the

 

' Tom Beauchamp, “Who Deserves Autonomy, and Whose Autonomy Deserves Respect?” in Personal
Autonomy: New Essays on Personal Autonomy and Its Role in Contemporary Moral Philosophy
(Cambridge University Press 2005), p. 327 footnote 1.

2 I suspect that there will be concerns about my use of the term “metaphysical.” Let me explain why I use
the term, justify my use of the term, and clarify what I do not see myself as committed to by my use of the
term. I use the term because it indicates the context in which the autonomy theorists that I discuss see
themselves as working. Harry Frankﬁirt, the originator of the idea of autonomy as free will, and free will
as a certain relationship between desires, takes himself to be doing metaphysics. And all of the autonomy
theories that came aﬁer Frankﬁrrt are responses to Frankfurt. In defense of Frankfurt considering what he
is doing to be metaphysics, and hence in defense of my use of the term “metaphysical autonomy,” let me
say that I take metaphysics to be the study of ontology and cosmology, where ontology involves claims
about existence—claims about the fumiture of reality, so to speak (11 what sense the things in one’s
ontology exist needs to be cashed out, of course), and cosmology involves claims about how causation and
how the ontological elements interact. So, certainly Frankfurt is justiﬁed in claiming to do metaphysics, for
he is claiming that autonomy or ﬂee will exists, and discussing the sense in which it exists and the

context of bioethics (what I will call “bioethical autonomy theories”). After surveying
theories of metaphysical autonomy, I will explore the connections between metaphysical
autonomy and bioethical autonomy.

The shifting meaning of the term “autonomy” has not gone completely unnoticed.
Tom May recognizes two senses in which the term “autonomy” is used: “autonomy as
autarkeia” and “autonomy as self-rule.” Autonomy as autarkeia is basically a conception
of autonomy as self-sufﬁciency. External inﬂuences cannot be present in order for one to
be self-sufﬁcient. May draws on Aristotle for examples of self-sufﬁciency. A self-
sufﬁcient city state is independent ﬁom outside powers; it provides all that is necessary to
ﬂourish. A self—sufﬁcient ﬁ'iendship is one that is not based upon need; the ﬁ'iend is not
there to fulﬁll some lacking. A self-sufﬁcient person (which Aristotle calls a “great
souled man”) is one who does not pursue things because she needs them. So, overall,
autonomy as autarkeia is independence ﬁ'om external inﬂuences such as a need for
utility.3

The second sense in which the term “autonomy” is used, according to May, is
autonomy as self-rule. This sense of autonomy allows for external circumstances to be
part of the picture. May again draws on Aristotle, this time for examples of self-rule.
Aristotle uses the analogy of a helsman. A helsman steers the ship, but he does so within

the context of external considerations. While external circumstances are allowed to

 

conditions for its existence—hence, I am justiﬁed in calling Frankfurt’s theory of autonomy (and the
theories that came into response to it) a metaphysical theory of autonomy. What I do not mean to do by
using the term “metaphysical autonomy” is to make any commitments to indeterminism—Frankfurt’s
theory of free will and the theories that followed are all compatible with both determinist and libertarian
accounts of the ultimate origins of our desires etc. I also do not mean to make any commitments to exactly
I’how“ the [free] will or autonomy exists. Neither Frankfurt nor his responders analyze the sense in which
these things exist—they simply acknowledge that they are part of our reality and examine the conditions
for their existence.

affect how one behaves, they cannot determine it. The autonomous agent actively
assesses the external inﬂuences and then acts.4

Manuel Vargus has also noted the shifting meaning of the term “autonomy” in the
literature. Vargus states, “Autonomy is variously characterized as: bare agency; a species
of self-governed agency; a kind of relation to the world; an ideal of self-control that may
rarely be had; a kind of rule—govemed activity that is frequently had; ownership-taking
for what one does; interchangeable with ﬁeedom; a conception of morally responsible
agency; neither ﬁ'eedom nor morally responsible agency; competence for medical
decision-making; authority over personal choices; self-rule; a designation for agents
bound by political principles governing the basic institutions of society, freedom from
external inﬂuence; freedom ﬁ'om external control or restriction on choice; and, the kind

”5 In more general

of thing for which external restrictions on choice are largely irrelevant.
terms, Vargus identiﬁes three strands of discussion in the autonomy literature: agential
autonomy, individual political autonomy, and minimal medical autonomy. Agential
autonomy discussions focus on notions of autonomy tied to the work of Harry Frankfurt
(which will be discussed later in this chapter), individual political autonomy discussions
are tied to social and political aims, and minimal medical autonomy discussions focus on
competency and medical decision-making.6

Joel F einberg has also noted the shiﬁing sense of the term “autonomy” in the

literature. He identiﬁes four meanings of the term “autonomy”: the capacity to govern

 

3 Thomas May, “The Concept of Autonomy” in American Philosophical Quarterly Vol. 31 No. 2 April
1994,p.135.

’ May, “The Concept of Autonomy,” pp. 139-141.

5 Manuel Vargus, “Review of James Stacey Taylor (ed.), Personal Autonomy: New Essays on Personal
Autonomy and Its Role in Contemporary Moral Philosophy (Cambridge University Press 2005)” in Notre
Dame Philosophical Reviews, 2004.

oneself and make one’s own decisions, the actual condition of self-govemment, an ideal
of character, and the right to govern oneself.7

Also noting the shifting meaning of the term “autonomy” in the literature is Nomy
Arpaly. Arpaly discusses eight senses in which the term “autonomy” is used: agent
autonomy concerns the relationship that an agent has to her motivational states,
autonomy as personal efﬁcacy concerns physical independence and not relying on others,
autonomy as independence of mind concerns not blindly accepting the views of others,
normative autonomy concerns one’s moral right to have her decisions respected,
autonomy as authenticity concerns the absence of external desires and values, heroic
autonomy concerns an ideal condition that a great majority of persons do not have, and
autonomy as acting rationally concerns the ability to respond to reasons.8 In other work,
Arpaly reduces the eight senses in which the term “autonomy” is used into two basic
categories: complex autonomy theories and autonomy theories in applied ethics.
Complex autonomy theories construe autonomy in terms of hierarchies of mental states,
non-alienation, a subjective sense of activity, absence of mental conﬂict, or
wholeheartedness.9 Autonomy theories in applied ethics construe autonomy in terms of
decision-making capacity.

Clearly, the different senses of autonomy are now many and varied. A USCfUI

taxonomy for the various uses of the term “autonomy” is the following:

 

6 Vargus, “Review of James Stacey Taylor (ed.), Personal Autonomy: New Essays on Personal Autonomy
and Its Role in Contemporary Moral Philosophy (Cambridge University Press 2005).”

7 Joel Feinberg, Harm to Self (Oxford University Press, 1986).

8 Nomy Arpaly, Unprincipled Virtue: An Inquiry Into Moral Agency (Oxford University Press, 2003), pp.
1 18-125.

Figgre 1: Types of Autonomy

 

 

Types: Personal Autonomy Kantian Autonomy
Contexts: Social/Political Metaphysics Bioethics
Strands
of Discussion: Autonomy as

Decision Making Capacity
Autonomy as Identiﬁcation Autonomy as Coherence Substantive Accounts

Decision and Reason-Giving Historical Non-Alienation
Satisfaction

The initial biﬁrrcation is between personal autonomy and Kantian autonomy. A brief
explication of Kantian autonomy will be helpful so that readers can get a sense of what
contemporary autonomy theory (“personal autonomy” theory) is not. Kant held that
autonomy is a property of the will that all adult human beings have, although only some
of us exercise it.10 The will is autonomous when it is self-determining and accords with a
law.11 In order for the will to be self-determining and at the same time accord with a law,
the will must have its own law.12 The categorical imperative is the law of the

autonomous will, and this is the case simply because it was designed this way by

 

9 Nomy Arpaly, “Responsibility, Applied Ethics, and Complex Autonomy Theories” in Personal
Autonomy: New Essays on Personal Autonomy and Its Role in Contemporary Moral Philosophy edited by
James Stacey Taylor (Cambridge University Press 2005), p. 174.

'0 Thomas Hill Jr., “The Kantian Conception of Autonomy” in Dignity and Practical Reason In Kant ’s
Moral Theory edited by Thomas Hill Jr. (Cornell University Press 1992), p. 85.

” Christine Korsgaard, The Sources of Normativity (Cambridge University Press, 1996), p. 98.

‘2 Korsgaard, The Sources of Normativity, p. 98.

nature. '3 The categorical imperative tells us to act only on a maxim which we could will
to be a [universal] law. When following the categorical imperative, our will is
autonomous and we are acting autonomously, because we are acting on a law that came
from our own will. And for Kant, everything outside of the will—including our desires
and inclinations—counts as alien causes, and as interferences with autonomy. '4
Contemporary personal autonomy theories, on the other hand, view certain types of
desires and inclinations as essential elements of autonomous action.

The difference between Kantian autonomy and personal autonomy is explained by
a difference in purpose. While Kant developed an account of autonomy for the purpose
of developing an account of right action, contemporary personal autonomy theorists have
developed accounts of autonomy for the purpose of developing accounts of ﬁ’ee action.
But free action is important to theorists for different reasons, depending on the context in
which they are discussing free action. In the social political context, ﬁ'ee action is often
argued for as a socio-political right, and theorists spend time constructing ways to
structure society so that free action (as opposed to forced or coerced actions) is
maximized. In the bioethical context, free action is similarly argued for as a patient
right, and theorists spend time constructing tools for evaluating and enhancing decision
making capacity and informed consent so that ﬂee action is maximized. In the
metaphysical context, ﬁ'ee action is presented as an ideal, as something that persons
should strive for. Ideally, persons should act freely; they should reﬂect on their desires
and strive to act on desires which are wholeheartedly theirs, and resist acting on desires

that are not. I will say much more about the connection between ﬁ'ee action and

 

’3 G.A. Cohen, “Reason, Humanity, and the Moral Law,” in The Sources of Normativity by Christine
Korsgaard (Cambridge University Press, 1996), p. 172.

autonomy in the contexts of bioethics and metaphysics later in this chapter, after I survey
the various conceptions of each.

Before surveying the various conceptions of autonomy, I will say a word about
why free action (autonomy) is thought to be so valuable. Immanuel Kant and John Stuart
Mill are traditionally pointed to as sources for explaining the value of autonomy. Kant
argued that we should respect autonomy because it is what makes us persons. We (unlike
lower animals) are capable of reﬂecting on our desires, and then deciding whether or not
to will them and act on them.15 Mill argued that we should respect autonomy because
letting people act ﬁ'eely and choose for themselves is what is most likely to enable people
to secure their own well-being, and because exercising their own autonomy/making their
own choices (even if they later regret them, or they are not in their best interests) is part
of well-being itself. ‘6
Autonomy as a Metaphysical Concept
AUTONOMY AS IDENTIFICATION
Identification as Decision and Satisfaction

Contemporary personal autonomy theories are largely inspired by the work of
Harry Frankfurt. In his 1971 seminal paper, “Freedom of the Will and the Concept of a
Person,” Frankfurt argued that what makes us persons is the ability to reﬂect on our
desires (our ability to form “second order desires”). When we have a particular “ﬁrst
order desire” to do something (say, eat a cookie), we have the ability to reﬂect on that
desire (to form a “second order desire”). Frankqu argues that it is when our 2nd order

desires are in accord with our 1St order desires that our will is free. It is when I both want

 

'4 Korsgaard, The Sources of Normativity, p. 97.
‘5 Korsgaard, The Sources of Normativity, p. 93.

to desire the cookie, and I want my desire for the cookie to be effective in action, that my
will is ﬂee and the action of eating the cookie is free.17 It is when I identiﬁ with my ﬁrst
order desire to eat the cookie that my will is ﬂee. But, what does it mean to “identify”
with a desire? Frankfurt distinguishes between desires that a person identiﬁes with and
ones that she views as outlaws. To view a desire as outlaw is to feel like I am a bystander
to it, it disturbs me, it makes no sense to me, I’d never think of acting on it, it has no
recognizable warrant, it happens to me/enters my mind, it feels oddly disconnected ﬂom
me or even dangerously antithetical, it is an unacceptable intruder, I feel an anxious
disposition to resist it, it is outlawed and disenﬂanchised, I refuse to recognize it as
grounds for what I think and do, I treat it as categorically unacceptable and try to
suppress it or rid myself of it entirely, regardless of how insistent it may or how
powerfully moved by it I am I give it now rational claim, I am determined to give it no
position at all in the order of my preferences.l8 On the other hand, to identify with a
desire is to acknowledge that satisfying it is to be assigned some position in his
preferences. It may be that the desire doesn’t please me or make me proud, but I am
willing to have it represent me; I accept it; I make no determined effort to dissociate
myself ﬂom it; in weary resignation I consent to having it and to being inﬂuenced by it.19

Or, I could identify more strongly with the desire and regard it with welcoming approval,

 

'6 John Stuart Mill, On Liberty (bnpublishing.com, 2008).

'7 Harry Frankfurt, “Freedom of the Will and the Concept of a Person” in The Importance of What We Care
About (Cambridge University Press, 1988), p. 15.

'8 Harry Frankfurt, “Taking Ourselves Seriously” in Taking Ourselves Seriously and Getting It Right
(Stanford University Press, 2006), pp. 8-11.

'9 Harry Frankfurt, “Taking Ourselves Seriously” in Taking Ourselves Seriously and Getting It Right
(Stanford University Press, 2006), pp. 7-8.

or even feel like not having that desire would be unthinkable.20 Frankfurt further
explains, “When a person identiﬁes himself decisively with one of his ﬁrst-order desires,
this commitment “resounds” throughout the potentially endless array of higher

orders. . . .The fact that his second-order volition to be moved by this desire is a decisive
one means that there is no room for questions concerning the pertinence of volitions of
higher orders. . . .The decisiveness of the commitment he has made means that he has
decided that no further questions about his second-order volition, at any higher order,
remain to be asked.”21 In his 1987 paper, “Identiﬁcation and Wholeheartedness,”
Frankfurt attempts to clarify the nature of decisive identiﬁcation.22 He says, “The fact
that a commitment resounds endlessly is simply the fact that the commitment is decisive.
For a commitment is decisive if and only if it is made without reservation, and making a
commitment without reservation means that the person who makes it does so in the belief
that no further accurate inquiry would require him to change his mind. It is therefore
pointless to pursue the inquiry any further. This is, precisely, the resonance effect.”23
And Frankfurt remarks that this characterization of decisive identiﬁcation is inﬂuenced
by Descartes’s discussion of clear and distinct perception.24 Further revision on the
concept of identiﬁcation came in his 1992 paper, “The Faintest Passion.” Here he moves
from a notion of identiﬁcation that has to do with decisiveness to one that has to do with

wholeheartedness and satisfaction. To wholeheartedly identify with certain ﬁrst order

desires is to be “. . .fully satisﬁed that they. . .should be among the causes and

 

20 See Harry Frankfurt, “Rationality and the Unthinkable” in The Importance of What We Care About
(Cambridge University Press, 1988), pp. 177-190. Also see Harry Frankfurt, “Discussion With Harry
Frankfurt,” in Ethical Perspectives Vol. 5 No. 1 1998, pp. 26-27.

2‘ Frankfurt, “Freedom of the Will and the Concept of a Person,” pp. 21-22.

22 Harry Frankfurt, “Identiﬁcation and Wholeheartedness,” in The Importance of What We Care About
(Cambridge University Press, 1988), p. 167.

23 Frankfurt, “Identiﬁcation and Wholeheartedness,” pp. 168-169.

considerations that determine his cognitive, affective, attitudinal, and behavioral
processes.” And, “What satisfaction does entail is an absence of restlessness or
resistance. A satisﬁed person might willingly accept a change in his condition, but he has
no active interest in bringing about a change. Even if he recognizes that he could be
better off, the possibility does not engage his concern. . ..To be satisﬁed with something
does not require that a person have any particular belief about it, nor any particular
feeling or attitude or intention. . ..There is nothing that he needs to think, or to adopt, or to
accept; it is not necessary for him to do anything at all.”25 To sum up Frankfurt’s view
then, a person is autonomous with respect to the ﬁrst-order desire that moves her to act if
she identiﬁes with it. To identify with a desire is to make a decisive commitment that
fulﬁlling it should be assigned some position in a person’s preferences, and to be satisﬁed
with that commitment.26
Identiﬁcation as Reason-Giving

More recently, Michael Bratrnan has developed a hierarchical account of personal
autonomy that has become prominent. For Bratman, to identify with a ﬁrst order desire
is to treat it as reason-giving. Bratman’s view of identiﬁcation involves notions of
planning, directed self-governance, and personal identity. In his 1996 paper,

“Identiﬁcation, Decision, and Treating as a Reason,” which was reprinted in his 1999

 

2’ Frankfurt, “Identification and Wholeheartedness,” p. 169 footnote 12.

25 Harry .Frankﬁrrt, “The Faintest Passion” in Necessity, Volition, and Love (Cambridge University Press,
1999), pp. 103-104.

26 Note that this account of autonomy is local in that it is focused on autonomy with respect to a
‘particular" decision or action. This is different than a more global account that emphasizes being the sort
of person who is usually autonomous with respect to her decisions or actions. Another way to cash this out,
and this will become important later, is the difference between synchronic unity and diachronic unity.
Synchronic unity concerns unity at a particular time and diachronic unity concerns unity over time.
Autonomy as Frankfurt and others talk about it is a matter of synchronic unity between the desire that one
identiﬁes with and the desire that moves one to action at a particular time. It is not a matter of uniﬁcation
between the desires that one identiﬁes with over time.

10

collection, Faces of Intention, Bratman argues that to treat a desire as reason-giving is to
treat it as end-setting and to take steps towards that end. Bratman also argues that the
decision to treat the desire as reason-giving is one that the person must be satisﬁed with.27
In his 2002 paper, “Hierarchy, Circularity, and Double Reduction,” Bratman argues that a
person is autonomous if the desire that moves her to act is one that she supports as
ﬁrnctioning as end-setting, which is done by having a non-instrumental higher order self-
goveming policy in favor of it functioning as end—setting. And, she must be satisﬁed
with that policy.28 In his 2003 paper, “Autonomy and Hierarchy,” Bratman emphasizes
that autonomy involves considering whether or not ﬁrst order desires are justifying
reasons for action.29 Bratman further develops his notion of autonomy in his 2005 paper,
“Planning Agency, Autonomous Agency.” He again argues that being autonomous
involves more than being purposive or intentional. It also involves appealing to
considerations treated as in some way legitimizing or justifying—identifying with ﬁrst
order desires involves considering them in light of a self-governing policy with normative
content. In addition, the agent must be satisﬁed with her self-governing policy.30
Bratman has also given consideration to personal identity in relation to autonomy theory.
His focus on identiﬁcation involving policies or plans is partly because he thinks that
these policies constitute an important part of our identity. In “Planning Agency,

Autonomous Agency,” he notes that a Lockean account of personal identity (the

 

27 Michael Bratman, “ldentiﬁcation, Decision, and Treating as a Reason” in Faces of Intention (Cambridge
University Press, 1999), pp. 200-201.

23 Michael Bratman, “Hierarchy, Circularity, and Double Reduction,” in Contours ongency edited by
Sarah Buss and Lee Overton (The MIT Press, 2002), p. 77.

29 Michael Bratman, “Autonomy and Hierarchy,” Social Philosophy and Policy (2003), p. 168.

30 Michael Bratman, “Planning Agency, Autonomous Agency,” in Personal Autonomy: New Essays on
Personal Autonomy and Its Role in Contemporary Moral Philosophy edited by James Stacey Taylor
(Cambridge University Press, 2005), p. 43.

ll

sameness of a person over time) involves overlapping and interwoven plans.31 To sum
up Bratman’s view, then, a person is autonomous with respect to the ﬁrst order desire that
moves her to action if she identiﬁes with it. And, to identify with it is to treat it as
reason-giving in light of one’s over-arching policies.
Historical Identiﬁcation

Recently, John Christman has developed an account of autonomy that has also
become prominent. Christrnan’s View is similar to Frankfurt’s and Bratman’s in that he
is concerned with the attitude one takes towards her ﬁrst order desires. Christman,
however, is concerned with one’s attitude towards the process by which the desire was
formed. On Christrnan’s view, being autonomous is partly about how one came to have
those ﬁrst order desires. In his 1991 article, “Autonomy and Personal History” and his
1993 article, “Defending Historical Autonomy: A Reply to Professor Mele,” he argued
that a person is autonomous if she:

(i) Did not resist the development of the ﬁrst order desire when attending to the
process of development. Or (in cases where she did not attend to the
process of development), she would not have resisted the development had
she attended to the process.

(ii) And, the lack of resistance was not under the inﬂuence of factors that might
inhibit self reﬂection.

(iii) And, the self reﬂection was “minimally rational” (meaning that it did not
involve self-deception, or mistakes in logical inference).

(iv) And, the agent is minimally rational with respect to desire at t (where

minimal rationality demands that an agent experience no manifest conﬂicts

 

3' Bratman, “Planning Agency, Autonomous Agency,” p. 42.

12

of desires or beliefs that signiﬁcantly affect the agent’s behavior and that are

not subsumed under some otherwise rational plan of action).32

Like Christman, Gerald Dworkin is concerned with the process by which one
came to have the desires that she has.3'3 In his article “Autonomy and Behavior Control,”
Dworkin argued that autonomy = identiﬁcation with the desire + the desire was formed in
an independent manner.34 Years later, Dworkin would revise his view in his 1988 book,
The Theory and Practice of Autonomy. In this book, Dworkin dropped identiﬁcation as a
requirement for autonomy. He argued that when we think of paradigm cases of lack of
autonomy (such as lobotomization of an unsuspecting person, or manipulation), what is
crucial is not whether or not they identify with their ﬁrst order desires; it’s that they don’t
even have the capacity to reﬂect on their ﬁrst order desires. Hence, on Dworkin’s revised
View, a person is autonomous if she has the capacity to reﬂect (in a procedurally
independent way) on her ﬁrst order desires, and either accept them or try to change them
in light of her second order desires. Because of his concern about the general capacity to
reﬂect on desires, Dworkin’s view of autonomy is global rather than local. He is not

concerned with whether a person is autonomous with respect to any particular ﬁrst order

 

32 John Christman, “Autonomy and Personal History,” Canadian Journal of Philosophy 21 (1991), pp. 10-

1 6.

Christman added criterion four to his account in response to a criticism made by Alﬂed Mele. See

“Defending Historical Autonomy: A Reply to Professor Mele,” Canadian Journal of Philosophy 23 (1993),
. 228.

g3 Gerald Dworkin, “Acting Freely,” Nous 9 (1970), pp. 367-383.

Although I am mentioning Gerald Dworkin’s work on autonomy rather late in my analysis of the different

theories of autonomy, Dworkin was the ﬁrst to introduce the notion of a hierarchical self, shortly before

Frankﬁrrt in 1970. And, Dworkin was the ﬁrst to directly link the hierarchical view with the term

“autonomy.”

3’ Gerald Dworkin, “Autonomy and Behavior Control,” Hastings Center Report 6 (1976), pp. 23-28.

Dworkin actually used the term “authenticity,” not “identiﬁcation;” but current literature has replaced the

terminology of authenticity with that of identiﬁcation. It is also worth noting that independence in the

13

desires. A person is autonomous to the extent that she reﬂects on ﬁrst order desires and
changes them or accepts them based on that reﬂection.35 It is worth noting, however, that
in his scathing criticism of Dworkin’s dropping of identiﬁcation, Haworth shows how
dropping identiﬁcation causes Dworkin’s theory to be ﬂawed and how Dworkin’s
concerns could be addressed without the dropping of identiﬁcation.36 To sum up, for (the
early) Christrnan and (the early) Dworkin a person is autonomous with respect to the ﬁrst
order desire that moves her to act if she identiﬁes with the process by which it was
formed.
Identiﬁcation as Non-Alienation

Christman further developed his view of autonomy in his 2001 paper,
“Liberalism, Autonomy, and Self-Transformation.” He developed an account that he
calls “autonomy as authenticity.” On this account, a person is autonomous so long as she
is minimally competent and she is authentic in the sense of being moved by values that, if
she undistortedly reﬂected on them and the process by which they were formed, she
would not feel deeply alienated ﬂom them.3'7 So, it is not that she must identify in a
positive sense with the process by which the desires were formed, she just must not feel
alienated ﬂom it. Christman goes on to describe alienation ﬂom X (where X is a desire or
trait) as experiencing a negative affect about X, experiencing diluted or conﬂicted
motivation stemming ﬂom X, feeling constricted by X, as though it were an external

force. Moreover, to feel alienated ﬂom X is to feel a need to repudiate X, to reject X and

 

manner which the desire was formed is what Dworkin terms “procedural independence,” and independence
in the content of 2‘Id order desires he terms “substantive independence.”

35 Gerald Dworkin, The Theory and Practice of Autonomy (Cambridge University Press, 1988), p. 20.

36 Haworth, Lawrence, “Review: Dworkin on Autonomy,” Ethics 102 (1991), pp. 129-139.

37 John Christman, “Liberalism, Autonomy, and Self-Transformation,” Social Theory and Practice 27
(2001), p. 202.

14

alter it as much as possible, and to resist X’s effects.38 To sum up the more recent
Christman, a person is autonomous with respect to the ﬁrst order desire that moves her to
act if she does not feel alienated ﬂom the process by which the desire was formed.
AUTONOMY AS COHERENCE

Laura Waddell Ekstrom has departed ﬂom hierarchical identiﬁcation analyses of
personal autonomy, and developed a coherence account of personal autonomy. In her
1993 article, “A Coherence Theory of Autonomy,” Ekstrom explains that she developed
her account based on a skepticism about certain aspects of hierarchical accounts: namely,
about where second order desires get their authority and about why the self is thought to
be determined by second order desires.39 On Ekstrom’s view, which is developed ﬁrrther
in her 2005 paper, “Autonomy and Personal Integration,” one acts autonomously when
the cause of her action is a preference that was un-coercively formed and that coheres
with her other preferences and beliefs.40 Ekstrom argues that a preference is a desire that
one wants to be effective in action and was formed in the search of the good (which can
be intrinsic or instrumental“).42 Ekstrom views the set of preferences that cohere
(“authorized preferences”) as making up the true self because they are long-lasting guides
for action that will likely remain because they are supported by reasons, they are “fully
defensible” against external challenges, and they are the ones that the person will act on

wholeheartedly.43 And, it is when we act ﬂom our true self that we act autonomously.

 

38 Christman, “Liberalism, Autonomy, and Self-Transformation,” p. 202.

39 Laura Waddell Ekstrom, “A Coherence Theory of Autonomy,” Philosophy and Phenomenological
Research 53 (1993), p. 602.

4° Laura Waddell Ekstrom, “Autonomy and Personal Integration,” in Personal Autonomy: New Essays on
Personal Autonomy and Its Role in Contemporary Moral Philosophy edited by James Stacey Taylor
(Cambridge University Press, 2005), pp. 148-152.

4' Ekstrom, “A Coherence Theory of Autonomy,” p. 603.

’2 Ekstrom, “Autonomy and Personal Integration,” pp. 148-152.

’3 Ekstrom, “Autonomy and Personal Integration,” p. 158.

15

SUBSTANTIV E ACCOUNTS OF AUTONOMY

Identiﬁcation and coherence accounts of autonomy have been subject to many
feminist critiques. Feminist philosophers have argued that non-substantive requirements
like identiﬁcation and coherence fail to acknowledge that desires and preferences are
adaptive“—so that while identiﬁcation with one’s desires or coherence among one’s
preferences may be necessary for autonomous action, it is not sufﬁcient (e. g. the
deferential wife who identiﬁes with her desire to spend all of her time serving her
husband is not autonomous). Feminist philosophers have also argued that hierarchical
identiﬁcation accounts make individualistic metaphysical assumptions about persons that
are false45 and make psychological assumptions about persons that are false (such as
self-transparency)"6

In response, some feminist philosophers have argued for content-speciﬁc accounts
of autonomy47 where the desire or decision that one ends up identifying with must be one

that is in their best-interest in order for the decision or action to be a ﬂee one. Other

 

’4 See Diana Meyers, “Intersectional Identity and the Authentic Self?: Opposites Attract!” in Relational
Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self edited by Catroina Mackenzie
and Natalie Stoljar (Oxford University Press 2000), pp. 151-180.; Marilyn Friedman, Autonomy, Gender,
Politics (Oxford University Press 2003), pp. 38, 30, 41.

’5 See John Christman, “Feminism and Autonomy” in Nagging Questions: Feminist Ethics in Everyday Life
edited by Dana Bushnell (Rowman & Littleﬁeld Publishers Inc. 1995), pp. 28, 32; Marilyn Friedman,
Autonomy, Gender, Politics (Oxford University Press 2003), pp. 15, 17, 30-35.

46 See Diana Meyers, “Intersectional Identity and the Authentic Self?: Opposites Attract!” in Relational
Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self edited by Catroina Mackenzie
and Natalie Stoljar (Oxford University Press 2000), pp. 151-180.; Marilyn Friedman, Autonomy, Gender,
Politics (Oxford University Press 2003), pp. 38, 30, 41.

47 For examples of content speciﬁc (substantive) accounts of autonomy see Paul Benson, “Feminist Second
Thoughts about Free Agency,” Hypatia 5 (1990): 47-64; Paul Benson, “Autonomy and Oppressive
Socialization,” Social Theory and Practice 17 (1991): 385-408; Sarah Buss, “Autonomy Reconsidered,”
Midwest Studies in Philosophy 19 (1994): 95-121; Sigurdur Kristinsson, “The Limits of Neutrality: Toward
a Weakly Substantive Account of Autonomy,” Canadian Journal of Philosophy 30 (2000): 257-286;
Marina Oshana, “Personal Autonomy and Society,” Journal of Social Philosophy 29 (1998): 81-102;
Natalie Stoljar, “Autonomy and the Feminist Intuition,” in Relational Autonomy, 94-111; Susan Wolf,
“Sanity and the Metaphysics of Responsibility,” in Ferdinand Schoeman, ed., Responsibility, Character,
and the Emotions (New York: Cambridge University Press, 1987): 46-62; and Susan Wolf, Freedom within
Reason (New York: Oxford University Press, 1990).

I6

feminist philosophers have suggested ways to improve the process by which a person
comes to identify with certain desires and to decide that she wants them to be effective in
action. There are certain skills that work to make sure that a person’s preferences are not
adaptive or oppressive, and these skills also make it more likely that she will be
autonomous (in Christrnan’s sense of being satisﬁed with the process by which her

desires were formed). Some of these skills are suggested by Diana Meyers:

I. Imaginative skills that enable individuals to envisage a range of self-
concepts they might adopt
2. Introspective skills that sensitize individuals to their own feelings and

desires, that enable them to interpret their subjective experience, and
that help them judge how good a likeness a self-portrait is

3. Memory skills that enable individuals to recall relevant experiences not
only ﬂom their own lives but also experiences that associates have
recounted or that they have encountered in literature or other artforms

4. Communication skills that enable individuals to get the beneﬁt of
others’ perceptions, background knowledge, insights, advice, and
support

5. Analytical and reasoning skills that enable individuals to compare
different self-concepts and to assess the relative merits of these
alternatives

6. Volitional skills that enable individuals to resist pressure ﬂom others to
embrace a conventional self-concept and that enable them to maintain
their commitment to the self-portrait that they consider genuinely their
own, that is, authentic

7. Interpersonal skills that enable individuals to join forces to challenge
and change social norms48

 

‘8 Diana Meyers, Gender In the Mirror: Cultural Imagery and Women ’s Agency (Oxford University Press
2004), p. 166. Some may ﬁnd Meyers’s skills to require too much introspection, reﬂection, and activity.
As Richard Double remarks, “The man-of-action, so the paradigm goes, shoots ﬁrst and asks questions
later. The true ﬂee spirit may not ask questions at all. ...For many persons, life is to be lived, not worried
over. . .” Double suggests the possibility that unreﬂectiveness may be alright, so long as unreﬂectiveness is
part of her “individual management style” (how she believes she should go about making choices). See
Richard Double, “Two Types of Autonomy Accounts,” Canadian Journal of Philosophy Vol. 22 No. 1
(1992), p. 73.

17

Autonomy as a Bioethical Concept
AUTONOMY AS DECISION MAKING CAPACITY

When bioethicists talk about patient autonomy, they are concerned with warding
off unjustiﬁed paternalism. When they talk about a patient’s right to exercise her
autonomy, they mean the patient’s right to make her own decision, so long as she has
decision making capacity (DMC) and so long as the patient’s choice does not wrongfully
interfere with the rights of others.49 A patient is autonomous with respect to a choice so
long as she has DMC with respect to that choice and she is not subject to controlling
inﬂuences.so Before explaining what it means to have DMC with respect to a choice and
what it means to not be subject to controlling inﬂuence, it will be useful to discuss the
ground for why we think that allowing patients to make their own decisions/exercise their
autonomy is valuable.

Allowing a person who has DMC and who is not subject to controlling inﬂuences
to make their own decisions is valuable. But what does it mean to have DMC and what
does it mean to not be subject to controlling inﬂuences? In order to have DMC, the
patient must have the ability to understand relevant facts, the ability to appreciate
relevance of facts to personal situation, the ability to rationally manipulate information
and consider it in light of values and goals to arrive at a choice, and the ability to

communicate that choice.5 '

 

’9 Technically, “decision making capacity” is the clinical term used by bioethicists and clinicians, and

“competency” is the legal term used by the courts.

50 Tom Beauchamp and James Childress, Principles of Biomedical Ethics (Oxford University Press: 1994),
. l 2 l .

a M. Bauer, The Field Guide to Psychiatric Assessment and T reatrnent (Philadelphia: Lippincott/Williams

and Wilkins, 2003).

18

In sum, the basic elements of DMC are understanding, appreciating, deliberating,
concluding, and communicating. When assessing understanding it is important to note
that patients may not be familiar with medical terminology and they may be susceptible
to information overload. A useful way to help patients understand is to explain things
using analogies (e. g. the risk of procedure X is about the same as the risk of using a
power tool).52 Even if a patient understands the information that she is given, she needs
to accept it as true and appreciate what it means for her.

To assess that a patient has DMC, the following questions are usually asked: Does
the patient understand his/her medical condition? Does the patient understand the risks
and beneﬁts of the proposed interventions? Can the patient weigh the burdens and
beneﬁts of each proposed intervention (test, medication, procedure)? Does the patient
understand the concept of life saving interventions? Can the patient express his/her
health care values?53

The other component of being autonomous with respect to a particular decision is
that the patient is not subject to controlling inﬂuences. Three forms of controlling
inﬂuence are coercion, persuasion, and manipulation. Coercion occurs when one person
intentionally uses a credible and severe threat of harm or force to control another.
Persuasion occurs when a person is convinced to believe in something through the merit
of reasons advanced by another person. Manipulation occurs when a person is being
swayed to do what the manipulator wants by means other than coercion or persuasion

(e. g. information management).54

 

52 Tom Beauchamp and James Childress, Principles of Biomedical Ethics, p. 158.

53 PL. Schneider and K.A. Brarnstedt, “When Psychiatry and Bioethics Disagree About Patient Decision
Making Capacity” in Journal of Medical Ethics Vol. 32 February 2006, pp. 90-93.

5’ Tom Beauchamp and James Childress, Principles of Biomedical Ethics, pp. 164-165.

19

In practice, when determining whether or not a patient is autonomous with respect
to a particular decision, little attention is paid to whether or not the patient is subject to
any of the abovernentioned controlling inﬂuences. If, however, it was obvious that the
patient was subject to a controlling inﬂuence, clinicians do not simply revoke the
patient’s right to have her decision respected. They attempt to remove the controlling
inﬂuence and then (provided that she still has DMC), follow whatever decision is made
once the controlling inﬂuence is removed. Likewise, if a patient lacks DMC, clinicians
would not simply revoke the patient’s right to have her decisions respected. They would
attempt to restore her DMC. Often DMC cannot be restored, but sometimes a patient
lacks DMC for physiological reasons such as encephalopathy, dehydration, or severe
clinical depression. In these cases DMC can be restored provided that there is time to do
so before any decisions need to be made. I do want to note that there may be cases where
a patient no longer has DMC, but we permit their decision anyway for other reasons (e. g.
it is in their best-interest). So, having DMC and hence bioethical autonomy is sufﬁcient
but not necessary for permitting someone’s decision.

The Connection Between Metaphysical Autonomy and Bioethical Autonomy

In a way, the conception of autonomy in bioethics (a right to make decisions so
long as there is an adequate understanding of the external world/the medical condition)
seems completely different ﬂom the conceptions of autonomy operating in the theoretical
and metaphysics discussions (an ideal for persons to aim for that involves a certain
relationship among their internal states). It is not as if physicians go around forcing
patients to act in ways that ﬁt better with their settled values, and it is not as if we hold

persons less responsible for their decisions if they are half-hearted. Frankfurt himself has

20

admitted that he is interested in metaphysics (just want it means to identify with one
desire rather than the other) and not with ethics (the ethics surrounding the cause of the
identiﬁcation or the bearing on ethical questions). Frankfurt says, “If this were a question
of public policy, if we were trying to decide how to conduct our judicial system or our
system of rewards and punishments, it would not be tenibly useful to talk about
wholeheartedness. . . 3’55 Part of the reason is because we could never really know
whether a person is/was wholehearted about a desire moving her to act, for we cannot get
inside the person’s head, so to speak.

Although it is true that there is a big difference in the ways that the terms are used
and the purposes that the two conceptions have been formed for, I think that Frankfurt is
incorrect to say that metaphysical autonomy theories are useless for applied issues such
as bioethical autonomy. There many connections between metaphysical autonomy
theories and bioethical autonomy. In particular, metaphysical theories of autonomy have
inﬂuenced bioethical autonomy in two ways: (1) certain classes of people (namely young
children and certain psychiatric patients) are have been judged to not have bioethical
autonomy because they lack metaphysical autonomy, and (2) more generally, the criteria
for judging decision making capacity and bioethical autonomy draw on metaphysical
autonomy theories.

CHILDREN

Children (mature minors excluded) are not granted bioethical autonomy. They
are not permitted to make medical decisions and have them followed so long as they are
not interfering with the rights of others. And this is because children lack DMC.

Children may lack DMC because they lack understanding of their condition, the possible

 

55 “A Discussion With Harry Frankfurt,” in Ethical Perspectives Vol. 5, 1998, p. 35.

21

therapies, and the consequences of reﬁrsing. But even if they did understand and have a
grasp of these things, they would lack DMC for another reason. They lack DMC and
bioethical autonomy because they lack metaphysical autonomy. And they lack
metaphysical autonomy because they are not able to form a will. They are not able to
reﬂect on and deliberate about their ﬁrst order desires in light of their values, and then
settle on a ﬁrst order desire that they want to move them to action. This is partly because
their values are not really formed. Allen Buchanan notes, “An important issue is the
extent to which children’s values adequately reﬂect their future interests. While children
in the seven to thirteen age range have largely left the earlier magical stage of thinking
and now view the world in concrete, naturalistic terms, they can have great difﬁculty
anticipating their future. . . .A related problem is the instability in children’s values in this
period. . . 3’56
PSYCHIATRIC PATIENTS

Some psychiatric patients are not granted bioethical autonomy (the ability and
permission to make their own medical decisions)lt is true that in many cases of mental
illness, particularly psychosis, the patient lacks DMC because they lack understanding.
Delusional beliefs common in some forms of schizophrenia can lead to a lack of
understanding, as a patient may, for example, think that a medication is really a poison.57
But even if they did not lack understanding, they would lack DMC for another reason.
They lack DMC and bioethical autonomy because they lack metaphysical autonomy.

And, they lack metaphysical autonomy because there is a disorder of their will. They

may understand their condition, the possible therapies, and the consequences of reﬁrsing

 

56 Allen Buchanan and Dan Brock, Deciding For Others: The Ethics of Surrogate Decision Making
(Cambridge University Press 1990), p. 221.

22

treatment, but may express decisions that are not consistent with their professed values;
indicating a problem in the process of deliberation in light of values (a problem in the
formation of their will).58 Or, their values may be distorted because of their illness.
Buchanan notes that severe mental illness can seriously distort a person’s underlying and
enduring aims and values, “For example, severe depression can result in a person no
longer caring about the aims and projects that before the onset of depression had given
meaning and value to life, while mania may lead a person to pursue grandiose and

”59 So, a person with mental illness may lack

unrealistic new plans and projects.
bioethical autonomy not because they cannot form a will (as in the case of children), but
because their decisions are not consistent with their will, or their will is distorted by their

mental illness.60

 

57 Buchanan and Brock, Deciding For Others, p. 318.
58 Mental illness is not necessary for the patient to express decisions that are not consistent with their
professed values. A patient may, for example, be weak of will. For a discussion of this, see footnote 57.

9 Buchanan and Brock, Deciding For Others, p. 318.
60 Note that in cases of “weakness of will,” a similar thing occurs. The agent’s desires, choices, or actions
are not consistent with their will (with the desire that they want to have and want to move them to action).
A dieter has a desire to eat a piece of cake (chocolate cake, with buttercream ﬂosting, and chunks of
chocolate delicately sprinkled on top), chooses to eat the cake, and does eat the cake. But, upon reﬂection,
he is not satisﬁed with his desire to eat the cake, and he does not want it to move him to action. In eating
the cake anyway, he is being “weak of will.” Or, a slight variation, perhaps he does want his desire to eat
the cake to be effective in action, but this is only because his craving has distorted his professed value of
healthiness. Most of us, however, would grant him permission to eat the cake. Most of us would not,
however, grant the mentally ill patient permission to refuse treatment. There is an inconsistency here, as
both cases are ones where the person has a similar disorder of the will—their decisions are not consistent
with their will, or they are, but only because their will has been distorted (either by a craving or by a mental
illness). One could argue then, that in the context of bioethics, we should revoke bioethical autonomy ﬂom
all cases where a person’s decision is not consistent with her will (or it is but only because her will has
been distorted); and not just limit the revocation to cases where the cause of this is mental illness. If we are
pushed, and want to be consistent, we may want to say that anyone who (for whatever reason) acts contrary
to their settled will is not acting in an autonomous way, and that therefore the resulting medical decision is
also not an autonomous one. The weak of will patient should not be granted bioethical autonomy. Now,
whether that means we should override their judgment will depend on other facts (e. g. their best interest).
Someone may respond by arguing that we do not need to be consistent in the granting of permission to the
mentally ill patient and the akratic cake-eater because the two cases are disanalogous: the consequences of
refusing treatment are much more serious than eating cake. But one can easily make the cases analogous:
simply substitute having unprotected sex with an HIV + person every night for eating cake. Now, both
situations (the mentally ill one and the weak of will one) are ones with dire consequences. However,
without reﬂecting, we may say that the promiscuous weak of will chap should be granted permission to

23

INFLUENCE ON THE CRITERIA

Apart ﬂom these two particular cases and more generally speaking, the concern
for bioethical autonomy and the criteria by which we judge decision making capacity
have been inﬂuenced by and draws on metaphysical autonomy theories. A major focus of
most accounts of metaphysical autonomy is the hierarchical nature of persons—the
unique ability of persons to reﬂect on their ﬁrst order desires. As Buchanan notes, “The
capacities for deliberation, choice and action that normal humans possess make it
possible for them to form, revise over time, and pursue in action a conception of their
own good. Having a conception of the good is more than merely possessing desires that
support goal-directed behavior—a feature that persons share with animals. Persons have
a capacity for reﬂective self-evaluation, for considering what they want their motivations
to be, what kind of persons they want to become.”61 We value this ability, and as a
result, this value has led to a concern about bioethical autonomy—a concern for allowing
persons to exercise this ability in the realm of medical decision making. Moreover,
determinations of DMC that grant a patient bioethical autonomy have been inﬂuenced by
metaphysical autonomy theories in that the criteria require deliberation in light of patient
values. Patients are required to do more than just understand the consequences of their
choices. They are also required to reﬂect on the choices they make

Although the criteria for evaluating DMC and granting bioethical autonomy are

inﬂuenced by metaphysical autonomy theories like Frankﬁrrt’s in that patients are

 

carry out his choice, but the mentally ill patient should not. But, upon reﬂection, we see that the cases are
the same: they are both cases of agents whose actions are inconsistent with their will. Therefore, cases of
mental illness and weakness of will may both be cases where the patient lacks decision-making capacity,
and hence bioethical autonomy. And, we do not permit their decision (unless there is another reason for
doing so—such as it is in their best-interest).

6' Buchanan and Brock, Deciding For Others, p. 39.

24

required to reﬂect on their choices and consider them in light of their values (i.e. they are
required to form 2"d order desires and volitions, I would argue that they are not
inﬂuenced enough. In practice, little attention is paid to whether or not the decisions and
values that the patient is professing are her “real” decisions and values (ones that she
identiﬁes with or endorses). Paying closer attention to this in determinations of decision
making capacity and aiming for a patient’s decisions and values to be ones that they
“really want” (to use Frankfurt’s terminology) would enhance bioethical autonomy.621n
closing Chapter One, let me note what I hope to have accomplished. I have illustrated
that the term “autonomy” has been used very arnbiguously in the literature. I have
developed a taxonomy to help organize the ways in which the term is used. I have drawn
a distinction between the way that the term is used in bioethical (applied) contexts and in
metaphysical (theoretical) contexts. Although metaphysical autonomy theories are very
different ﬂom bioethical autonomy theories, metaphysical autonomy theories do have
things to offer bioethical autonomy. In some cases (e. g. young children and some
psychiatric patients), they help explain why a person should not be granted bioethical
autonomy (e. g. they lack a will or they have a distorted will, respectively). In other
cases, they illustrate how bioethical autonomy could be enhanced. An interesting case
that has not yet been discussed is the case of ambivalence, where a person’s will is
divided. The remainder of the dissertation will focus on this speciﬁc case and the impact

that it has on both metaphysical and bioethical autonomy.

 

‘52 It is especially important to pay close attention to this in high stakes decisions. As some bioethicists have
argued, the standards for DMC should become more stringent as the stakes rise. See Buchanan and Brock,
Deciding For Others, pp. 48-57.

25

Chapter Two: A Philosophical Analysis of Ambivalence

Susanna: I'm ambivalent. In fact that's my new favorite word.

Dr. Wick: Do you know what that means, ambivalence?

Susanna: I don’t care.

Dr. Wick: If it's your favorite word, I would've thought you would...

Susanna: It *means“ I don‘t care. That's what it means.

Dr. Wick: On the contrary, Susanna. Ambivalence suggests strong feelings... in opposition. The
preﬁx, as in "ambidextrous," means "both." The rest of it, in Latin, means "vigor." The word
suggests that you are torn... between two opposing courses of action.”

Surprisingly little attention has been paid to the phenomenon of ambivalence64 in
contemporary philosophical literature. I say surprising because much attention has been
paid to autonomy, weakness of will, self-deception, and integrity. All of these
philosophical discussions rely on the assumption that there is something that the person
really wants, or really values. But ambivalence calls this into question; it calls into
question whether there is something that the person really wants or really is. So, the issue
of ambivalence needs to be properly addressed if the issues of autonomy, weakness of
will, self-deception, and integrity are to be properly addressed. As Lynn McFall says, “In
order to sell one’s soul, one must have something to sell.”65
First Person Accounts of Ambivalence

Before engaging in a conceptual analysis of the phenomenon of ambivalence, it
will be useful to have an idea of the ﬁrst person experience of the phenomenon. Steve
Harrist conducted a phenomenological study of ambivalence, focusing on what the

experience of ambivalence is like for those who experience it. His method of

 

63 Girl Interrupted

6’ As I am using the term, ambivalence is a feeling that arises out of a certain relationship between mental
states. More speciﬁcally, it is the feeling of internal conﬂict that arises when a person is drawn both
towards and away ﬂom an object, a desire, or a course of action.

65 Lynn McFall, “Integrity,” Ethics Vol. 98 No. 1 (1987), p. 10.

26

investigation was serni-structured interviews that were transcribed and then interpreted.
He found that the experience of ambivalence has the following movements: background,
disorientation, exploration, and resolution."6 Background is a state of pre-ambivalence
where the person is not really paying attention; they have a taken for granted the ﬂow and
direction of life. When ambivalence occurs it interrupts this taken for granted ﬂow, and
urgently calls it into question.67 The person enters a state of disorientation. The person
experiences a feeling of something not being right with the self, a feeling of being
trapped and powerless, a sense of things not being under the person’s voluntary control,
and feelings of being unsettled and unanchored.68 Part of the phenomenological
experience of ambivalence is an impulse for a return to a simpler experience of the
world.69 The person then enters a stage of exploration where they weigh options,
question, and make judgments.70 It is at this point that some feelings may take on greater
value or importance than others and this may lead the person into the stage of resolution.
Or, the person may not move into the stage of resolution, in which case they experience
feelings of being stuck, being powerless, obsessing, being unable to decide, and
anxiety“. The subjects who did experience a resolution of their ambivalence did not have
all of their conﬂicting feelings disappear, but they did feel capable of moving forward in
a determinate direction.72 “Resolution involves a sense that there is less uncertainty and
confusion as it becomes clearer which of the feelings may predominate. As one feeling

“overrides” other feelings to a signiﬁcant extent, it is not necessarily the case that other

 

6" Steve Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” Journal of
Phenomenological Psychology Vol. 37 No. 1 2006, p. 99.

67 Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 101.

68 Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” pp. 100-102.

69 Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 99.

7° Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” pp. 103-104.

7' Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 89.

27

feelings simply disappear. Rather, they may recede into the background of the
individual’s experience and become less ﬁgural.”73 Resolution of ambivalence was
experienced as a change in perspective» and this change had to come before the choice,
or else the choice felt arbitrary and unsatisfactory. 74
A Conceptual Analysis of Ambivalence
LOCATIN G AMBIVALENCE

With the experience of ambivalence in mind, let us move to a conceptual analysis
of the phenomenon. Perhaps the philosopher who has written the most on ambivalence is
Harry Frankfurt. Frankfurt has characterized ambivalence as a conﬂict that occurs during
will formation. According to F rankﬁrrt, an ambivalent person experiences conﬂict during
the process of reﬂecting on a desire that she has, and the conﬂict prevents her ﬂom
forming a will”; ﬂom taking a position on whether it is a desire that she wants to have
and to be effective in action. Ambivalence prevents there ﬂom being a “something that

she really wants”/a truth about herself.76 Recall that for Frankfurt, the objects, people,

 

72 Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 105.

73 Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 106.

7’ Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 106.

7‘ Cuypers has identiﬁed three different senses in which Frankfurt talks about the will. First, the
“appetitive will” is the ﬁrst order desire that is effective in action. For example, I desire a certain person
and as a result of that desire I enter into a relationship with that person. The desire that causes me to enter
into the relationship is my will. Second, the “active will” is the activity of identifying with the ﬁrst order
desire that is effective in action. When one identiﬁes with the ﬁrst order desire that is effective in action,
she forms what Frankfurt calls a second order volition, or will. Imagine, for example, that I identify with
the desire that causes me to enter into a relationship with that person. I want to have it and I want it to
move me to action. That identiﬁcation is my will. It so happens that in this case the will that I do have (my
appetitive will) matches the will that I want to have (my active will). Third, the “substantial will”, which is
the set of our volitional necessities—what we care about, love, and ﬁnd unthinkable. The substantial will
inﬂuences what desires we do or do not identify with. Imagine, for example, that I cannot help loving the
person that I enter into a relationship with. This love is an essential part of my will. What I am talking
about when I characterize ambivalence (and what I think Frankfurt means) is a conﬂict at the level of the
active will such that I do not settle on what I want to move me to action; I do not form a will.

See Stefaan Cuypers, “Harry Frankfurt on the Will, Autonomy, and Necessity” in Ethical Perspectives Vol.
5, 1998, pp. 44-45. Note that the terms in quotations are Cuypers’s, not Frankfurt’s.

7" Harry Frankfurt, “The F aintest Passion” in Necessity, Volition, and Love (Cambridge University Press,
1999), p. 100.

28

courses of action etc. that we desire are objects of ﬁrst order desires. We, as humans,
have the ability to reﬂect on these desires and either want to have them or not want to
have them—these desires about our ﬁrst order desires are second order desires. If, upon
reﬂecting on a desire, we want to have it and we want it to be effective in moving us to
action, then it is our will (the will is also referred to by Frankfurt as a second-order
volition).77 In ambivalence, however, a person is indecisive about whether he wants a
particular desire to move him to action. For example, in Aeschylus’s play Agamemnon,
Agamemnon is faced with the conﬂict of whether or not to kill his daughter Iphigenia in
order to spare the lives of his cavalry. The anguish of his mental conﬂict can be felt in
his words:

"A heavy doom, sure, if God's will were broken;

But to slay mine own child, who my house delighteth,

Is that not heavy? That her blood should ﬂow

On her father's hand, hard beside an altar?

My path is sorrow wheresoe'er I go.

Shall Agamemnon fail his ships and people,

And the hosts of Hellas melt as melts the snow?

They cry, they thirst, for a death that shall break the spell,
For a Virgin's blood: 'tis a rite of old, men tell.
And they burn with longing.--O God may the end be well

! "78
Agamemnon has a desire to spare the lives of his cavalry. His attitude towards that

desire, however, is a conﬂicted one. He is drawn towards the desire because he feels a

duty to his cavalry. He is drawn away ﬂom the desire because he loves Iphigeneia.79 He

 

77 Although Frankfurt does refer to the will as a second order volition (see Harry Frankfurt, “Freedom of
the Will and the Concept of a Person” in The Importance of What We Care About (Cambridge University
Press, 1988)), I thank James Stacey Taylor for pointing out that Frankfurt is wrong by deﬁnition in
referring to the will as a second order volition , since an X of the second-order is an X about an X; so a
second-order desire is a desire about a desire. A person’s volitional endorsement of a ﬁrst-order desire thus
involves merely a ﬁrst-order volition, as its object is not another volition.

78 Aeschylus, Agamemnon.

79 Ambivalence could either be feeling drawn toward and drawn away ﬂom X at t1 (simultaneous) or it
could be feeling drawn toward X at t1 and t3 and away ﬂom X at t2 and t4 (vacillation).

29

thus cannot decide which desire he wants to be effective in action; he is ambivalent. As a
result, he does not form a will.

Ambivalence can occur at either the ﬁrst order level (e. g. a person is ambivalent
about an object that can be described propositionally—for example eating steak for
dinner or eating chicken for dinner) or the second order level (e. g. a person is ambivalent
about a desire that she has for the object that can be described propositionally—for
example about having the desire to eat chicken (or steak) for dinner and about it being
effective in action). Ambivalence at the ﬁrst order can further be distinguished ﬂom
ambivalence at the second order by noticing that a ﬁrst order ambivalence does not
usually involve any values or goals (the person is just drawn towards the object of
chicken and the object of steak), whereas a second order ambivalence does (the person is
ambivalent about her desire for health (chicken) being effective in action or her desire for
pleasure (steak) being effective in action. Henceforth, when I talk about ambivalence I
will be referring to ambivalence occurring at the second order level. This is because it is
the most philosophically signiﬁcant occurrence of ambivalence for the following reasons.
First, it is likely the case that ﬁrst order ambivalence boils down to second order
ambivalence.80 Often, the reason that a person is ambivalent about an object that can be
described propositionally (e. g. steak or chicken) is because she is ambivalent about which
of her values (second order desires) she wants to be effective in that circumstance (e. g.

her value of pleasure or her value of health). Second, the second order level is the locus

 

80 It is interesting to think about whether ﬁrst order ambivalence and second order ambivalence always boil
down to the same thing. I do not think that they always do, for I can think of an example where a person is
ambivalent at a ﬁrst order level but not at a second order level: a person is ambivalent about smoking a
cigarette. She has a desire to smoke it, and a desire not to. But she is not ambivalent at the second order
level. She simply does not want to have the desire to smoke the cigarette, nor does she want it to be
effective in action.

30

of the theories of autonomous action—and I am ultimately concerned about the effect of
ambivalence on autonomy, hence I am concerned with ambivalence that occurs at the
second order level.

Returning to F rankfurt’s characterization of ambivalence, much still remains to be
answered. For example, what is the nature of the reﬂection on ﬁrst order desires? What
does it mean to want to have a ﬁrst order desire and to want it to be effective in action?
And, why is the person drawn towards and away ﬂom the desire?

In his seminal paper, “Freedom of the Will and the Concept of a Person,”
Frankfurt has the following to say about reﬂection on ﬁrst order desires, “In speaking of
the evaluation of his own desires and motives. . .I do not mean to suggest that a person’s
second-order volitions necessarily manifest a moral stance on his part towards his ﬁrst-
order desires. It may not be ﬂom the point of view of morality that the person evaluates
his ﬁrst-order desires. Moreover, a person may be capricious and irresponsible in
forming his second-order volitions and give no serious consideration to what is at stake.
Second-order volitions express evaluations only in the sense that they are preferences.
There is no essential restriction on the kind of basis, if any, upon which they are

formed.”81

 

8' Frankﬁrrt, “Freedom of the Will and the Concept of a Person,” p. 19 footnote 6. Some have responded to
Frankfurt by distinguishing between identiﬁcation and endorsement and arguing that our second order
volitions (our wills) are not things that we come to through identiﬁcation (in the sense of forming some sort
of psychological connection), but through endorsement (considering them to be endorsable)—see Cheshire
Calhoun, “Standing For Something,” The Journal of Philosophy Vol. 42 No 5 (1995), p. 243. I think
though that the example of Ivan Illych that Lynn McFall mentions shows the superiority of idendiﬁcation
to endorsement. Ivan Illych endorses his desires but does not identify with them. He says, “Maybe I did
not live as I ought to have done. . . .But how can that be, when I did everything properly?” Ivan Illych had
lived in conformity with “legality, correctitude, and propriety.” “It occurred to him that his scarcely
perceptible attempts to struggle against what were noticeable impulses which he had immediately
suppressed, might have been the real thing, and all the rest false.” See Lynn McFall, “Integrity,” Ethics
Vol. 98 No. 1 (1987), p. 6.

31

To form a second order volition with respect to a particular desire is to not only
want to have it, but to will that it be effective in action. In “Freedom of the Will and the
Concept of a Person,” Frankfurt draws a distinction between wanting to have a desire and
wanting that desire to be effective in action.82 He gives the example of a psychotherapist
who wants to have a desire to smoke because he thinks that it will help him relate to his
narcotics patients, but he does not want the desire to move him to actually smoke. It is
when a person wants to have a desire and wants it to be effective in moving her to action
that she wills it.

In Chapter One, I outlined Frankfurt’s explanation of what it means to want a
desire. To want a desire is to identify with it; to make a decisive commitment that
fulﬁlling it should be assigned some position in a person’s preferences, and to be satisﬁed
with that commitment. To identify with a desire is to commit to it being assigned some
position in a person’s preferences. What is important, and what Frankfurt largely
ignores, is that the person has to make a second decision. She has to decide on the
ordering of the desires that she identiﬁes with. In order to form a will, a person has to
decide which of her desires she wants to be effective in action at a particular time. And, I
would argue, it is in the process of ordering desires that most ambivalence occurs.
Perhaps a visual illustration will be useful:

1. Reﬂect on ﬁrst order desires and decide whether they are ones with which I:

\.

Identify (assign some position) Reject/Outlaw (assign no position)

i

2. Decide on the order

 

82 Frankfurt notes that it is possible for a person to want to have a particular ﬁrst order desire, but to not
really care which ﬁrst order desire moves him to action. Frankfurt terms this person a “wanton.”

32

1.4— The one that I want to move me to act at t1 (the one that I will)
2.
3.

Ambivalence can occur during the ﬁrst decision process (of whether or not to identify
with or reject a particular ﬁrst order desire), but it is most likely to occur during the
second decision process. That is, when a person is deciding which desire (of the desires
she identiﬁes with) she wants to be eﬂective in action at a particular time; when she is
deciding which she wills.
PARALYZIN G AMBIVALENCE

In what I will call “paralyzing ambivalence” a person actually fails to make a
decision about which desire she wants to be effective in action at a particular time; she

I.83 For example, Agamemnon succeeds in making the ﬁrst decision;

fails to form a wil
he decides to identify with his desire for his daughter’s well-being, and he decides to
identify with his desire for his cavalry’s well-being. His paralyzing ambivalence occurs
when he must decide which of those desires he wants to be effective at the time that he is
presented with the dilemma of his daughter or his cavalry perishing. Agamemnon is

par aIyZingly ambivalent and does not form a will.8485

 

Frankfurt, “Freedom of the Will and the Concept of a Person,” p. 21.

83 Although Frankfurt does not discuss it, I imagine that a person could be retrospectively ambivalent as
well. That is, ambivalent when he reﬂects on which desire has moved him to action.

8’ Of course, Agamemnon does later resolve his paralyzing ambivalence and decide that he wants his desire
for his cavalry’s well-being to be the one that is effective in action.

85 A general note about my explanation of ambivalence in this section: a person may be unambivalent about
the ﬁrst decision (whether or not the desire is one that she identiﬁes with), she may even be unambivalent
about the second decision in that she wants either desire I or 2 to move her to action at t1 but not desire 3 or
4. What she is ambivalent about is whether she wants desire I or desire 2 to move her to action at t.. For
example, imagine that a person is unambivalent that she identiﬁes with her desire to go to the beach, her
desire to go the park, her desire to go to the bookstore, and her desire to have sex. She is even
unambivalent that she wants either the desire to go to the beach or the park this afternoon. What she is
ambivalent about is whether she wants her desire to go to the beach or her desire to go to the park to be the

33

RESIDUAL AMBIVALENCE

While paralyzing ambivalence is a conﬂict so deep that the person does not form
a will, a person may experience ambivalence but nonetheless form a will. I will call this
“residual ambivalence”. A person who resolves her paralyzing ambivalence will likely
still experience residual ambivalence. Even though Agamemnon resolves his paralyzing
ambivalence by deciding that he wants his desire to save his cavalry to be effective in
action (Ire forms a will), he is likely still strongly moved and inﬂuenced by the other
desires that he identiﬁes with (namely the desire to save his daughter). I will present a
series of examples to illustrate the phenomenon of residual ambivalence. Consider the

following passage in Evelyn Waugh’s Decline and F all:86

“If I take the money,” he said to himself, “I shall never know whether I have
acted rightly or not. It would always be on my mind. If I reﬁrse, I shall be
sure of having done right. I shall look back upon my self-denial with
exquisite self-approval. By refusing I can convince myself that, in spite of
unbelievable things that have been happening to me during the last ten days,
I am still the same Paul Pennyfeather I have respected so long. It is a test
case of the durability of my ideals. ...I’m aﬂaid you’ll ﬁnd my attitude rather
difﬁcult to understand,” he said. “I suppose it’s largely a matter of
upbringing. There is every reason why I should take the money. ...By any
ordinary process of thought, the money is justly mine. But, I can’t help it;
it’s born in me. I just can’t take that money.”

Pennyfeather feels ambivalent, he feels drawn towards and drawn away ﬂom his ﬁrst
order desire to take the money. This is because he identiﬁes with his desire to take the
money (the money is justly his and he needs it) but he also identiﬁes with his desire to
preserve his upbringing. He is forced to order these identiﬁcations and decide which he
wants to be effective in action at that particular time (he is forced to decide which he

wills); and he does make this decision, he wills for his desire to preserve his upbringing

 

one that is effective in action this aftemoon. This footnote arises ﬂom Geofﬂey Sayre-McCord pushing me
to reﬂect on the way that dividing up desires effects the classiﬁcation of ambivalence.

34

to be the desire that is effective in action.87 But, even though he has taken the stand that
he does not want the desire to take the money to move him to action and remains
determined, he is still strongly inﬂuenced by the desire for the money, and this leaves
him uneasy (residually ambivalent) with his choice.

Also consider Winston’s reaction to Julia’s proposition in George Orwell’s I 984.
Julia passes Winston a note that says “I love you,” in a society where such human
interaction is severely punished by The Thought Police. This note stirred up a mix of
emotions in Winston. He was excited by the idea of a secret relationship with Julia, but
also terriﬁed. This is because Winston identiﬁes with both his desire for safety and his
desire for love. Winston is forced to choose which desire he wants to be effective in
action, and he does. He forms a will when he decides that he wants his desire for love to
be effective in action. Winston says that the idea of refusing Julia’s advances did not
even really cross his mind.88 But of course he is still strongly inﬂuenced by the desire for
safety, and this leaves him uneasy (residually ambivalent) with his choice to meet her.

Imagine a parent who has a desire for a relationship with his son who is addicted
to drugs. He wants to continue his relationship with his son, but he also wants to protect
his own safety and psychological well-being (in technical terminology: he identiﬁes with
his desire for a relationship with his son, but in this particular situation it conﬂicts with

his. desire for his own safety and psychological well-being, which he also identiﬁes with).

 

86 Evelyn Waugh, Decline and Fall, Chapter 6.

87 We might be curious about what causes the person to take the stand that they do, especially if the other
desires are ones that it would be in her best interest to stand by, and she realizes this. One explanation
might be that the person cannot do otherwise. Sometimes we are subject to what Frankfurt calls “volitional
necessities.” To illustrate the concept, he uses the example of Martin Luther declaring “Here I stand; I can
do no other.” Frankfurt says of Luther, “What he was unable to muster was not the power to forbear, but
the will. . .A person who is subject to volitional necessity ﬁnds that he must act as he does.” Frankﬁrrt, “The
Importance of What We Care About,” in The Importance of What We Care About (Cambridge University
Press, 1988), p. 87.

35

So, he commits to and stands behind his decision to not let his desire for a relationship
with his son to be effective in action. Of course the parent is still strongly inﬂuenced by
this desire for a relationship; he even insists that this decision does not mean that he does
not love his son, and so he is uneasy (residually ambivalent) with his choice.

The three cases presented illustrate the phenomenon of residual ambivalence: a
phenomenon where a person forms a will, but still experiences the inner conﬂict of
ambivalence due to the pressure of the other desires that she identiﬁes with. Although
Frankfurt does not draw the distinction between paralyzing ambivalence and residual
ambivalence as I have, nor does he focus on ambivalence within the set of desires that a
person identiﬁes with as I have, he does talk about the general compatibility of taking a
stand and experiencing conﬂict, which I think helps explain the phenomenon of residual
ambivalence.89 In addition, I will further discuss what it means to form a will despite
identifying with and still being drawn towards the opposing course of action in Chapter
Three when I discuss the compatibility of residual ambivalence and autonomous action.

Frankfurt explains:

“When someone identiﬁes himself with one rather than with another of his
own desires, the result is not necessarily to eliminate the conﬂict between those
desires, or even to reduce its severity, but to alter its nature. Suppose that a
person with two conﬂicting desires identiﬁes with one rather than with the
other. This might cause the other—the desire with which the person does not
identify—to become substantially weaker than it was, or to disappear
altogether. But it need not. Quite possibly, the conﬂict between the two
desires will remain as virulent as before. What the person’s commitment to the
one eliminates is not the conﬂict between it and the other. It eliminates the
conﬂict within the person as to which of these desires he prefers to be his
motive.”

 

88 George Orwell, 1984, Chapter 2.

89 Frankfurt, “Identiﬁcation and Wholeheartedness,” p. 172. Also see: Frankfurt, “Identiﬁcation and
Externality,” in The Importance of What We Care About (Cambridge University Press, 1988), p. 68.
Frankfurt, “The Faintest Passion,” p. 100. Frankfurt, “Identiﬁcation and Wholeheartedness,” pp. 172-174.

36

“Since it is most conspicuously by making a decision that a person identiﬁes with
some element of his psychic life, deciding plays an important role in the
formation and maintenance of the self. . ..The closest parallel among English
synonyms for “to decide” is the phrase “to make up one’s mind,”. . ..What
appears to be fundamentally common to all occurrences of the notion of making
something up is. . .the theme of creating an orderly arrangement. It seems to me
that in this light the closest analogue to a situation in which someone makes up
his mind is, rather surprisingly perhaps, a situation in which two people make up
their differences. People who do that after a quarrel pass ﬂom a condition of
conﬂict and hostility to a more harmonious and well-ordered

relationship. . .Moreover, people who have been quarreling may resolve harmony
between themselves even though their disagreement continues. Making up
concerns healing a relationship disrupted by conﬂict, and it has nothing directly
or necessarily to do with whether or not the conﬂict has ended.”

The experience of residual ambivalence has some similarities with and some
differences ﬂom the experience of regret. There is of course, a tinge of regret in many
cases of residual ambivalence; for example, in the case of the parent who chooses not to
have a relationship with his drug-addicted son. The regret, however, may not be regret
about the ordering of his desires; it is not that he regrets that he decided that he wanted
his desire for his safety to be effective in action instead of his desire for a relationship
with his son. The regret is that he cannot have both; the tinge of regret is directed
towards the world. The sort of classic, “I wish things could have been different,” line.

There is, however, a sense in which this seems too simple; for sometimes the
regret may be about the ordering of the desires. If the decision that the agent had to make
about the ordering of her desires is one that involved other persons then it seems too
simplistic to characterize the agent’s regret as simply a regret directed towards the world.

If Agamemnon chose (as he ﬁnally did) his army over his daughter, then it may be too

37

simplistic to describe his regret as just “I wish the world could have been different.” He
may regret doing wrong to Iphigenia. It may be more accurate to describe his regret as
directed towards her and not towards the world or the circumstances.
Causes of Ambivalence

Earlier I remarked that there is much to be answered about the phenomenon of
ambivalence; such as why the ambivalent person is both drawn towards and drawn away
ﬂom a desire that he has. This is akin to asking for the cause of ambivalence. In the case
of Agamemnon, he is drawn towards his desire to spare the lives of his cavalry because
he feels that he has a duty to them. He is drawn away ﬂom his desire to spare the lives of
his cavalry because he loves Iphigeneia. This case illustrates that a cause of ambivalence
is contingent facts about the world. It is a contingent fact about the world that
Agamemnon is living in that sparing the lives of his cavalry means killing his daughter.

Additionally, part of the reason why Agamemnon is both drawn towards and
away ﬂom his desire to kill Iphigeneia is because of, as Frankfurt puts it, “. . .deﬁning
elements of his own nature: his love for his daughter and his love for the army he
commands. ...”9° Cheshire Calhoun also argues that lack of wholeheartedness does not
necessarily signal some personal failure on the part of the agent to make up her mind
what she really wants. Calhoun points to Maria Lugones’s example of identifying both as
Latino and Lesbian and these being incompatible because of contingent facts about the
world.91

One may even dig deeper and examine why we (humans) are the sort of creatures

that are often ambivalent. We humans are often ambivalent—as children we feel

 

9° Frankfurt, “Autonomy, Necessity, and Love,” in Necessity, Volition, and Love (Cambridge University
Press, 1999) p. 139 footnote 8.

38

ambivalent towards our parents, as young adults we feel ambivalent about what to do
with our lives, as adults we feel ambivalent about careers, marrying, having children,
about our romantic partners, and our colleagues. I think that part of the reason is because
of our psychological complexity. As Phillip Koch notes in his work on ambivalence, as
humans we can psychologically adopt many points of view, or perspectives. We try to
organize these into one uniﬁed perspective, but it is always in ﬂux—it is breaking down
and being synthesized with new perspectives.92 Moreover, some perspectives and their
associated emotions are in the foreground and others are in the background. Koch
provides an example of a person walking towards some ﬂowers when the ground begins
to become very so ggy. The person does not forget about the ﬂowers, but they recede into
the background. As the cost of approaching them increases, the person starts to like them
less, and this is a drawing away. The person is both drawn towards and drawn away ﬂom
the ﬂowers.93 The complexity of human psychology also causes us to be able to
experience intense psychological suffering, and the more we experience this suffering,
the more ambivalent we may become. In Chekhov’s play “The Cherry Orchard,”
Madame Ranevsky (who is ambivalent about selling her cherry orchard) says, “What
truth? You can see what’s truth, and what’s untruth, but I seem to have lost the power
and vision; I see nothing. You settle every important question so boldly; but tell me,
Peter, isn’t because you’re young, because you have never solved any question of your
own yet by suffering? You look so boldly ahead; isn’t it only that you don’t see or divine

anything terrible in the future; because life is still hidden ﬂom your young eyes.”94

 

9' Calhoun, p. 239.

92 Koch, “Emotional Ambivalence,” p. 268.
93 Koch, “Emotional Ambivalence,” p. 268.
9’ Chekhov, “The Cherry Orchard,” p. 32.

39

Another possible cause of ambivalence is what Spinoza calls “imitation.”95

Imagine that one of my rivals is up for a university award. I have a desire for him to get
the award, but I am both drawn towards and away ﬂom my desire that he get the award.
I am drawn away ﬂom my desire that he get the award because his getting the award
gives him an edge over me. I am drawn towards my desire that he get the award because,
according to Spinoza, him getting the award resembles me getting the award (we are both
philosophers working on autonomy). Patricia Greenspan has expanded on Spinoza’s
theory of imitation. She has argued that ambivalence is partly due to imitation and partly
due to emotions resisting qualiﬁcation. If the ambivalence was about a judgment, then it
could be resolved. One might, for example, say, “There is a sense in which putting a
criminal in prison is a bad thing, and a sense in which it is a good thing, but overall, it is a
good thing.” Ambivalence as a phenomenon involving emotions and the will, however,
is not resolved in a similar manner because one cannot come to an all things considered
emotion. My ambivalence about my desire for my rival to get the award remains.96
At an even deeper level of causal analysis, Judith Farr Tormey has argued that the
cause of humans being ambivalent creatures, the cause of us being drawn towards and
drawn away ﬂom a desire or a course of action to the extent that we are indecisive about
it is our ﬂeedom. Tormey notes that ambivalence causes me to ask questions of myself—

“What should I do?” I expect some determinants about myself or my situation to give the

answer, but I experience their absence (because the two contradictory impulses cancel

 

9’ Patricia Greenspan, “A Case of Mixed Feelings: Ambivalence and the Logic of Emotion,” in Explaining
Emotions edited by Amelie Oksenberg Rorty (University of California Press 1980), p. 224.
9" Greenspan, “A Case of Mixed Feelings: Ambivalence and the Logic of Emotion,” p. 224.

40

each other out). I honestly do not know what I will do; I experience radical
freedom/indeterminacy.97

Finally, part of the cause of ambivalence may be the times and the society that we
- live in. Frankfurt makes this suggestion; however, he does not go into detail.98 This is a
major thesis of Andrew Weigert’s Mixed Emotions: Certain Steps Toward Understanding
Ambivalence. Wei gert argues that a cause of ambivalence is that modern society has
become increasingly pluralistic and complex.99 We have multiple goals, commitments
and values, and this makes it very likely that these will come into conﬂict. To see
Weigert’s point, consider the typical student entering college. She has goals of making
ﬂiends, of having new experiences, and of succeeding in her studies. She has
commitments and values tied to her traditional conservative family back home, and
commitments and values tied to her more liberal education. These aspects of her life will
often come into conﬂict, and she may be ambivalent about decisions that she is making
because of this tension. Weigert also notes how societal and professional roles shape
emotions, and given that many of these roles and rules have contradictory aspects, it is no

'00 Take for example, the role of a physician. One

surprise that ambivalence results.
aspect of this role is to be caring, yet another aspect of this role is to be detached. In her
practice, the physician may sometimes behave in a caring, personal way towards a

patient, and at other times behave in a detached way. Through her behavior, the

physician begins to develop feelings of empathy and also feelings of detachment towards

 

97 Judith Farr Tormey, “The Ambivalent Self” in Philosophic Exchange: Annual Proceedings Vol. 14,
1983, pp. 112-113. Note that “freedom” here is understood as “indeterminacy,” which is of course not
what Frankfurt means by it.

98 Frankfurt, “The Faintest Passion,” p. 99.

99 Andrew Weigert, Mixed Emotions: Certain Steps Toward Understanding Ambivalence (State University
of New York Press 1991), p. xiv.

'00 Weigert, Mixed Emotions: Certain Steps Toward Understanding Ambivalence, p. 14, p. 22.

41

her patients. As a result, she may begin to often feel ambivalent about her behavior
towards her patients.
Ambivalence and the Importance of the Decision

Some of the things that we are ambivalent about matter more than others. I may
be ambivalent, even paralyzingly ambivalent, about how to spend my money; but the
decision about how to spend my money is not as important as the decision about which
person (if any) I should marry. In his book, The Theory and Practice of Autonomy, Carl
Schneider points out how many choices we are faced with: “We walk down an aisle full
of breakfast cereals; national catalogs complete with local stores to sell us shirts and
socks; we have to choose not only a long-distance phone company, but a cellular phone
service and soon an electric utility. But these consumer choices are just the beginning.
People make decisions about the most basic aspects of their lives. All afﬁnities are
elective. Your “lifestyle” is yours to select, your religion is yours to reject, your family
yours to deﬁne. Even commitments like marriage are “nonbinding,” open to daily
reconsideration.”'°l Notice how the ﬁrst set of choices that Schneider mentions pales in
comparison to the choices about religion and marriage. Some of our choices are less
important then others. And, important choices are more prone to generate ambivalence
than ones that we consider to be unimportant. For example, the choice of which person
(if any) I should marry is more prone to generate ambivalence in me than the choice of

which pair of socks to buy. '02 Moreover, of the choices that are important to us, some are

 

'0' Schneider, 174.

'02 Because different people consider different things to be important, of two different persons facing the
same choice, one may be ambivalent about it and the other may not. It may well be the case that there is
someone who considers which pair of socks to buy more important than which person (if any) to marry.
So, even though her and I are faced with same choice of which pair of socks to buy, she will be ambivalent
about it whereas I will not. And even though her and I are faced with the same choice of which person (if
any) to marry, I will be ambivalent about it whereas she will not.

42

identity-conferring and some are not. As Lynn McFall points out, some of our principles
or commitments are “. .. identity-conferring commitments: they reﬂect what we take to be
most important and so determine, to a large extent, our (moral) identities.” And, I would
argue, it is worse for a person (worse for her) to be ambivalent (or otherwise non-
autonomous) about commitments that she considers to be identity-confening. As
Frankfurt notes, “. . .to remain persistently ambivalent concerning issues of substantial
importance in the conduct of life is a signiﬁcant disability?”03
Ambivalence and Information

One might ask whether or not ambivalence, either paralyzing or residual, is really
just a problem of lack of information. Oscar is ambivalent about which car to buy. He
has a desire to buy car X. He is drawn towards his desire to buy car X because car X is a
good car, but he is also drawn away ﬂom his desire to buy car X because he also wants
car Y and cannot have both. He experiences paralyzing ambivalence and does not form a
will about which desire he wants to move him to action. But suppose that Oscar knew
that car Y had an exhaust problem that was going to be costly. Then, Oscar would not be
ambivalent. Also consider the case of Winston in 1984. Winston experiences residual
ambivalence about his desire to meet Julia. He has decided to meet Julia, and stands by
the desire to meet her (he wills it), but still experiences conﬂict because of his desire for
safety. But suppose that Winston learned that meeting Julia was completely safe. Then,
Winston would not experience any residual ambivalence.

I will talk much more about ambivalence and information in Chapter Four, but for

now I have two responses. First, additional information will not always help resolve

ambivalence. Second, additional information may sometimes make the ambivalence

 

'03 Frankfurt, The faintest passion, 102.

43

even worse. Frankfurt notes, “Since ambivalence is not a cognitive deﬁciency, it cannot
be overcome merely by acquiring additional information.”104 To illustrate Frankfurt’s
point, consider the example of Agamemnon. There is no piece of information that could
alleviate his paralyzing ambivalence. Often, cases of ambivalence are like this. Consider
the case of someone who takes a job in New York for a year. During that year she
becomes very involved with a charity for cats in NYC. She is both drawn towards and
away ﬂom her desire to continue to be involved in this charity. She is drawn towards it
because she likes working with this charity, but she is drawn away ﬂom it because she
knows that she is going to have to return to LA. after one year, and she knows that it is
going to be painful because she will have grown attached to this charity. Whether or not
her ambivalence is paralyzing (she does not form a will), or residual (she wills the
continuation of the charity work despite continued contrary desires), no additional
information would resolve the ambivalence. That further knowledge does not necessarily
resolve ambivalence is supported by the ﬁndings of psychologist Richard Petty who
found that attitudes change in response to new information but old attitudes can remain
and inﬂuence behavior.105 Even with new information and new attitudes, old conﬂicting
attitudes may remain and cause the person to remain ambivalent.

Not only does more information often not resolve ambivalence, it sometimes
makes it worse. Consider again the case of Winston in 1984. Suppose that Winston

learned that it is even more dangerous to know Julia than he had ﬁrst thought. This

 

’04 Frankfurt, “The F aintest Passion,” p. 100. Note that Frankfurt’s claim that ambivalence “cannot be
overcome with more information” is too strong. It sometimes can, as my examples in the preceding
paragraphs illustrate. It would be more correct to say that ambivalence “cannot always be overcome with
more information.”

'05 Richard Petty, “Implicit Ambivalence ﬂom Attitude Change: An Exploration of the PAST Model,”
Journal of Personality and Social Psychology Vol. 90 No. 1 2006, pp. 21-41.

44

knowledge would actually increase his residual ambivalence. Even if his decision and his
will do not change, he may constantly be questioning himself on whether he has made the
right choice. Or, to return to the paralyzing ambivalence of Agamemnon; suppose that
Agamemnon learned that the death that his daughter would suffer would be particularly
painful. This knowledge might actually increase his paralyzing ambivalence in that he
would feel even more strongly conﬂicted ﬂom a subjective point of view. As Weigert
noted, “Rationality won’t resolve ambivalence, actually it generates it.”106

Someone might object that cases where a person’s ambivalence can be resolved
by additional propositional information are not true cases of ambivalence to begin with.
Consider an analogy where someone objects that cases where cultural relativism can be
explained by a disagreement about facts are not true cases of cultural relativism: ‘07
Someone cites an example of cultural relativism where a norm of person A’s culture is
that it is morally impermissible to eat cows and a norm of person B’s culture is that it is
morally permissible and even desirable to eat cows. But suppose that we discover that
the reason that person A’s culture believes that it is morally impermissible to eat cowls is
because they believe that the souls of their ancestors are reincarnated in cow’s bodies.
Person B’s culture agrees with the normative judgment that one should not kill and eat
one’s ancestors; they just disagree with what person A’s culture sees as a fact (that
ancestors take the form of cows). So, person A and person B’s disagreement is not truly
a case of cultural relativism.

Although it does make sense to describe person A and person B’s disagreement as

not a true case of cultural relativism (because as it turns out they have the same normative

 

'06 Weigert, Mixed Emotions: Certain Steps Toward Understanding Ambivalence, p. 165.
'07 I thank Tamra Frei for raising this helpful objection and accompanying analogy.

45

belief about not eating ancestors); the case of ambivalence is different. I have deﬁned
ambivalence as the feeling of internal conﬂict that arises when a person is drawn both
towards and away ﬂom an object, a desire, or a course of action. The very deﬁnition of
ambivalence focuses on the subjective experience of the person—the person is drawn
both towards and away ﬂom something and experiences internal conﬂict as a result of
this tension. To argue that cases where a person’s ambivalence can be resolved by
additional pr0positional information were not true cases of ambivalence to begin with is
akin to arguing that cases where a person’s anger can be resolved by additional
propositional information were not true cases of anger to begin with. For example, a
person is angry with a ﬁiend for not meeting her for dinner when she promised. The
person then ﬁnds out that her friend did not meet her for dinner because she was involved
in a serious car accident. She is no longer angry. Now, it would not make sense to say
that this was not a true case of anger to begin with, and similarly it does not make sense
to say that cases where ambivalence subsides with the introduction of more information
were not true cases of ambivalence to begin with.
Ambivalence and Action

One remaining issue is whether or not ambivalence and action are compatible.
Certainly residual ambivalence and action are compatible. Pennyfeather acted despite his
residual ambivalence, Winston acted despite his residual ambivalence, and we act despite
our residual ambivalence all of the time. The more interesting question is whether or not
action is possible in cases of paralyzing ambivalence. I would argue that acting and
paralyzing ambivalence are not necessarily incompatible. Despite the fact that the

paralyzingly ambivalent person is so conﬂicted about which desire she wants to move her

46

to action that she does not form a will; she can still perform the act of leaping in one
direction or the other (I will discuss this leap extensively in Chapter Four). To deny this
possibility, one would have to maintain that a donkey that is equidistant ﬂom two
equivalent piles of hay would have to starve to death (this example comes ﬂom
“Buridan’s paradox,” which I will discuss in Chapter Four). Surely the donkey is not
going to starve to death. Despite his being so conﬂicted about which desire he wants to
be effective in action that he does not form a will (pretend for a moment that donkeys can
form wills), he is going to eat one or the other piles of hay.

Someone may object to my claim that the paralyzing ambivalent person can act
by pointing out that the paralyzing ambivalent person does not have a will, and then
arguing that acting without a will is impossible. I would reply that acting without a will
is possible—children, animals, and even normal adult humans (when they are acting
unreﬂectively) do it all of the time. My opponent may respond that children, animals,
and unreﬂective adults behave in a habitual/reﬂexive way; they do not act. Once a
person has started to go down the intentional/reﬂective road of action, they need a will to
act. I would respond with the counter-example of the wanton. The wanton is engaging in
the intentional/reﬂective road of action—he is thinking about which of his ﬁrst order
desires he wants to be effective in action—but acts without a will (he does not care which
of them actually move him to action) nonetheless.

My arguments are intended to defeat the claim that it is impossible for the
paralyzingly ambivalent person to act. But, I do not mean to argue that it is always
possible for the paralyzingly ambivalent person to act. It depends on individual human

psychology. A person may feel as if they are literally physically paralyzed and cannot

47

leap in one direction or the other. For example, if both courses of actions are ones that
the agent repudiates, she may be unable to bring herself to leap towards either course of
action. As F rankﬁrrt has argued, there are some things that we are simply unable to bring
ourselves to do; we are unwilling to have our wills shaped in a particular way.108
Indifference and Temptation

Ambivalence is not to be confused with indifference. Consider the following

quote by the character Meursault in Albert Camus’s The Stranger:'09

"She was wearing a pair of my pajamas with the sleeves rolled up. When she
laughed I wanted her again. A minute later she asked me if I loved her. I told
her it didn't mean anything but that I didn't think so. She looked sad. But as
we were ﬁxing lunch, and for no apparent reason, she laughed in such a way
that I kissed her."

Meursault does not reject his desire for Marie, nor does he endorse it. He is not
ambivalent, for he is not experiencing strong feelings drawing him in conﬂicting
directions. He is what we might call, indifferent. To be indifferent to one’s own motives
is to, as Frankfurt says, “...take no evaluative attitude toward the desires that incline him
to act. If there is a conﬂict between those desires, he does not care which of them proves
to be the more effective. In other words, the individual does not participate in the
conﬂict.”1 ‘0 This type of indifference we might call, borrowing partly ﬂom Frankfurt,
“wanton indifference.”l l 1

Another type of indifference might be someone who does not even have ﬁrst

order desires that incline him to act one way or the other. There is no conﬂict between

 

'08 Frankfurt, “Rationality and the Unthinkable,” pp. 181-183. Frankfurt argues that these are necessities of
the will, and while it may seem as if they restrict the ﬂeedom of our will, actually, it is when we act in
accordance with them that our will is ﬂee. For an extensive discussion of this see Frankfurt, “Getting It
Right,” in Taking Ourselves Seriously and Getting It Right (Stanford University Press 2006).

'09 Albert Camus, The Stranger, Part 1 Chapter 4.

”0 Frankfurt, “Identification and Wholeheartedness,” p. 164.

“' Frankfurt calls this person a wanton in “Freedom of the Will and the Concept of a Person,” p. 18.

48

the desires that incline him to act (ﬁrst order desires) because he does not even have
inclinations one way or the other. Imagine that you have to decide whether to spend the
day at the beach or the park—those are your only two options. You are indifferent about
where you spend the day, in the sense that you do not have a ﬁrst order desire inclining
you to spend the day at the beach, nor do you have a ﬁrst order desire inclining you to
spend the day at the park. This type of indifference we might call “ﬁrst order
indifference”. Ambivalence is also not to be confused with temptation. Ambivalence
occurs when either a person cannot decide on which desire (of the desires that she does
identify with) she wants to be the one that moves her to action, or when she does decide
but still feels pulled by the other desires that she identiﬁes with. Temptation occurs when
a person has decided on viewing a desire that she has as outlaw, but it continues to
inﬂuence her; it tempts her.112 For example, I decide that I do not identify with my ﬁrst
order desire to eat ice cream for dinner once again, but the desire to eat ice cream for
dinner continues to inﬂuence me; it tempts me. This is different ﬁom me not being able
to decide on whether or not to identify with my ﬁrst order desire to eat ice cream for
dinner, or not being able to decide whether I want that desire to be the one that moves me
to action tonight.

In closing Chapter Two, let me note what I hope to have accomplished. I have

provided both a ﬁrst person and a conceptual analysis of the phenomenon of

 

”2 This is, perhaps, a specialized use of the term ‘temptation’ in that we may think that temptation need not
occur only ﬁ'om desires that a person repudiates (does not identify with, views as bad, evil etc.).

Sometimes the way that the term is used leaves open the possibility that a person can be “tempted” by the
good, or by other desires that she identiﬁes with (desires lower in the ordering of the desires she identiﬁes
with). In my use of the term ‘temptation’ I am describing a phenomenon in which a person is drawn
towards desires that she repudiates, does not identify with, views as evil etc. in order to distinguish it from
the phenomenon in which a person is drawn towards other desires that she identiﬁes with even though she
has committed to those not being the ones that move her to action at a particular time (she has committed to
willing a different desire).

49

ambivalence. I have drawn a distinction between paralyzing ambivalence (a phenomenon
in which a person is both drawn towards and away from a ﬁrst order desire that she has,
such that she does not form a will), and residual ambivalence (a phenomenon in which a
person is both drawn towards and away from a ﬁrst order desire that she has, but has
formed a will despite this conﬂict). I have also argued that most ambivalence occurs not
in deciding whether or not to identify with a desire (as Frankﬁirt characterizes it), but in
deciding on the ordering of the desires that the person identiﬁes with; in deciding which
of her desires she wants to be effective in action at a particular time. Aﬂer explaining
some of the possible causes of ambivalence, I discussed the difference between being
ambivalent about identity conferring commitments and ones that are less important, I
argued that ambivalence is not simply a problem of lack of information, and I analyzed
the relationship between ambivalence and action, arguing that paralyzing ambivalence
and action are not incompatible. In the next chapter, I will argue that cases of paralyzing
ambivalence result in a loss of metaphysical and bioethical autonomy, and in Chapter
Four, I will propose methods to resolve paralyzing ambivalence that preserve an agent’s

autonomy.

50

Chapter Three: The Compatibility of Autonomy and Ambivalence

In Chapter One I drew a distinction between bioethical autonomy and
metaphysical autonomy, but I also discussed the ways in which metaphysical theories of
autonomy have inﬂuenced the granting of bioethical autonomy. In Chapter Two, I
provided a conceptual analysis of the phenomenon of ambivalence, arguing that there are
two types of ambivalence: paralyzing and residual. In this Chapter, I will argue that it is
the cases of paralyzing ambivalence that result in a loss of metaphysical autonomy, and
hence bioethical autonomy. The cases of residual ambivalence, however, do not result in
a loss of metaphysical autonomy, nor of bioethical autonomy, despite the fact that
contemporary autonomy theories imply that it does.

Paralyzing Ambivalence and Metaphysical Autonomy

In Chapter One, I discussed the various conceptions of metaphysical autonomy.
Conceptions of metaphysical autonomy either focus on identiﬁcation or coherence.
Within the identiﬁcation camp there are many different theories of identiﬁcation; but
despite their differences, all of the various conceptions are in agreement that paralyzing
ambivalence results in a loss of autonomy. Coherence conceptions of autonomy preclude
paralyzing ambivalence as well.

For example, Frankfurt’s theory of autonomy requires that a person identify with
the desire that moves her to action; but the paralyzed ambivalent person cannot settle on
whether or not she does identify with that desire, or on whether or not she wanted that
desire to be the one that moves her to action at a particular time. Consider an example of

a person who is paralyzingly ambivalent about getting married. She has a ﬁrst order

51

desire to marry, but she also has a ﬁrst order desire to devote herself entirely to her
career. She identiﬁes with both desires, and as a result is both drawn towards and away
from marriage. She cannot settle on which desire she wants to move her to action—her
desire for marriage or her desire for a career. She has not formed a will, and so she is not
acting on a desire that matches her will (she is not acting on any desire at all). Hence, she
is not autonomous, on Frankfurt’s account.

Bratman’s account of autonomy also precludes the paralyzed ambivalent person
ﬁ'om being autonomous. To be autonomous with respect to a desire, a person must
decide to treat it as reason-giving and be satisﬁed with that decision. And, “One is
satisﬁed with such a decision when one’s will is, in relevant ways, not divided: The
decision to treat as reason-giving does not conﬂict with other standing decisions and
policies about which desires to treat as reason-giving.”l ‘3 Consider the case of the woman
who is ambivalent about getting married. Imagine that she decides to treat the desire to
get married as reason-giving (she treats it as end-setting). She does this because she has a
higher order, non-instrumental policy in favor of it functioning as end-setting (perhaps
something to do with her religious beliefs). She is not, however, satisﬁed with her self- -
governing policy; it is challenged by another of her other self-goveming policies.114 She
has a feminist self-governing policy that challenges her religious self-governing policy.
The conﬂict between the two is causing her ambivalence about marriage. Because she is
ambivalent and not undivided about her decision to treat the desire for marriage as

reason-giving, she is not autonomous with respect to that desire.

 

”3 Bratman, “Identiﬁcation, Decision, and Treating as a Reason,” p. 201.
114 For Bratman’s account of satisfaction see “Reﬂection, Planning, and Temporally Extended Agency” in
The Philosophical Review Vol. 109, 2000 pp. 49-50.

52

Christman’s historical account of autonomy precludes the paralyzed ambivalent
person from being autonomous as well. For Christman, a person is autonomous with
respect to a desire or decision if she did not resist the development of the ﬁrst order
desire when attending to the process of development. Or (in cases where she did not
attend to the process of development), she would not have resisted the development had
she attended to the process. But, in cases of ambivalence the person does sometimes
resist the development of the ﬁrst order desire. Return to the case of the woman who is
ambivalent about getting married; at the times when she is drawn away ﬁ'om the desire to
marry she has struggled against its development.

In Christman’s 2001 revised account of autonomy (autonomy as non-alienation)
there is an indication that the paralyzed ambivalent person could be autonomous.
Christrnan says, “Non-alienation is also a different condition from the familiar
requirement of identiﬁcation, which one typically ﬁnds in discussions of autonomy. On
the one hand, I can feel no alienation toward a characteristic but not fully identify with it,
in the sense of wholehearted endorsement without regret. We all contain some measure
of internal conﬂict and complexity, and an attitude of ironic acceptance of the tensions of
our own psyches is inevitable and perhaps healthy in a multidimensional and perplexing
world.”1 15 While this account may be tolerant of some ambivalence, it is not tolerant of
the sort of paralyzing ambivalence seen in the case of Agamemnon, or in the case of the
woman who is ambivalent about marriage. Christman’s non-alienation account of
autonomy requires that one not feel any alienation towards a characteristic, act, or desire,

in order to be autonomous with respect to it. But the ambivalent person does feel

 

”5 John Christman, “Liberalism, Autonomy, and Self-Transformation” in Social Theory and Practice Vol.
27, 2001, p. 202.

53

alienation towards the characteristic, act, or desire in question. She feels both alienated
from it and drawn towards it.

Paralyzing ambivalence is also precluded on coherence accounts of autonomy.
The division of ambivalence is almost the exact opposite of cohesion. Ekstrom says,
“Inner turmoil, disharmony in the self, indicates lack of autonomy.”l '6 Ekstrom does
admit that we are all conﬂicted to some extent and that it is difﬁcult to achieve cohesion
among the elements of one’s inner life; but she certainly would not consider someone
who is paralyzingly ambivalent to be autonomous. Moreover, she argues that it is only
when we act ﬁ'om the non-conﬂicting, cohering parts of our self that we are
autonomous.117 Coherence theorists like Ekstrom could possibly allow for someone who
is paralyzingly ambivalent to be autonomous by arguing that being autonomous is not a
matter of how one feels, it is a matter of whether or not preferences and desires that
motivate action cohere. Hence, Agamemnon could feel conﬂicted to a paralyzing extent,
and at one moment be swept away by one of the desires such that it moves him to action,
but be autonomous so long as it happened that the desire that moved him was one that (as
a matter of fact) cohered with other preferences that he had. But, this is an ad hoc way
for coherence accounts to allow for paralyzing ambivalence; for the fact that the person is
acting on a desire that coheres with her other preferences is merely accidental. Thus,
coherence accounts would need to add the condition that the coherence between the
desire that moves the person to action and the person’s other preferences be an intentional

and not accidental coherence. With this added condition, the paralyzed ambivalent

 

”6 Ekstrom, “Autonomy and Personal Integration,” p. 147.
”7 Ekstrom, “A Coherence Theory of Autonomy,” p. 602.

54

person cannot be considered autonomous on coherence accounts of autonomy, for her
will is divided and not coherent.
Paralyzing Ambivalence and Bioethical Autonomy

Thus far I have argued that the various conceptions of metaphysical autonomy all
converge on the point that paralyzing ambivalence results in a loss of autonomy. Recall
that in Chapter One I argued that there are certain cases where decision making capacity
and hence bioethical autonomy are gone because metaphysical autonomy is gone. Two
such cases that have been discussed in the bioethics literature are young children and
some psychiatric patients. Both often lack decision making capacity and hence the
autonomy to make their own medical decisions because they have disorders of the will
that block them from being metaphysically autonomous. Young children are not yet
developed enough to form a will, and some psychiatric patients are too unstable or
delusional to form a will. There is in addition a third class of patients who suffer from a
disorder of the will called paralyzing ambivalence, that (as I have argued above) blocks
metaphysical autonomy. This class of patients has not been discussed in the bioethics
literature, but they clearly lack decision making capacity as they lack the ability to make
and communicate their decisions. To see this, consider the real life case of Mr. X
(introduced in the preface of the dissertation).

Recall that the overall trend in discussions about autonomy in the metaphysical
context is a focus on a person’s internal states. And, the overall trend in discussions
about autonomy in the bioethical context is a focus on a person’s understanding of the
world. But, I argued that there are cases where bioethical autonomy is gone (the ability

and permission to make one’s own decisions is gone) because metaphysical autonomy is

55

gone (there is a disorder of the will). Paralyzing ambivalence is one of those cases. In
the preface of my dissertation, I introduced the case of Mr. X. Mr. X, a youthful 70 year
old man, was involved in a head on motor vehicle accident. As a result, he lay in a
hospital bed unconscious, on a ventilator, and paralyzed from the neck down. Mr. X’s
wife and adult son and daughter all agreed that there was no way that Mr. X would want
to live this way. In fact, they pointed to Mr. X’s advance directive to afﬁrm this view.
Shortly thereafter, Mr. X regained consciousness. To everyone’s surprise, Mr. X asserted
that he wanted to remain on the ventilator; he wanted to stay alive. The following day,
we went in to talk with Mr. X and he asserted that he did not want to remain on the
ventilator; he wanted to die. The family, the physicians, and the ethicists talked to Mr. X
to try to ascertain what he really wanted. These conversations spanned over weeks, but
Mr. X did not know what he really wanted. Mr. X could not decide which of his desires
(his desire to live or his desire to avoid living a life paralyzed and on a ventilator) he
wanted to be effective in action; he was paralyzingly ambivalent.

Because of Mr. X’s paralyzing ambivalence, he lacked decision making capacity
and was not granted bioethical autonomy. The usual protocol for such cases is to attempt
to restore decision making capacity to the patient, or if that cannot be done, to consult
advance directives or next of kin. In Chapters Four and Six, I will discuss various
methods for restoring decision making capacity to paralyzingly ambivalent patients such

as Mr. X.

56

Residual Ambivalence and Metaphysical Autonomy

Unlike the paralyzed ambivalent person, the residually ambivalent person has
formed a will; she stands by one of her desires being the one that moves her to action,
despite the fact that she still experiences contrary desires (that she also identiﬁes with).
Broadly speaking, on identiﬁcation accounts of autonomy, the residually ambivalent
person should count as autonomous (so long as the one that she stands by is the one that
moves her to action), for she has identiﬁed with the desire that moves her to action.
Narrowly speaking, however, the various accounts of identiﬁcation seem to differ on
whether or not a residually ambivalent person is autonomous. Some accounts have very
narrow criteria for identiﬁcation, such that they preclude the experience of any
ambivalence, division, or conﬂiction.

For example, sometimes Frankfurt writes as if identiﬁcation requires a complete
absence of conﬂict. Frankfurt argues that to wholeheartedly identify with a ﬁrst order
desire is to be fully satisﬁed with it causing attitudes and behaviors, and that full
satisfaction entails an absence of restlessness or resistance.118 He writes, “Being or
becoming satisﬁed is like being or becoming relaxed.”1 ‘9 Similarly, sometimes
Christrnan writes as if identiﬁcation (with the process by which the desire was formed)
requires a complete absence of conﬂict. Christrnan argues that the reﬂection on the
process by which one comes to have desires must be minimally rational; meaning that
there cannot be “manifestly inconsistent” desires or beliefs (ones that are easily brought

120

to consciousness and recognized as incompatible). Christrnan says, “If the ‘self’ doing

the ‘governing’ is dissociated, fragmented, or insufﬁciently transparent to itself, then the

 

“8 Harry Frankfurt, “The Faintest Passion,” 103-104.
”9 Harry Frankfurt, “The Faintest Passion,” 105 footnote 16.

57

process of self-determination sought for a concept of autonomy is absent or
incomplete.”121 Similarly, sometimes Bratman writes as if identiﬁcation requires a
complete absence of conﬂict. Recall that for Bratman, identifying with a desire involves
a decision to treat it as reason-giving, and being satisﬁed with that decision. Bratman
says, “One is satisﬁed with such a decision when one’s will is, in relevant ways, not
divided: The decision to treat as reason-giving does not conﬂict with other standing
decisions and policies about which desires to treat as reason-giving.”122
I would argue that any account of autonomy ought to allow for residual
ambivalence, and so identiﬁcation accounts of autonomy should conceptualize
identiﬁcation in a way that allows for some ambivalence. The argument for this has two
premises: (1) autonomy theorists have claimed to be offering an account of autonomy as
a property that is empirically possible and fairly commonly possessed by persons, and (2)
residual ambivalence is common and so must be accounted for. With respect to premise
one, Gerald Dworkin argues, “There should be no empirically grounded or theoretically
derived knowledge which makes it impossible or extremely unlikely that anybody ever
has been, or could be, autonomous.”123 Nomy Arpaly seconds this criterion. She says,

"...agent-autonomy, or at least a substantial degree of it, is supposed to be the default

condition of the average adult human being...", and “. . .caution must be exercised when

 

12° Christman, “Autonomy and Personal History,” p. 15.
12' Christrnan, “Autonomy and Personal History,” p. 17.
'22 Michael Bratman, “Identiﬁcation, Decision, and Treating as a Reason,” pp. 200-201.

‘23 Gerald Dworkin, The Theory and Practice of Autonomy,, p. 9.

58

mixing intuitions about autonomy as an ideal condition and intuitions about autonomy as
a property that a great majority of humans have, to a large extent, most of the time.”124
Premise two is that residual ambivalence is common. We humans are often
ambivalent—as children we feel ambivalent towards our parents, as young adults we feel
ambivalent about what to do with our lives, as adults we feel ambivalent about careers,
marrying, having children, about our romantic partners, and our colleagues. We are often
unsure, hesitant, of two minds, uncertain, anxious, or conﬂicted about things. Even if we
make up our minds about something, we often still have desires and feelings in opposing
directions; and these desires in the other direction are not necessarily ones that we view
as outlaws. As Philip Koch writes, “It’s really astonishing how feelings, or feeling-
shadows, endure: an old photo, a phrase, a stream of light striking us just so—and
suddenly the old longing is ﬂesh and urgent. One realizes that it has lain there always,
though covered with layers of brush and dry leaves.”125
Hence, identiﬁcation accounts of autonomy should conceptualize identiﬁcation in
a way that allows for residual ambivalence. What then would identifying with a desire
and willing that it be effective in action at a particular time mean in these cases? It
cannot mean that we feel completely relaxed about it, or that we feel there are no further
questions to be asked, or that we are not ﬁ'agmented or divided. I think that the best way
to conceptualize it is to return to Frankfurt’s original notion of identiﬁcation as decision

and commitment. To will that one desire rather than another be effective in action at a

particular time is to decide that that one be effective in action rather than the other one.

 

'24 Nomy Arpaly, “Responsibility, Applied Ethics, and Complex Autonomy Theories” in Personal
Autonomy: New Essays on Personal Autonomy and Its Role In Contemporary Philosophy edited by James
Stacey Taylor (Cambridge University Press 2005), p. 124.

'25 Koch, “Emotional Ambivalence,” p. 274.

59

But, what does it mean to decide that desire X be effective in action at t1? It means that
the person makes a commitment to that desire being effective in action at t]. Now, we
have to ﬁgure out what making a commitment to that desire means. What it cannot mean
is that you feel as if there is no room for questions, or that nothing could change your
mind; for you are still strongly drawn towards a conﬂicting course of action.126 That’s
not the sort of commitment that it is. The commitment that one makes to a certain desire
being effective in action when they are residually ambivalent is more like a resignation
followed by action. Imagine the following example: Sally and John are engaged to be
married and have a young child together. John was raised Catholic, but has not seriously
practiced his religion since his teens. Lately however, John has been rekindling his
relationship with the Church and over the past few months he has begun to experience a
deep feeling that he is being called to become a deacon. Now, a deacon is allowed to
retain a wife if he already has one, but if he is not yet married he is not permitted to do
so. John has a desire to marry Sally and share a life with her and their daughter, but he
also has a desire to become an ordained deacon. Both desires are ones that he identiﬁes
with in that he in some sense endorses them as being among the inﬂuences on his
behavior. Aﬂer some time and inner turmoil, John makes a decision and forms a will.
He commits to his desire for a life with Sally and their daughter being the one that is
effective in action. Meaning, he has resigned himself to that course of action and has
began to take steps towards it. Now, taking steps towards it can mean small and gradual
steps. For example, whenever John is feeling like getting on the webpage for deacons he

redirects his thoughts to something else. Whenever he begins to miss his deep

 

'26 Frankfurt, “Freedom of the Will and the Concept of a Person,” pp. 21-22.

60

involvement with the Church, he reminds himself of what he would be losing in pursuing
that option. John begins therapy with his ﬁancée to make their relationship even better
and more attractive. John is however residually ambivalent. He is still drawn towards
(and at times strongly drawn towards) his desire to become a deacon, especially when he
drives by a church or sees a book on Catholicism while he is browsing at his local
bookstore.

Despite his residual ambivalence, John is still autonomous with respect to his
decision to marry Sally. He has formed a will (he has resigned himself to a certain order
of his desires and taken steps towards making the desire that he has ordered as primary
effective). Moreover, John is not acting on a desire that he views as outlaw, he is acting
on a desire that he identiﬁes with. It is true that he prefers to be in a different situation—
one where he could act on both desires—or perhaps one where he had never met Sally
and could act on his deacon desire. But the key to John being metaphysically
autonomous with respect to his choice to marry Sally is that the desire to marry her is not
an outlaw and John wills to marry her despite his residual ambivalence.

Coherence theories of autonomy can and should allow for residual ambivalence as
well. On Ekstrom’s coherence theory, a person acts autonomously when the cause of her
action is a preference that Was un-coercively formed and that coheres with her other
preferences and beliefs.127 The set of preferences that cohere (“authorized preferences”)
make up the true self, and it is when we act from our true self that we act autonomously.
So long as a person is acting intentionally from an authorized preference, she is

autonomous. It does not matter if she still experiences being drawn away from (in

61

addition to being drawn towards) that preference. In fact, it is likely that she will
experience being drawn away from (in addition to being drawn towards) the preference
because she still has other conﬂicting preferences that will cause her to feel somewhat
drawn away ﬁ'om her authorized preference. The focus should be that despite her feeling
both drawn towards and drawn away, she has taken a stand about which desire she wants
to move her to action; and so long as that desire coheres with her other authorized
preferences, she is autonomous with respect to it.
Residual Ambivalence and Bioethical Autonomy

Thus far I have argued that the various conceptions of metaphysical autonomy can
and should allow for residual ambivalence. To carry the point over to thinking about
bioethical autonomy, even if a patient is residually ambivalent about a decision that they
have made, they still retain bioethical autonomy and their decision should be followed.
This matches our intuitions. Imagine that (returning to our case study) Mr. X resolves his
paralyzing ambivalence and decides that he wants to remain alive on the ventilator. He
has resigned himself to this course of action and taken action by telling the doctors and
his family that this is what he wants to do. He has also requested that he be connected
with a counselor and with groups for paraplegics that he can contact when he starts to feel
like giving up. But, he is residually ambivalent about this decision; he feels uneasy about
it and he still often feels strongly inﬂuenced by his desire to just let go. I imagine that
many of our intuitions tell us that despite the residual ambivalence, Mr. X’s decision
should be followed. Not only should it be followed, but it should also be respected in the

sense that we should not harass Mr. X with continuous conversations about his decision.

 

'27 Laura Waddell Ekstrom, “Autonomy and Personal Integration,” in Personal Autonomy: New Essays on
Personal Autonomy and Its Role in Contemporary Moral Philosophy edited by James Stacey Taylor

62

Now, someone might object that the fact that we are inclined to permit Mr. X’s
decision does not necessarily illustrate that we think that he is autonomous. There may
be other explanations for why we are inclined to permit Mr. X’s decision; for example we
may think that it is in Mr. X’s best interest to remain alive on the ventilator. But if we
change the example to one where Mr. X’s decision is one that we do not think is in his
best interests (e. g. he decides to just let go), I argue that we would still be inclined to
permit his decision—illustrating our recognition that autonomy and residual ambivalence
are compatible.

In conclusion, paralyzing ambivalence results in a loss of metaphysical autonomy
and bioethical autonomy. We should attempt to restore bioethical autonomy to a patient
who lacks it, and in Chapters Four and Six I will discuss ways to restore metaphysical
and bioethical autonomy to a person who is paralyzingly ambivalent. Residual
ambivalence, on the other hand, does not result in a loss of metaphysical autonomy or
bioethical autonomy. Autonomy theorists need to be clearer about their position on
residual ambivalence. I have suggested that both identiﬁcation and coherence accounts
of autonomy can and should allow for residual ambivalence, and I have suggested a way

to conceptualize identiﬁcation in cases of residual ambivalence.

 

(Cambridge University Press, 2005), pp. 148-152.

63

Chapter Four: Resolving Paralyzing Ambivalence

Lopakhim: You must make up your mind once and for all. Time waits for no man. The question is
perfectly simple. Are you going to let off the land for villas for not? Answer in one word; yes or no. Only
one word!

Trophimof: Whether the property is sold today or whether it’s not sold, surely it’s all one. . .You mustn’t
deceive yourself any longer; for once you must look the truth straight in the face.128

In Chapter Three, I argued that paralyzing ambivalence results in a loss of
metaphysical and hence bioethical autonomy. Thus, resolving paralyzing ambivalence is
a necessary condition for restoring metaphysical and bioethical autonomy. '29 In this
Chapter, I examine possible responses and resolutions to paralyzing ambivalence, paying
close attention to whether they promote autonomy or not.. I aruge that in cases where the
cause of the paralysis is an epistemological deﬁcit (i.e., the person cannot decide which
desire she wants to be effective in action at a particular time because she is either missing
information or she has not reﬂected enough on how each desire coheres with her other
values, desires, and preferences), then the response should be to help the person come to
will one course of action or the other by providing the relevant available information and
by helping her to reﬂect on which course of action best coheres with her other values and
desires. In cases where that does not work, or where the cause of the paralysis is
contingent facts about the world (i.e., the person identiﬁes with both desires, but they

conﬂict because of contingent facts about the world that are attached to these desires)

 

'28 Anton Chekhov, The Cherry Orchard, (Dover Publications Inc. 1991), pp. 19, 32. Lopakhim and
Trophimof are addressing Madame Ranevsky, who is ambivalent about selling her cherry orchard.

'29 Note that I am claiming that resolving paralyzing ambivalence is a necessary condition for metaphysical
autonomy. Resolving paralyzing ambivalence is not a sufﬁcient condition for metaphysical autonomy.
Moreover, the possession of metaphysical autonomy is not a sufficient condition for the possession of
decision making capacity or bioethical autonomy.

64

then the response should be to assess which course of action best coheres with the
person’s other values and desires and invite her to take a leap in that direction. .
Provide Basic Information and Information about Coherence

Left alone, the paralyzed ambivalent person may remain paralyzed and not form a
will. As a result, she will not be metaphysically autonomous. Nor will she have the
ability to make a decision in a medical context, and so she lacks bioethical autonomy as
well. If we want to restore autonomy to this person, then we must help her to make a
decision about which desires she identiﬁes with and which desire she wills be effective in
action at a particular time. ’30 Understanding the cause of her ambivalence may help us
help her to resolve it. Often, it is the case that persons are paralyzingly ambivalent
because of some sort of epistemological deﬁcit. There may be a basic sort of
epistemological deﬁcit where the person lacks knowledge about the consequences of
identifying with one desire rather than another. I discussed this in Chapter Two with the
example of Oscar. 13‘ Oscar is ambivalent about which car to buy. He has a desire to buy
car X. He is drawn towards his desire to buy car X because car X is a good car and
attractive, but he is also drawn away ﬁom his desire to buy car X because he also wants
car Y—also a good car and attractive—and cannot have both. He experiences paralyzing
ambivalence and does not form a will about which desire he wants to move him to action.

But suppose that Oscar knew that car Y had an exhaust problem that was going to be

 

'30 Resolving paralyzing ambivalence is a necessary condition for restoring autonomy, but it is not
sufficient, for the person could form a will/stand behind a desire that she wants to move her to action, but
then act on a different one. In such a case the person would not be autonomous with respect to her action,
but her lack of autonomy would be for a new reason (not because she is paralyzingly ambivalent, but
because she is weak of will).

'3 ' One might want to consider Hamlet another example of a case of paralyzing ambivalence due to a basic
epistemological deﬁcit—he does not know if the commands to kill Claudius are really from his father’s
ghost. There are, however, other interpretations of Hamlet’s case that do not read Hamlet as being

65

costly. Then, Oscar would not be ambivalent. So, in these cases of basic epistemological
deﬁcit, we should help the paralyzed ambivalent person resolve her ambivalence by
providing her with as much information as possible about the consequences of each
course of action. This may help her to come to identify with one desire or another.132

It may be that the cause of the paralyzing ambivalence is a more complex
epistemological deﬁcit. That is, the person cannot decide which desire to identify with
because she has not reﬂected enough to obtain knowledge about how well each desire
coheres with her other (2“d order) values, desires, and preferences. Consider a variation
of the example that I posed in Chapter Two about the woman who is ambivalent about
continuing to be involved with a charity for cats while working in New York: A woman
takes a job in New York for a year. During that year she becomes involved in a charity
for cats. She is both drawn towards and away from her desire to continue to work for this
charity. She is drawn towards it because she likes working with this charity, but she is
drawn away from it because she knows that she is going to have to return to LA. after
one year, and she knows that it is going to be painful because she will have grown
attached to this charity. Her ambivalence is paralyzing in that she does not form a will,
she does not take a stand about whether or not she wants her desire to continue to work
for this charity to be effective in action. Imagine that this woman’s ambivalence is due to
an epistemological deﬁcit of the complex sort: she has not reﬂected enough on how well
this desire coheres with her other (2“d order) values, desires, and preferences. In these
cases of complex epistemological deﬁcits, we should help the paralyzed ambivalent

person resolve her ambivalence by pointing out to her how the desire coheres with her

 

ambivalent due to an epistemological deﬁcit. For example, one may read Hamlet as being weak of will in
not killing Claudius.

66

other (2“‘1 order) values, desires, and preferences. For example, we point out to this
woman that her desire to continue her charity work coheres well with most of the other
2“d order desires and values that she has—they are for the most part values of altruism
and self-sacriﬁce.l33 In more technical terms, Alﬁed Mele has argued that we have
proximal desires (e. g. the desire to continue enjoyable work) and complex proximal
desires, which are simply the conjunction of our proximal desires (e. g. the desire to
continue enjoyable work and the desire to avoid pain and the desire to be altruistic). In
attempting to resolve a person’s paralyzing ambivalence by ﬁlling a complex
epistemological deﬁcit, we are in a sense pointing out to the person her complex
proximal desires. '34 Pointing this out may help her come to identify with one desire or the
other. Another way to approach the conversation is to focus not so much on which course
of action can support more of the person’s values, but on which course of action the
person feels will bring out the best in her. Harry Frankfurt has written that “. . .the
decision of what to identify yourself with is not made on the basis of an evaluation of the
various objects, it is not which is more or less valuable or important than the other; the
primary basis for the decision lies in what the person feels he an live most fully as, what
will bring out the best in him, what will enable him to realize most completely the
capacities for an invigorating and robust emotional and intellectual and volitional life.”135

Is this a method of resolution that promotes the agent’s autonomy? The answer

depends, of course, on which account of metaphysical autonomy you adhere to (refer to

 

'32 Or, as I noted in Chapter Two, it may also have the opposite effect and make her even more paralyzed.
'33 It is true that she may react by chucking her altruistic ambitions, but most likely a person will resolve
their paralyzing ambivalence in favor of their coherent set of ambitions. But, it is worth pointing out that
coherence can be achieved in many different directions.

'34 Alﬁ'ed Mele, Autonomous Agents: From Self Control to Autonomy, (Oxford University Press 2001), p.
42.

'35 Harry Frankfurt, “Discussion With Harry Frankfurt,” in Ethical Perspectives Vol. 5 No. l 1998, p. 27.

67

Chapter One for various accounts). Identiﬁcation theorists about autonomy claim that a
person is autonomous with respect to a desire that moves her to act if she identiﬁes with
it, and some have an additional requirement that the process or method by which the
desire was formed must be one that she identiﬁes with. Providing basic information and
information about coherence is an autonomous way of resolving paralyzing ambivalence
so long as the person has (or would have once it was pointed out to her) no resistance to
this method. And, the lack of resistance does not take place under the inﬂuence of factors
that might inhibit self-reﬂection. It is improbable that a person would feel alienated from
the process of being given basic information and information about how their desires
cohere,I36 and so this would be an autonomous way to resolve paralyzing ambivalence.
Coherence theorists about autonomy would likely be quite supportive of this method of
resolution, for it is likely that the person will end up resolving her paralyzing
ambivalence (if she does indeed resolve it) in favor of the desire that best coheres with
her other preferences once they are pointed out to her.

Although providing information is generally a method of resolution that promotes
the agent’s autonomy, the issue is more complicated. As I mentioned, information may
help a person resolve her ambivalence, but it could also make it worse. From this, the
question arises of whether we should selectively give the person only information (all
true of course) that is likely to resolve her paralyzing ambivalence and hence restore her

autonomy. And similarly, should we give the person information (all true of course) that

 

'36 Although, as Tom Tomlinson has pointed out, this depends a lot on how this is done—I can imagine it
being done in a very presumptuous way that suggests someone else knows better than I do what I should
want, which I would very much resent.

68

is likely to make her ambivalence worse and more excruciating so that she is motivated to
resolve it and hence have her autonomy restored?

Imagine a person who is paralyzingly ambivalent about whether she should put
her money into high risks stocks or into a fund for her children’s college. She is drawn
towards her desire to put the money in stocks because she would like to be able to
achieve a certain sort of lifestyle for her and her children. But she is also drawn towards
her desire to put the money in the college fund, for obvious reasons. Should we try to
give her information that we think will make her ambivalence even more excruciating so
that she is motivated to identify with and will one or the other course of action and regain
her autonomy in this situation? I would argue that we should not. It is a mistake to think
that making someone even more excruciatingly paralyzed is a way to get them to resolve
their ambivalence. Imagine that we insist on meeting this woman for coffee every day
and saying “Martha, you’ve got to decide—your children are getting older and they’re
not collecting any money for college nor are they enjoying their meager lifestyle right
now.” It may work, but for many it may cause them to curl up in the fetal position.
There are better ways to help persons overcome their paralyzing ambivalence than by
making it even more acute. What about giving Martha information (all true of course)
that we think is likely to resolve her paralyzing ambivalence? I would argue that this is
ethically permissible and perhaps even advisable. There is certainly a thought that
autonomous decisions are informed ones, but no one ever has all of the information when
they make a decision. If having all information were a requirement of autonomy then
none of us would ever be autonomous in decisions we make (not to mention that we

would be very boring people, spending all of our time collecting information). So, it is

69

permissible to provide the paralyzingly ambivalent person with some and not all
information. There is, however, a difference between selectively withholding
information because it would cause the person to resolve her ambivalence in a direction
that you don’t want her to (e. g. a broker does not tell Martha how risky the stocks are)
and selectively withholding information because you do not think that it would do
anything to resolve her ambivalence or may make it worse. The latter is compatible with
promoting her autonomy, the former is not. The former is called manipulation. Now, in
selectively providing information that one thinks will help resolve the person’s
paralyzing ambivalence, we should consider what information would help anyone
resolve their ambivalence in the situation (e.g. in Martha’s situation, information about
the likelihood of an increase in happiness if the investment allows for a very materialistic
lifestyle, would be helpful for most anyone making that decision). We should also
consider what information would help this particular person resolve their ambivalence in
the situation (e. g. we may know that Martha’s children are not the best students and that
it is unlikely that they will get scholarships to college) and inform them of that.
Assess Coherence and Invite a Leap

It may be the case that a person has all of the information available and has
considered how the desire that they are ambivalent about coheres with her other (2“d
order) desires, values, and preferences, and is still paralyzingly ambivalent. Often
paralyzing ambivalence is the result of a conﬂict between deﬁning elements of a person’s
nature; such as in Maria Lugones’s example of a woman who identiﬁes as a Latina and a

137

Lesbian. In cases like these, the ambivalence is not due to any epistemological deﬁcit.

As Frankfurt says, “Since [this type of] ambivalence is not a cognitive deﬁciency, it

70

cannot be overcome merely by acquiring additional information.”138 To illustrate
Frankﬁrrt’s point, consider the example of Agamemnon. Agamemnon does not suffer
from any basic epistemological deﬁcit. He knows the consequences of saving his cavalry
(his daughter will be killed), and he knows the consequences of saving his daughter (his
cavalry will perish). Nor does he suffer from any complex epistemological deﬁcit. Let
us say that he has reﬂected on how his desire to save his cavalry coheres with his other
(2“d order) desires and values, and how his desire to save his daughter coheres with his
other (2"d order) desires and values. Both of them cohere equally well with his other (2“(1
order) desires and values. Agamemnon identiﬁes equally with his desire to save his
daughter and with is desire to save his cavalry, it is just that because of contingent facts
about the world, those two desires conﬂict, causing him to be ambivalent about which
desire he wants to be effective in action. Agamemnon’s paralysis could have been
resolved by more information or reﬂection.

And this brings us to a second method for resolving a person’s paralyzing
ambivalence. We should look at the courses of action that the person is torn between,
judge which one we think best coheres with her other preferences and values, and invite
her to take a leap in that direction. Return to the example of Martha; the woman who is
paralyzingly ambivalent about whether she should put her money into high risks stocks or
into a fund for her children’s college. We have tried to help Martha resolve her
paralyzing ambivalence by giving her the information about the riskiness of the stocks,
about her children’s likelihood of getting scholarships, and about how happiness

correlates with material wealth; but Martha is still paralyzingly ambivalent about what to

 

'37 Cheshire Calhoun, “Standing For Something,” in Journal of Philosophy Vol. 92 No. 5 1995 p. 239.

71

will. We should assess which course of action best coheres with Martha’s other values
and desires and invite her to take a leap in that direction. After talking with Martha we
discover that she really values her children’s well-being, so we drive her to the bank and
invite her to put the money into a trust fund for her children’s college.

Before addressing whether or not this is a method of resolution that promotes the
agent’s autonomy, let me brieﬂy note the connections between this method and
existential philosophy. Existentialist philosophers have long emphasized the moment in
decision in which thought and reason can take a person no further. In “Existentialism Is
A Humanism,” J can-Paul Sartre recalls the story of a pupil of his who had to choose
between going to England to join the Free French Forces or staying near his mother and
helping her to live. The young man was torn, but no matter where he looked (externally
towards religious or moral doctrines or internally towards feeling and reason) he did not

ﬁnd the answer. Sartre describes: '39

At the same time, he was hesitating between two kinds of morality; on the one side the
morality of sympathy, of personal devotion and, on the other side, a morality of wider scope
but of more debatable validity. He had to choose between those two. What could help him to
choose? Could the Christian doctrine? No. Christian doctrine says: Act with charity, love your
neighbour, deny yourself for others, choose the way which is hardest, and so forth. But which
is the harder road? To whom does one owe the more brotherly love, the patriot or the mother?
Which is the more useful aim, the general one of ﬁghting in and for the whole community, or
the precise aim of helping one particular person to live? Who can give an answer to that a
prion? No one. Nor is it given in any ethical scripture. The Kantian ethic says, Never regard
another as a means, but always as an end. Very well; if I remain with my mother, I shall be
regarding her as the end and not as a means: but by the same token I am in danger of treating
as means those who are ﬁghting on my behalf; and the converse is also true, that if I go to the
aid of the combatants I shall be treating them as the end at the risk of treating my mother as a
means.

 

'38 Harry Frankfurt, “The F aintest Passion,” in Necessity, Volition, and Love (Cambridge University Press
1999), p. 100. Inside brackets are mine, not Frankfurt’s.

'39 Jean-Paul Sartre, “Existentialism is a Humanism,” in Existentialism ﬁom Dostoevsky to Sartre edited by
Walter Kaufmann (World Publishing Company 1956), p. 296.

72

Consider a humorous example from Benjamin Kunkel’s novel Indecision where

paralyzing ambivalence is resolved by leaping in direction or the other: '40

“I experienced a flashback to a childhood Thanksgiving. Probably dad did too.
I’d loved cranberry sauce, the savory stufﬁng, the turkey itself with such
equality of love that after a gabbled grace I’d been unable to begin eating, and
the more ludicrous the spell of indecision became, the harder to break. I’d
been salivating and paralyzed in front of my plate, plunged in what later came
to be known as the Zone, until ﬁnally dad raised his fork at me saying “Eat!
Eat! Dammit, eat!” So I’d shut my eyes, loaded my fork with mystery, and
raised it toward the cave of my mouth. The tart surprise of the cranberries I
could remember still.”

Is this invited leap an autonomous way to resolve paralyzing ambivalence? I
argue that it is. It is not as if the person is unreﬂectively leaping in one direction or the
other. The person has reﬂected about which desire she wants to move her to action but
has reached a point where reﬂection will take her no ﬁirther. In determining whether or
not it is an autonomous way to resolve paralyzing ambivalence, identiﬁcation theorists
about autonomy would be concerned about whether she identiﬁes with (e. g. is satisﬁed
with, treats as reason-giving, etc.) the desire that she ends up acting on. And, if the
person is paralyzingly ambivalent between two courses of action that she identiﬁes with
but which conﬂict because of contingent facts about the world, then either way she will
be autonomous with respect to the desire that she acts on because either way she will be
acting on a desire that she identiﬁes with.141 Coherence theorists about autonomy would
likely respond that whether or not it is an autonomous resolution depends on whether the

desire that the person ends up acting on is one that coheres with her other values and

 

14° Benjamin Kunkel, Indecision (Random House Trade Paperbacks 2005), p. 83.

m It is true that she is likely to experience residual ambivalence with respect to her decision, and residual
ambivalence (although compatible with metaphysical autonomy, as I argued in Chapter Three), may pose a
threat to metaphysical autonomy (as I argue in Chapter Five).

73

preferences. And, since we are inviting the agent to take a leap in the direction that we
judge best coheres with her other preferences and values, it likely would.

I can anticipate four major objections to the claim that an invited leap is an
autonomous resolution of paralyzing ambivalence. One is that if a person is paralyzingly
ambivalent between two courses of action that she repudiates then there is no way for an
autonomous resolution of paralyzing ambivalent, for either way that she leaps she will be
acting on a desire that she does not identify with. In reply, a person rarely (if ever) is
choosing between courses of action that she categorically repudiates. The person is
choosing between courses of action that she independently identiﬁes with, but the
satisfaction of one course of action negatively affects the satisfaction of another.

The paralyzingly ambivalent person identiﬁes with each of the competing desires.
Suppose that she leaps towards one of them. I have claimed that (on identiﬁcation
accounts of autonomy) the person will be autonomous because any course of action that
she leaps towards is one that she identiﬁes with. But, someone might object, it seems
that that person’s overall metaphysical autonomy is somehow reduced because acting on
one desire that she identiﬁes with prevents her from acting on another that she identiﬁes
with. For example, in Maria Lugones’s example, acting on her desire to live her life as a
Latina prevents her from living her life openly as a Lesbian. It is true that in this case the
agent cannot effectuate the whole of her will, but the world presents us with this all of the
time. To say that a person is not autonomous at a particular time if she cannot effectuate

the whole of her will would result in no one (or very few of us) being autonomous.142

 

142 A similar objection can be found in an example that Cheshire Calhoun gives of a man who is
ambivalent between going to therapy to treat his homosexuality and going to gay bars. Do we really want
to say that he is autonomous if he leaps towards or identiﬁes with his desire to treat his homosexuality?
This objection is essentially the objection of adaptive preferences. John Christman’s requirement that in

74

A third objection that someone might have to the claim that an invited leap is an
autonomous resolution of paralyzing ambivalence is that a person who makes the leap too
early is not autonomous. Consider a person who is indecisive about what desire she wants
to be effective in action (she is paralyzingly ambivalent). Instead of waiting a bit to see if
more information comes in, or if feelings change that could help her come to decide
which desire she wants to be effective in action, she leaps towards one course of action
almost immediately. The objection is that someone with very little or no reﬂection on
their desire, decision, or action cannot be autonomous with respect to it. Richard Double
discusses this objection and counters it by arguing, “. . .there are autonomy exemplars that
count against requiring reﬂectiveness. Rugged individualists may be preoccupied with
evaluating their ﬁrst order desires, but they need not. The man-of-action, so the
paradigm goes, shoots ﬁrst and asks questions later. The true free spirit may not ask
questions at all. And what about the millions of persons to whom reﬂecting on their
lower-order psychological states is not only an infrequent occurrence, but is anathema to
their individual management styles? For many persons, life is to be lived, not worried
over.””3 Double proposes that as long as unreﬂectiveness is part of this person’s
“individual management style” (how she believes she should go about making choices)
then she is autonomous with respect to her desire.144 I think that Double’s proposal is

persuasive.

 

order for a person to be autonomous with respect to a desire she must identify with the process by which
the desire was formed addresses this criticism.

For Calhoun’s example, see Cheshire Calhoun, “Standing For Something,” p. 243.

”3 Richard Double, “Two Types of Autonomy Accounts,” Canadian Journal of Philosophy 22 (1992), p.
73.

'4’ Double, “Two Types of Autonomy Accounts,” p. 73.

75

It might be objected that the notion of a management style does not ﬁt well with
our common experience, as it is not as if we sit down and think about the matter and then
declare a management style. Although it is true that we do not decide on a management
style in this deliberate way, I think that we do have latent styles, or ways that we think we
should go about making choices. What is key, I think, is that if a person’s management
style were pointed out to her, she would endorse it or defend it. Another concern about
the notion of management style is that someone could adopt a management style of, say,
ﬂipping a coin whenever there is a decision to be made. The character Dwight in

Benjamin Kunkel’s novel Indecision does exactly this:145

“. . .people were always calling me and asking me to do things, and since only
pretty rarely was I really sure I wanted to, my system was to ﬂip a coin. ...l was
proud of the system. Statistically fair, it also kept my whole easy nature from
forcing me to do everyone’s bidding; it ensured a certain scarcity of
Dwightness on the market; it contributed the prestige of the inscrutable to my
otherwise transparent persona; and above all it allowed me to ﬁnd out in my
own good time whether I would actually have liked to do the thing in
question.”

Although unusual, as long as Dwight would endorse or defend this method, then it is an
autonomous way to resolve paralyzing ambivalence.146
The fourth and ﬁnal objection that someone might have to the claim that an
invited leap is an autonomous resolution of paralyzing ambivalence is that an leap of the
will is not possible. While it may be possible to leap towards doing one thing or the

other, it is not possible to leap towards willing one thing or the other; and so, the leap will

not resolve paralyzing ambivalence because ambivalence is essentially a paralysis of the

 

”5 Kunkel, Indecision, p. 19.
”6 I discuss “bad faith” later in this chapter, but here I will note that the coin ﬂipping need not be an
example of bad faith. Flipping a coin to make decisions is bad faith if the person thinks that in ﬂipping a

76

will. To quote Ludwig Wittgenstein, “Willing too is merely an experience, one would
like to say. . .It comes when it comes, and I cannot bring it about.”147 Frankﬁrrt has also

argued that one cannot have whatever will she wants to have. He argues: '48

“We do not control, by our voluntary command, the spirits within our own
vasty deeps. We cannot have, simply for the asking, whatever will we want.
We are not ﬁctitious characters, who have sovereign authors; nor are we gods
who can be authors of more than ﬁction. Therefore, we cannot be authors of
ourselves. Reducing our own volitional indeterminacy, and becoming truly
wholehearted, is not a matter of telling stories about our lives. Nor, unless we
wish to be as foolish as Owen Glendower, can we propose to shape our wills
by stipulating peremptorily at some moment that now we are no longer divided
but have become solidly resolute. We can be only what nature and life make
us, and that is not so readily up to us

Frankfurt has also claimed that ambivalence cannot be overcome voluntarily: “A
person cannot make himself volitionally determinate and thereby create a truth where
there was none before, merely by an “act of will.” In other words, he cannot make
himself wholehearted just by a psychic movement that is fully under his immediate
voluntary control.”149 In fact, Frankfurt seems almost to anticipate my leap proposal and
reply to it, arguing, “To be sure, a person may attempt to resolve his ambivalence by
deciding to adhere unequivocally to one of his alternatives rather than to the other; and he
may believe that in thus making up his mind he has eliminated the division in his will and
become wholehearted. Whether such changes have actually occurred, however, is
another matter. When the chips are down he may discover that he is not, after all,
decisively moved by the preference or motive he supposed he had adopted.”150 If he

discovers that in fact he did not resolve his ambivalence, Frankfurt explains, “No doubt

 

coin he is escaping choosing. If the person recognizes that ﬂipping a coin is making a choice; if he projects
himself as a coin-ﬂipper, as someone who gambles with decisions/life, then it is not a case of bad faith.

”7 Ludwig Wittgenstein, Philosophical Investigations, paragraph 611.

”8 Frankfurt, “The Faintest Passion,” p. 101.

"9 Frankfurt, “The F aintest Passion,” p. 100.

'50 Frankfurt, “The F aintest Passion,” p. 101.

77

he made us his mind, but doing that manifestly failed to shape his will. . . .since no
effective volitional commitment was actually accomplished, insisting that a decision was
made would require conceding that the “decision” was no more than a merely verbal or
intellectual event.”15 I
I have two responses to Frankfurt’s argument. First, while I recognize that we
cannot directly shape our will, we can indirectly shape it. Second, action may shape the
will, and so leaping into action may be equivalent to leaping into will. Alﬁed Mele has
argued in his book, Autonomous Agents: From Self-Control to Autonomy, that we need
not wait for a desire that we do not endorse to grow weaker in order to exercise self-

‘52 We can exercise self-control indirectly by employing certain

control with respect to it.
techniques such as self-command. Mele presents an example of a man (Ian) who desires
to paint the shed and wants that desire to be effective in action, but ﬁnds himself
overpowered by a desire to watch TV. While he cannot directly defeat his desire to
watch TV, he does so indirectly by uttering a self command (“Ian, get up and paint the
shed you lazy sot”). Mele says, “Ian may be in the habit of obeying his self-

commands. . .and his uttering the command consequently may tap an additional source of
motivation.”153 I think that we can transfer Mele’s points about the extent of our control
over our desires during weakness of will to the extent of our control over our will during
paralyzing ambivalence. So, we need not wait to identify with one desire more than the

other in order to resolve paralyzing ambivalence. We can shape our will indirectly by

uttering a self-command such as (in the case of the example of the man that I will

 

'5' Harry Frankfurt, “Rationality and the Unthinkable,” in The Importance of What We Care About
(Cambridge University Press, 1988), p. 181.

’52 Mele, Autonomous Agents, p. 43.

'53 Mele, Autonomous Agents, p. 45.

78

introduce in a moment who is ambivalent about which woman to commit to), “Ian,
commit to Angela, you two-timing sot.” In this way, Ian can indirectly leap into will and
resolve his paralyzing ambivalence, for the self-command may provide additional
motivation that causes him to identify with one desire over the other, where before he
was dead-locked between the two desires. Frankfurt himself would agree with my
proposal for indirect control of the will, I think, for he says, “On the other hand, it is
surely open to someone for whom an action is unthinkable to try by other means, less
direct than the exercise of willpower alone, to alter his own will in such a way that the
action becomes thinkable for him. The fact that a person cannot bring himself to perform
an action does not entail that he cannot bring himself to act with the intention of changing
that fact.”154

My second point is that action may shape the will, and so leaping into action may
be equivalent to leaping into will. This line of argument can be nicely summarized by
Sartre’s comments in “Existentialism Is A Humanism:” “Moreover, as Gide has very well
said, a sentiment which is play-acting and one which is vital are two things that are
hardly distinguishable one ﬁom another. To decide that I love my mother by staying
beside her, and to play a comedy the upshot of which is that I do so — these are nearly the
same thing. In other words, feeling is formed by the deeds that one does. . .”'55 So, in
throwing myself towards acting in accordance with one or another of the desires that I
identify with, I may come to will that desire; I may come to resolve my paralyzing

ambivalence. This line of argument is not particular to existentialists. It has been widely

 

'54 Frankfurt, “Rationality and the Unthinkable,” p. 137.
'55 Sartre, “Existentialism is a Humanism,” p. 297.

79

recognized. CS. Lewis, for example, makes a similar argument when he suggests that if
one behaves as if they believe in God, then they will come to believe in God.156
Change The World

A third method of resolving paralyzing ambivalence is to change the situation
such that the person can attain both courses of action (if they are both desired), or avoid
both courses of action (if they are both dreaded). In Agamemnon’s case, it would be to
change the situation so that he does not have to choose between his daughter and his
cavalry; in Lugones’s case, it would be to change the social world so that identifying as a
Latina and identifying as a Lesbian are compatible.

This solution might work; in some cases of ambivalence the person could ﬁnd a
way to compromise so that she has some of both. For example, a graduate student is
ambivalent about whether to focus on applied psychology or academic psychology. She
could easily change the world so that she could have both by enrolling in, or transferring
to, a graduate school that emphasizes both. But the severity of ambivalence found in
paralyzing ambivalence is usually the result of having to decide between making a full
commitment to one thing or another; a compromise is not obvious or available. For
example, Agamemnon is paralyzingly ambivalent about whether to sacriﬁce his daughter
or his army. He has to fully commit to one or the other; a compromise is not obvious or
available. Consider another example: a man is paralyzingly ambivalent about whether to
remain with woman A or to go be with woman B. He has to fully commit to one or the
other; a compromise is not obvious or available. Finding a way to have both desires met
in these two cases is much more difficult. The person must be very creative and

imaginative in order to change the circumstances so that she can have both of her desires

 

'56 CS. Lewis, Mere Christianity, (Harper Collins Publishers 2001), p. 187.

80

met.157 If the person is able to do this, then this is indeed an autonomous way to resolve
her ambivalence, for she is ﬁnding a way to act on [both] desires that she identiﬁes with
(that she feels satisﬁed with, views as reason-giving, etc.).

Resolve For The Person:

Now let me address a fourth possible way to resolve a person’s paralyzing
ambivalence: we could resolve her paralyzing ambivalence for her. Meaning, we could
force her to take on one course of action or the other, or we could choose a course of
action for her without her knowledge. For example, in the case of the woman who is
ambivalent about continuing her charity work, we take over the situation and force her to
keep working (even if she is protesting or upset). Or, we could tell the charity that she
quits without her knowledge.

There are a couple of problems with this method of resolution. First of all, it may
not work to resolve her ambivalence. As I discussed in the last section, it is not
necessarily the case that action will shape the will and so the person may still remain
paralyzingly ambivalent, especially if her action is literally forced or if she does not even
know that the decision has been made for her. Second, it may not work to restore her
autonomy. When she becomes aware that we made the decision for her, she may actually
in psychological revolt will that the other conﬂicting desire be the one that is effective in
action—but then it may be too late for her to take that course of action and so she will
ﬁnd herself stuck acting on a desire that she does not will and hence she will not be

acting autonomously.

 

’57 For example, perhaps the ambivalent lover can move to a country where polygamy is accepted and
legal. Perhaps Agamemnon could have convinced Artemis (the goddess who is conﬂicted the wrath on his
army) to cease.

81

The third problem with this method of resolution is that it fails to distinguish
between autonomy as an ideal and autonomy as a right. As I argued in Chapter One, one
of the purposes of metaphysical autonomy theories is to develop an ideal of the person as
someone who acts on desires that she identiﬁes with, or that cohere with her other values,
and avoids acting on ones that she repudiates; whereas bioethical autonomy focuses on a
person’s right to make her own decisions. It might be true that forcing the woman to
keep working at her charity, or telling the charity that she quits without her knowledge
moves her closer to that ideal metaphysically autonomous person. But the reason that
this seems ethically egregious to us is because it interferes with her right to make her own
decisions.

Not Choose
“Not to resolve is to resolve; and many times it breeds as many necessities, and engageth

as far in some other sort, as to resolve.” — Francis Bacon

A ﬁfth possible response to paralyzing ambivalence is to just remain paralyzingly
ambivalent. In fact, we may think that there are sometimes good reasons to do so.
Imagine a case of a parent who has twins. Both twins need a heart transplant in the next
ﬁve hours, or they will die. A heart comes in after four hours and forty ﬁve minutes, and
the children’s parent has to decide whether he wants his desire for Twin 1 to get the heart
or for Twin 2 to get the heart to be effective in action. He is paralyzingly ambivalent
because he identiﬁes with both desires, but they conﬂict because of contingent facts
about the world. Imagine that he chooses to remain ambivalent—to not choose which

desire he wants to be effective in action.

82

Next, imagine a man who is in love with two women at once. Both women are
pressuring this man to take the relationship to the next level and to move in together. He
wants to commit himself to one person, to change his ways so to speak; but he is
paralyzingly ambivalent about whether he wants his desire to commit to Woman 1 to be
effective in action, or if he wants his desire to commit to Woman 2 to be effective in
action. Resolving his ambivalence (leaping towards commitment to Woman 1), has the
cost of losing Woman 2 and perhaps then being plagued by residual ambivalence, and so
he chooses to remain ambivalent—to not choose which desire he wants to be effective in
action. '58

We can see where the parent and the romancer are coming from, but are they
responding autonomously to their paralyzing ambivalence? Is remaining paralyzingly
ambivalent an autonomous response? I think that a distinction will help at this point; a
distinction between choosing to not choose and choosing to remain paralyzingly
ambivalent. I think that those who say that there may be reasons to remain paralyzingly
ambivalent really mean that there may be reasons for not choosing. A person can never
autonomously choose to remain paralyzingly conﬂicted, at least on attitudinal
identiﬁcation theories of autonomy (such as F rankfurt’s), for it is impossible to be
satisﬁed with or relaxed with paralyzing conﬂiction. But, can a person autonomously
choose not to choose? I think that most accounts of autonomy would allow for this.
Identiﬁcation theorists would allow for it so long as the person feels satisﬁed with her

decision to not choose, or views her not choosing to be in accord with a non-instrumental

 

'58 One might think that instead of leaping towards commitment to one woman or the other, he could (and
probably will) simply abandon or downgrade his desire for commitment (thinking commitment is not so
great after all). I would argue that this is not possible. The reason that he is so paralyzed and strongly
conﬂicted is because he knows that he must commit to one or the other, or he will lose both.

83

higher order self-governing policy that she has and is satisﬁed with, etc. Coherence
theorists would allow for it so long as her decision not to choose coheres with other
endorsed preferences that she has.

Although a person can autonomously choose not to choose, there is a difference
between someone who views their not choosing as a choice itself (as taking a position on
the issue) and someone who views their not choosing as an escape. The latter is a case of
what Sartre would call “bad faith.” In Sartre’s “Existentialism Is A Humanism” he says:
“. . .what is not possible is not to choose. I can always choose, but I must know that if I do
not choose, that is still a choice.” Sartre argued that the self is made up of two parts: the
things that are givens about us or that others ascribe to us (e. g. our hair color, weight,
height, job etc.), and the attitude that we take towards these things (e. g. not caring that
others ﬁnd me ugly, or resolving to change my job). Bad faith takes two forms: either
denying that I do have the ﬁ'eedom to choose how I react to others’ perceptions of me or
to facts about my situation, or denying the facts about my situation and thinking that I can
do anything just by wishing it 30.159 To think that I have the option of not choosing is
bad faith. Return to the example of the parent who is ambivalent about which of his
children should get the heart. If he views his refusal to choose as an escape then he is
doing so in bad faith. On the other hand, if he views his refusal to choose as itself a
choice, if he views it as him fashioning himself as a father who does not choose between
his children, realizing that he is constructing himself this way in front of others and that
he is then ready to accept or deal with whatever moral judgments they put on him; then

he is doing so in good faith.

 

84

I would argue, however, that good faith (choosing, or realizing that in “not
choosing” I am fashioning myself a certain way) is more important in some
circumstances than others. In Chapter Two I discussed how some of our choices are ones
that we consider to be more important than others, and how some are ones that we
consider to be identity-conferring. Being in bad faith about something that I consider to
be important and identity conferring (e. g. my religious beliefs) is worse for me than being
in bad faith about something that I consider to be neither important nor identity
conferring (e. g. which socks I buy).

Commit Suicide

A sixth possible response to paralyzing ambivalence is to commit suicide. '60 As
Lynn McF all notes, it has been argued that suicide may be chosen rationally and
deliberately as an act of rejecting both alternatives in a world where they are the only

1 Agamemnon could have killed himself as an act of

choices and both are intolerable.l6
rejecting the sacriﬁce of his daughter and the sacriﬁce of his cavalry. Is this an
autonomous way to resolve paralyzing ambivalence? Committing suicide is akin to
choosing not to choose, which I have argued, can be an autonomy promoting response,
and can also be done in good faith so long as he admits that he has indeed made a

choice—a choice to deﬁne himself a certain way or to make a certain statement (e.g. this

is what a responsible person does in this situation).

 

'59 See “Existentialism Is A Humanism,” p. 305. Also see J.P. Sartre, Anti-Semite and Jew (New York:
Schocken books, 1948).

'60 A seventh possible way that a person could respond to her paralyzing ambivalence is to alternate
between one course of action and the other (e. g. half the day go to the beach, the other half go to the park).
In most cases of paralyzing ambivalence (for example, Agamemnon), this is not feasible, and so I will not
discuss it as a method of resolution.

'6' Lynn McFall, “Integrity,” Ethics Vol. 98 No. 1 (1987), p. 10.

85

First Order Desires Take Over

Another possible way for paralyzing ambivalence to be resolved is if one of the
person’s ﬁrst order desires simply takes over.162 For example, a woman has a desire to
return to school after being a stay at home mom for a number of years. But she also has a
desire to continue to stay at home because she believes that it is best for her children.
When she reﬂects on her desires, she is uncertain about which one she wants to be
effective in action; she is conﬂicted and torn; she is paralyzingly ambivalent. In the
midst of her paralyzing ambivalence she suddenly becomes overcome with fear and self-
doubt about returning to school. Her fear and self-doubt take over and she decides to just
continue to stay at home with her children.

It is true that in the end her paralyzing ambivalence (which was blocking her from
being metaphysically autonomous) is resolved, but it is not resolved in an autonomous
way. She does not (in any way) identify with her fear and self-doubt, nor does it cohere
with her other endorsed desires and values. As Gerald Dworkin has argued, we regard
fear as cancelling liberty even though we do not view something such as obligation as

cancelling liberty. '63

The reason for this is that often persons endorse acting for reasons
of obligation (they have a second order desire to act in ways that fulﬁll obligations),
whereas persons normally do not endorse acting for reasons of fear. For example, if this
woman decided to continue to stay home with her children because she feels that it is her
duty and she generally endorses acting for reasons of duty and obligation, then this is not

a case of ﬁrst order desires just taking over. It is a case of someone acting in accordance

with her second order desires, and hence she is not metaphysically autonomous with

 

'62 Note the passivity of the agent (a way for ambivalence to be resolved) as opposed to the activity of the
agent in the other responses (a way for the agent to respond or resolve).

86

respect to action is autonomous. If, however, this woman decides to continue to stay
home with her children because she is swept away by fear and self doubt and she
generally does not endorse acting for such reasons, then it is a case of ﬁrst order desires
just taking over. It is a case of someone not acting in accordance with her second order
desires, and hence she is not metaphysically autonomous with respect to her action.

In closing Chapter Four, I will refer you to some comforting advice. Although I
have suggested ways to resolve paralyzing ambivalence, it is possible that in some cases
none of the suggestions work, and a person remains paralyzingly ambivalent. When
Frankfurt was pushed on what he had to say about cases of unresolved paralyzing
ambivalence, he had the following advice: “Sometimes a person is so ambivalent, or
vacillates so ﬂuidly, that there is no stable fact concerning what he thinks or
feels. . .suppose you are simply unable to make up your mind. No matter how you twist
or turn, you cannot ﬁnd a way of being satisﬁed with yourself. My advice is that, if your
will is utterly divided, and volitional unity is really out of the question, be sure at least to

hang on to your sense of humor?”64

 

'63 Gerald Dworkin, “Acting Freely,” in Nous Vol. 4 No. 4 1970, p. 376.
'6’ Frankfurt, “The Faintest Passion,” p. 107.

87

Chapter Five: Resolving Residual Ambivalence

“If a man will begin in certainties he shall end in doubts; but if he will be content to
begin in doubts, he shall end in certainties. ” — Francis Bacon

In Chapter Three I argued that residual ambivalence (the phenomenon in which a
person is both drawn towards and away from a particular ﬁrst order desire that she has,
but has taken a stand about whether or not she wants it to move her to action) does not
make a person non-autonomous; the person does have a will, and so as long as her action
is compatible with her will, then she is autonomous. Even though I have argued that
residual ambivalence does not preclude a person from being metaphysically and
bioethically autonomous, it may have other drawbacks.I65
Drawbacks of Residual Ambivalence

Here are several possible drawbacks of residual ambivalence: it can harm one’s
self, harm others, and put a person dangerously close to paralyzing ambivalence which
would result in a loss of autonomy. Let me begin with how residual ambivalence can
harm one’s self. There are two senses of “harm” that one could have in mind:
experiential and non-experiential. Residual ambivalence could cause experiential harm

to a person by creating psychological stress. Frankfurt argues that persons have a

 

"’5 It may help to recall my examples of residual ambivalence from Chapter Two: Paul Pennyfeather in
Evelyn Waugh’s Decline and Fall: even though he has taken the stand that he does not want the desire to
take the money to move him to action and remains determined, he is still strongly inﬂuenced by the desire
for the money, and this leaves him uneasy (residually ambivalent) with his choice; Winston in George
Orwell’s I 984: he forms a will when he decides that he wants his desire for love to be effective in action.
But of course he is still strongly inﬂuenced by the desire for safety, and this leaves him uneasy (residually
ambivalent) with his choice to meet Julia; Parent of the Drug Addicted Son: he commits to and stands
behind his decision to not let his desire for a relationship with his son to be effective in action. Of course
the parent is still strongly inﬂuenced by this desire for a relationship; be even insists that this decision does
not mean that he does not love his son, and so he is uneasy (residually ambivalent) with his choice.

88

primitive need for self-unity, and that “Any threat to this unity—that is, any threat to the
cohesion of the self—tends to alarm a person and to mobilize him for an attempt at ‘self—
preservation?”66 When presented with choices, a person who is residually ambivalent
may endure more anguish. Frankurt argues: “For someone who is unlikely to have any
stable preferences or goals, the beneﬁts of freedom are, at the very least, severely
diminished. ...The fact that he is ﬂee to choose between them is likely only to make his
anguish more poignant and more intense.”167 Making a similar point about residual
ambivalence and anguish is Steve Harrist, who through empirical studies found that an
overwhelming degree of ambivalence leaves one feeling obsessive, unable to decide, and
may result in identity diffusion.168

Residual ambivalence may also cause what could be called non-experiential harm
to the self. Non-experiential harm is often discussed in the literature about the harms of
death. Some philosophers argue that a person can be harmed non-experientially by their
death, or after their death. Presenting the example of a person who has honible lies
spread about her after her death, they argue that the harm lies in the tarnished reputation
of her character, regardless of her experience of it.169 A similar argument may be made
about the harm that residual ambivalence inﬂicts upon the integrity of a person’s
character. Lynn McFall, Cheshire Calhoun, and Harry Frankfurt all argue that integrity

requires that there be something that you would not do—something that you stand for

 

"’6 Harry Frankfurt, “Autonomy, Necessity, and Love,” p. 139.

I want to clarify the connection between unity and ambivalence. By deﬁnition, ambivalence is always a
kind of disunity. But not all disunity produces ambivalence.

"’7 Harry Frankfurt, “The Faintest Passion,” p. 102.

"’8 Steve Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” Journal of
Phenomenological Psychology Vol. 37 No. l 2006, p. 89.

'69 See John Martin Fisher, The Metaphysics of Death (Stanford University Press 1993).

89

over time and throughout changing circumstances.170 Calhoun writes, “. . .integrity is a
matter of endorsing and, should the occasion arise, standing on some bottom-line
principles. . . 3,171 Calhoun further discusses this picture of integrity in her discussion of
Gabrielle Taylor’s work: “. . .as Taylor also observes, unless the individual regards her
endorsements as prima facie committing her to making the same endorsements on future
occasions, she will be no more than shallowly sincere, wholeheartedly identifying with
one set of desires today and a different set tomorrow.”172
Residual ambivalence may pose a threat to a person’s integrity, because standing
for something over time requires that it is something about which the person is un-
conﬂicted. Imagine for example that I have a strong commitment to the animal rights
movement. Put more technically, I have a desire to commit to the rights of animals, and I
identify with that desire. Now imagine that it is a bit more complicated than this;
imagine that there is some residual ambivalence involved. Although I have taken a stand
that I want my desire to commit to the rights of animals to move me to action, I am both
drawn towards and away from this desire. I am drawn away ﬁ'om it due to my other
conﬂicting desires (that I also identify with). . .a desire to advance medicine (which I
think requires animal experimentation), a desire to keep up my health (which I cannot
seem to do as a vegetarian due to the fact that I dislike vegetables and so only eat pizza

and ice cream), etc. The argument might be made that the presence of these conﬂicting

desires may eventually cause me to cease to stand by my longtime commitment to the

 

”0 See Cheshire Calhoun, “Standing For Something,” The Journal of Philosophy Vol. 42 No 5 (1995), pp.
235-260; Lynn McFall, “Integrity,” Ethics Vol. 98 (1987), pp. 5-20; Harry Frankfurt, “Rationality and the
Unthinkable,” in The Importance of What We Care About (Cambridge University Press 1988), pp. 177-190;
and Harry Frankﬁrrt, “Autonomy, Necessity, and Love,” in Necessity, Volition, and Love (Cambridge
University Press 1994), pp. 129-142.

'71 Cheshire Calhoun, “Standing For Something,” p. 246.

'72 Cheshire Calhoun, “Standing For Something,” p. 237.

90

rights of animals. Residual ambivalence will have resulted in a loss of integrity; and
whether or not I experience any harm ﬁ'om this loss, it is harmful nonetheless because
self-integrity is valuable in and of itself.173

There are a couple of replies at this point. One is that just because I drop my
commitment to animal rights, this does not mean that there is not something to which I
stand for over time. And so, I may still be a person with integrity more generally. But,
suppose that what has happened with my commitment to animal rights has happened to
many other things in my life as well. That is, my residual ambivalence has a tendency to
in many situations cause me to cease standing by things that I previously stood for; one
day I stand for X and as a result of residual ambivalence I cease to stand for X and the
next day stand for Y. I am then, as Calhoun writes, “wholeheartedly identifying with one
set of desires today and a different set tomorrow;” I am lacking integrity.I74 While I see
this worry, it brings me to my second reply, which is that there is nothing intrinsically
wrong with being a person who lacks integrity. The “badness” of lacking integrity as a
result of residual ambivalence, or of residual ambivalence itself, lies either in some
experiential harm to oneself or some experiential harm to others. So, supposing that one
is not experiencing any harmful effects from a lack of integrity; the other possible harm
caused from lacking integrity is harm to others.

Residual ambivalence may cause harm to others. To see this, simply imagine a

person who makes a lifetime commitment to a romantic partner. Although this person

 

'73 Tom Tomlinson raised the interesting question of whether there would be a lack of integrity if I had not
vowed a commitment to animal rights, but only aspired to one (and kept it to myself). If not, the problem
would be with my actions, not with my residual ambivalence. I think that the problem is with my residual
ambivalence; I think that the lack of integrity is present even if I never told anyone else about my
commitment to animal rights.

174 Cheshire Calhoun, “Standing For Something,” p. 237 .

91

has taken a stand that she wants her desire to commit to this person for a lifetime to move
her to action, she experiences some residual ambivalence. The residual ambivalence
presses on her for awhile and eventually she ceases to stand by her lifetime commitment
to her romantic partner. This cessation hurts her partner deeply. Moreover, residual
ambivalence may harm personal relationships themselves. In his paper, “Emotional
Ambivalence,” Philip Koch notes that ambivalence often harms personal relationships.
Koch argues that having a personal relationship with an ambivalent person is difﬁcult
because ambivalence blocks the understanding and unity that we need to construct of the
person that we are trying to have a relationship with.175 Ambivalence makes it difﬁcult
for others to know me and have a relationship with me, and this loss deprives us both.
Similar to the worry about mental anguish, the worry about harm to others and our
relationships with them is a legitimate worry about residual ambivalence.

A third worry about residual ambivalence is that it poses a threat to autonomy.
While I have argued that residual ambivalence does not preclude metaphysical autonomy,
someone might argue that it does threaten it. Residual ambivalence may threaten
metaphysical autonomy in two ways. First, it may put a person dangerously close to
paralyzing ambivalence, which does, so I have argued, preclude metaphysical autonomy.
Second, because residual ambivalence can, as I have articulated, break up the consistency
of what a person stands for over time, it may cause a person to be unsatisﬁed with desires
that have moved her to action in the past. The ﬁrst way that residual ambivalence may
pose a threat to metaphysical autonomy is fairly clear. Suppose that, returning to the last
example, a person stands behind her desire to make a lifetime commitment to a romantic

partner, despite the fact that she is experiencing some residual ambivalence about it.

 

'75 Philip Koch, “Emotional Ambivalence,” pp. 267-268.

92

Now, suppose that over time her residual ambivalence builds and builds until soon she no
longer stands behind her desire to make a lifetime commitment to her romantic partner,
nor does she stand behind her conﬂicting desire to date a variety of other people. She is
paralyzingly ambivalent about which desire she wants to move her to action. And, as a
result, she has lost her metaphysical autonomy in this situation.

The second way that residual ambivalence may pose a threat to metaphysical
autonomy is a bit more complicated. Recall that in order for a person to be
metaphysically autonomous with respect to a particular desire or action, she must identify
with that desire (or, on coherence accounts, it must cohere with her other established
preferences). Because residual ambivalence might break up the consistency of what a
person stands for over time, she may not be satisﬁed with desires that have moved her to
action in the past. Although it may be the case that the person was at the time satisﬁed
with her desire to commit herself to her romantic partner for a lifetime moving her to
action, now she is not. Being a person with residual ambivalence about many of her
decisions and actions may cause the person to generally not be satisﬁed with desires that
moved her to action last week or last month; and if she am the sort of person that is
generally not satisﬁed with the desires that she has acted on, then she is generally not a
metaphysically autonomous person.

Beneﬁts of Residual Ambivalence

Although I have argued that there are drawbacks to residual ambivalence, I will
now argue that there are perhaps even more beneﬁts. Residual ambivalence can help
with personal growth, with social relationships and empathizing with others, and it can

actually (and counter-intuitively) secure metaphysical autonomy.

93

First, residual ambivalence can help with personal growth. Soren Kierkegaard
argued that people become greater by means of distress, anxiety, and paradox.176 He
recalls the story of Abraham who was commanded by God to kill his only son Isaac.
Abraham was severely conﬂicted about this action—believing both that he should obey
God and that killing Isaac would be wrong and hurt Abraham horribly. Because of the
distress, anxiety, and paradox of the conﬂiction, Abraham became greater—he became a
man of faith (which is the highest realm of being according to Kierkegaard). God
intervened and Abraham neither disobeyed God nor hurt Isaac. Also noting the
connection between internal conﬂiction and personal growth is Steve Harrist: “Another
advantage is that it presents the opportunity for growth and change because it involves
coming to terms with the rich complexity of experience that might otherwise be
hidden.”177 And, “. . .more fully experiencing the full spectrum of contradictory feelings
that were initially avoided may open the way to a richer emotional life with a deeper
appreciation for the signiﬁcance and importance of multi-valenced feelings in one’s
life.”178 And, ﬁnally, as most professional philosophers can, I can personally attest to the
connection between internal conﬂiction and personal growth. I remember my ﬁrst year
in college. Most of my values, beliefs, and desires were ones that I (even after reﬂection)
solidly endorsed and had no internal conﬂiction about. And then, I took a philosophy
course. I began to develop conﬂiction about which values, beliefs, and desires I
identiﬁed with. Struggling with this allowed me to grow and forced me to be more

engaged in life. Internal conﬂict is difficult, but as Jack London writes, “It is so much

 

”6 Soren Kierkegaard, Fear and Trembling, (Princeton University Press 1983), p. 65.

177 Steve Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” Journal of
Phenomenological Psychology Vol. 37 No. 1 2006, p. 108.

”8 Steve Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 109.

94

easier to live placidly and complacently. Of course, to live placidly and complacently is
not to live at all.”179

Second, residual ambivalence can help in social relationships and in empathizing
with others. Philip Koch, Patricia Greenspan, and Cheshire Calhoun have all noted this
positive aspect of ambivalence. Koch argues that feeling conﬂicting emotions and points
of View helps us to develop empathy and moral consciousness'so; Greenspan argues that
it allows us to commit to multiple persons’ interests, which causes them to feel as if they
can depend on usm; and Calhoun argues that “When one’s own and others’ judgments
come into serious conﬂict, ambivalence may be a way of acknowledging that
equality?”82

Third, residual ambivalence can actually secure metaphysical autonomy. Imagine
two versions of my earlier example: In Version One, person one has a desire to commit
to his romantic partner for a lifetime and he identiﬁes with that desire. Moreover, he has
ﬁgured out a way to rid himself of any residual ambivalence when it begins to creep up
(imagine that he takes a pill, or employs some psychological technique). He has ﬁgured
out a way to be completely wholehearted and completely un-conﬂicted about his desire to
commit to his partner. In Version Two, person two has a desire to commit to his
romantic partner for a lifetime and he identiﬁes with that desire. Person two, however,
allows his residual ambivalence to play out when it creeps up. He has still taken a stand

that he wants his desire to commit to this person for a lifetime to move him to action, but

he allows himself to feel his conﬂicting desires to date other people. Person two’s

 

”9 Xinli Chen, Being and Authenticity (Rodopi 2003), p. 57.

'80 Philip Koch, “Emotional Ambivalence,” p. 279.

'8' Patricia Greenspan, “A Case of Mixed Feelings: Ambivalence and the Logic of Emotion,” in Explaining
Emotions edited by Amelie Oksenberg Rorty (University of California Press 1980), p. 241.

95

decision has come out of a struggle; he is more an agent of his will than person one; the
nature of person two’s choosing is more autonomous. Moreover, fast forward ﬁve years:
I would argue that person two is more likely to be autonomous with respect to this
situation than person one. Because person two let himself feel and consider his
conﬂicting desires, should he continue to stand by his commitment to his partner, he is
likely more sure of and more satisﬁed with that commitment; he is still autonomous with
respect to it. Because person 1 did not allow himself to feel and consider his conﬂicting
desires, he is quite likely now unhappy about and unsatisﬁed with his situation; he is not

183

autonomous with respect to it. My prediction is supported by empirical evidence. In

Steve Harrist’s empirical study of ambivalence, one participant remarks (of the
ambivalence): “It was a powerful experience. It solidiﬁed a lot of things for me.”184
One ﬁnal and related note on how residual ambivalence can help secure
metaphysical autonomy. In Chapter Two, I discussed the causes of ambivalence, one of
them being that persons have multiple interests and identities that sometimes conﬂict
because of contingent facts about the world; for example, as Maria Lugones points out,
identifying as both a Latina and a Lesbian.185 If a person stiﬂes their internal conﬂict and
residual ambivalence about their identiﬁcations, then they will more than likely end up
being unsatisﬁed with their actions. As Amy Mullin writes, “. . .we are plural selves who

need to encourage a respectful conversation among the parts of the self if we are to avoid

splitting off and denying parts of ourselves associated with the disenfranchised?”86 I

 

‘32 Cheshire Calhoun, “Standing For Something,” The Journal of Philosophy Vol. 42 No 5 (1995), p. 241.
'83 Of course, this is a contingent, empirical claim. He could be just as happy as person two. He could
have kept his autonomy by making himself into a simpleton. I will discuss the simpleton later in this
chapter.

'8’ Steve Harrist, “A Phenomenological Investigation of the Experience of Ambivalence,” p. 107.

'85 Cheshire Calhoun, “Standing For Something,” p. 239.

'86 Amy Mullin, "Selves, Diverse and Divided: Can Feminists Have Diversity Without Multiplicity?" in

96

argue that if we deny parts of ourselves in order to avoid internal conﬂict and be
maximally uniﬁed and wholehearted about our identiﬁcations, we may in the long run
feel alienated from our actions. Cheshire Calhoun suggests that we simultaneously stand
behind our convictions and take seriously others’ doubts about them, and argues that
autonomy may sometimes require resisting the impulse to resolve inconsistencies and
ambivalence.187 Philip Koch has argued that ambivalence may produce expansion and an
outburst of activity.‘88 I propose that a person could use that expansion and creativity to
ﬁnd creative solutions for incorporating the diverse parts of the self, instead of denying
them. If a person does this, it is more likely that she will be satisﬁed with the desires that
are moving her to action; it is more likely that she will be metaphysically autonomous
with respect to them.
Resolving Residual Ambivalence: Types of Wholeheartedness

Residual ambivalence has both beneﬁts and drawbacks. Some persons may
consider the drawbacks to outweigh the beneﬁts, and may seek to rid themselves of their
residual ambivalence; to resolve it. The opposite of ambivalence is wholeheartedness.
But, there are different types of wholeheartedness. One kind of wholeheartedness is what
I will call “synchronic wholeheartedness.” Recall my diagram (ﬁom Chapter Two)

explaining where ambivalence commonly occurs (in deciding on the ordering of desires):

 

Hypatia Vol. 10 No. 4 1995, p. 13.

'87 Cheshire Calhoun, “Standing For Something,” p. 238.

188 Paul Koch, "Emotional Ambivalence" inPhilosophy and Phenomenological Research Vol. 48 No. 2
1987, pp. 257-279.

97

l. Reﬂect on ﬁrst order desires and decide whether they are ones with which I:

/ \

Identify (assign some position) Reject/Outlaw (assign no position)
2. Decide Oil the order

1&— The one that I want to move me to act at t1 (the one that I will)

i:

4.
Recall that the residually ambivalent person has decided which of her desires she wants
to move her to action at a particular time (she has formed a will), but she is still strongly
inﬂuenced by the other, competing desires that she identiﬁes with. So, one type of
wholeheartedness (synchronic wholeheartedness) would be where the other desires that
the person identiﬁes with are not competitors. There is not competition and conﬂict
within the ordered set of desires. Satisfying desire one does not make it harder to satisfy
desire two, or three, etc. A person who is synchronically wholehearted may decide that
she wants her desire to join the Peace Corps to be effective in action right now, but this is
perfectly compatible with the satisfaction of the other desires that she identiﬁes with. It
is true that satisfying her desire to join the Peace Corps right now may postpone the
satisfaction of some of her other desires (e. g. her desire to raise a child one day), but
deciding that she wants her desire to join the Peace Corps to move her to action (and
joining) does not make it harder or impossible for her to raise children one day—it just
postpones the satisfaction of it. She knows this, and so she is not uneasy with her

decision. She is not residually ambivalent. Of course, this is just one instance of

synchronic wholeheartedness. But imagine that she is the type of person who generally

98

does not experience competition or conﬂict among the desires that she identiﬁes with.
They are generally compatible; she is generally not ambivalent, she is generally
wholehearted in the synchronic sense.

A second type of wholeheartedness is what I will call “diachronic
wholeheartedness.” Diachronic wholeheartedness is where the person’s identiﬁcations
and orderings are consistent over time. It is the same at t2 and t3 as it is at t]. A person
who is diachronically wholehearted may decide at t; that she wants her desire for a very
successful career to be the desire that moves her to act (to be number one among her set
of ordered desires), and her desire for a rich social life to take a back seat. But she would
again arrive at or stand by this ordering ﬁve months ﬁ'om now (at t2), or even ﬁve years
from now (at t3). She is the sort of person who consistently puts work before pleasure. It
is not a decision that she is uneasy with because it has become a consistent part of who
she is.

A third type of wholeheartedness is what I will call “devotee wholeheartedness.”
Devotee wholeheartedness is where the person feels sure about and devoted to what she
identiﬁes with and the order of her identiﬁcations at t.. We all know examples of the
devotee. Persons who exhibit it are often referred to as zealots, or described as having
fanatic commitments. The devotee can often be found in political or religious settings.
Ambivalent they are not. They know where they stand and they are sure about it.

The fourth type of wholeheartedness is what I will call “mature
wholeheartedness.” In mature wholeheartedness, although the ordered set of desires may
have internal conﬂicts, the agent feels less distressed about them; she comes to accept, or

feel comfortable with what is lost by willing what she wills. Recall my example in

99

Chapter Two of the parent who is residually ambivalent about his decision to not have a
relationship with his drug addicted son. For him to resolve his ambivalence by becoming
maturely wholehearted would mean that he is less distressed by his decision; he is more
comfortable with, or at peace with, what he lost by willing what he wills.

Generally, diachronic wholeheartedness is thought to be desirable and devotee
wholeheartedness is thought to be undesirable. With regard to the former, we tend to
think that there is value in having a uniﬁed identity over time; in having the same values,
beliefs, and personality over time. The phrase, “She has changed so much” is often
accompanied by shocked exclamations or disparaging head-shaking. While the phrases,
“Still the same as she always was,” and “You can always count on her to stand by X” are
often accompanied by wide smiles or looks of admiration. I personally do not ﬁnd
diachronic wholeheartedness so admirable. I ﬁnd it static and boring; although I admit
that there are reasons why I would want others to be diachronically wholehearted.
Devotee wholeheartedness, on the other hand, is often considered undesirable. Persons
who exhibit it are often referred to as zealots and sirnpletons. Fanatic commitment to
anything, be it a political belief, a religious belief, a sports team, or a person, is often
quite ﬁightening. The remaining two types of wholeheartedness (synchronic and mature)
are interesting. Synchronic wholeheartedness seems to me something that cannot be
controlled much by the agent herself. Whether or not at a particular time the fulﬁllment
of some of the desires that a person identiﬁes with (e. g. living openly as a Lesbian, to
return to Lugones’s example) impedes the fulﬁllment of others that she identiﬁes with
(e. g. living openly as a Latina), seems to me to be largely up to the circumstances of the

society and the world in which the agent is living in. Mature wholeheartedness I ﬁnd to

100

be most interesting. It is admittedly difﬁcult to achieve, and I do not advise aiming for it
all of the time because too often a person may accept what is lost by willing what they
will, when they could have maybe found a way to satisfy desires that at one time seemed
incompatible. But, I do think that mature wholeheartedness is desirable in end of life
circumstances; and I will discuss this in Chapter Six.

Before closing, I want to note that although I have discussed the various
drawbacks to residual ambivalence, and the various types of wholeheartedness that may
in different ways resolve residual ambivalence, there is a sense in which residual
ambivalence is inevitable. Philip Koch argues for the near impossibility of resolving
residual ambivalence. Koch writes, “It’s really astonishing how feelings, or feeling-
shadows, endure: an old photo, a phrase, a stream of light striking us just so—and
suddenly the old longing is fresh and urgent. One realizes that it has lain there always,
though covered with layers of brush and dry leaves.”189 Koch argues that the presence of
residual ambivalence is similar to the presence of clouds; even if a cloud disappears,
there are still leftover vapors. Likewise, even if a person takes a stand about which desire
she wants to be effective in action (forms a will), there is still leftover conﬂiction. As
Koch says, “Resolving ambivalence can seem very much like trying to draw a map of the
sky.”190 Consider Simone DeBeauvoir’s description of the ambivalence present in her
[failed] relationship with her mother: “So our former relationship lived on in me in its
double aspect—a subjection that I loved and hated. It revived with all its strength when
Maman’s accident, her illness and her death shattered the routine that governed our

contacts. . .I thought I had made up my mind about our failure and accepted it; but its

 

‘89 Philip Koch, “Emotional Ambivalence,” p. 274.
19° Philip Koch, “Emotional Ambivalence,” p. 279.

101

sadness come back to my heart.”'91 Existentialist philosopher Soren Kierkegaard also
notes the inevitability of residual ambivalence (although he comes to quite harsh
conclusions about what this means!): “Satisﬁed, completely, absolutely satisﬁed in every
way, this one never is, and to be more or less satisﬁed is not worth the trouble, so it is
better to be completely dissatisﬁed. Anyone who has painstakingly pondered the matter
will certainly agree with me that it has never been granted to a human being in his whole
life, not even for as much as a half hour, to be absolutely satisﬁed in every conceivable
way.”192

In closing Chapter Five, let me note what I hope to have accomplished. Although
I have argued in Chapter Three that residual ambivalence does not preclude a person
from being autonomous, I admit that there are other drawbacks to residual ambivalence:
it harms the self, it harms others and makes personal relationships difﬁcult, and it puts a
person close to paralyzing ambivalence (which would result in a loss of metaphysical
autonomy). There are, on the other hand, beneﬁts to residual ambivalence: it helps with
personal growth, it helps with social relationships and in empathizing with others, and it
can secure metaphysical autonomy in the long run. Moreover, a certain amount of
residual ambivalence is inevitable anyway. As Amy Mullin writes, “Something in
between pathological fragmentation and strong integration is possible and actually

characterizes many people’s experience and may sometimes be preferable to strong

integration”193 Residual ambivalence is that “something in between.”

 

'9' Simone DeBeauvoir, A Very Easy Death (Pantheon, 1935): P- 103-

192 Soren Kierkegaard, “Repetition,” in The Essential Kierkegaard (Princeton University Press 2000), p.
109.

'93 Amy Mullin, “Selves, Diverse and Divided: Can Feminists Have Diversity Without Multiplicity?" p. 3.

102

Chapter Six: Ambivalence and End of Life Decisions

Thus far in the dissertation, I have argued that the term autonomy is used
differently in the applied philosophy literature (bioethics) than in the theoretical
philosophy literature, and that this difference has been largely ignored; that ambivalence
can either be paralyzing (such that a person cannot make a decision) or residual; that
paralyzing ambivalence results in a loss of autonomy; and that residual ambivalence does
not, despite the fact that most theoretical accounts of personal autonomy imply that it
does. I have also proposed ways for persons to resolve their paralyzing ambivalence so
that their autonomy may be restored. In the last chapter, I want to directly address the
case of ambivalence about end of life decisions. The issue is signiﬁcant because
ambivalence is often present in patients making end of life decisions, but little attention
has been paid to the phenomenon, and there is a danger of it being resolved by medical
paternalism.

The Context of End of Life Decisions

I want to spend some time conveying a sense of what the end of life is often like,
in order to paint a picture of the context in which end of life decisions are made (the
context in which the will is formed). The context lends itself to the experience of
ambivalence. So, it is no surprise that ambivalence (even the paralyzing sort) about end
of life decisions is common. The best way to convey a sense of what the end of life is
like is to hear from those who have experienced it for themselves, or from those who

have watched a loved one die.

103

Simone DeBeauvoir’s, A Very Easy Death, a book written about the death of her

mother, presents a poignant picture of the life of a dying person. Consider the following

quotes by DeBeauvoir:194

“I looked at her. She was there, present, conscious, and completely unaware of what
she was living through. Not to know what is happening underneath one’s skin is normal
enough. But for her the outside of her body was unknown to her—her wounded
abdomen, her ﬁstula, the ﬁlth that issued from it, the blueness of her skin, the liquid that
oozed out of her pores: she could not explore it with her almost paralyzed hands, and
when they treated her and dressed her wound her head was thrown back.”

“Then suddenly she cried out, a burning pain in her left buttock. It was not at all surprising.
Her ﬂayed body was bathing in the uric acid that oozed from her skin. ...All tense on the
edge of shrieking, she moaned, ‘lt burns, it’s awful; I can’t stand it. I can’t bear it any
longer.’ And half sobbing, ‘l’m so utterly miserable,’ in that child’s vice that pierced me to
the heart. How completely alone she was! I touched her, I talked to her; but it was
impossible to enter into her suffering. . . .Nothing on earth could possibly justify these
moments of pointless torment.”

In his book The Practice of A utonomy: Patients, Doctors, and Medical Decisions,

Carl Schneider talks to dying patients. A patient describes his/her experience: 195

“I am accosted by the everyday, overwhelmed by the mundane. And the symptoms are
terrifying; I break out in a hot sweat, become dizzy with the secret but powerful secretion of
adrenaline, my mind boils with the disparate thoughts as the world transforms itself into an
elaborate disaster. All I lcnow is that I am naked and alive. During the brief stretches of calm I
attempt to ponder my condition. . .all I now know is that I am deeply and irrevocably out of my
mind.”

Schneider makes a number of illuminating observations that paint a picture of the context
in which end of life decisions are made (the context in which the will is formed). These
quotes illustrate that end of life decisions are often made when patients are suffering, out

of their element, and processing incomplete and awkward information.196

 

19’ Simone DeBeauvoir, A Very Easy Death (Pantheon, 1985): pp. 77 and 81 (respectively).

'95 Carl Schneider, The Practice of Autonomy: Patients, Doctors, and Medical Decisions (Oxford
University Press 1999), p. 64.

'96 Schneider, The Practice of Autonomy, pp. 62, 50, xvi, and xvii (respectively).

104

“Simply being in the hospital can be intellectually disorienting and psychologically
debilitating, for you are deprived of much that is familiar and that makes you what you are,
and you are isolated ﬁom the world. the seasons. and even the time of dav.”

“. . .the information the doctor can present to the patient will ﬁ'equently and unavoidably be rife
with uncertainty, roiled by complexity, awkward to articulate, and hard to assimilate. This is
true even under the rare best of circumstances, where the doctor has taken serious time and
trouble to try to communicate with the patient and where the patient has reciprocated with
thoughtful attention, clarifying questions, and intelligent reﬂection.”

“As Daniel Chambliss observes, “Much of bioethics assumes that people are autonomous
decision makers sitting in a fairly comfortable room trying logically to ﬁt problems to given
solution-making patterns. The whole business is almost deliberately unreal...’”’

“And hyper-rationalism’s simpliﬁcations are particularly implausible in bioethics, a ﬁeld that
treats people in their least rational moments, in their most emotional travails, in their most
contextual complexity.”

Given this emotionally charged and uncertainty ridden context, it s not surprising
that ambivalence (even the paralyzing sort) about end of life decisions is common.
Supporting the view that ambivalence about end of life decisions is common is Lewis
Cohen et a]. who write: “Ambivalence is expected when individuals are faced with
momentous decisions. For example, when Chochinov et al. investigated whether
terminally ill patients desire death in order to prevent further suffering, they found the
desire for death to be transient; even in the less common cases in which patients had more
pervasive wishes to die, this desire was often unstable over time.”197 Author of The
Practice of Autonomy: Patients, Doctors, and Medical Decisions, Carl Schneider writes,

“. . .in important respects many of us have divided selves. One self may yearn to give up

 

'97 Lewis Cohen et al., “Ambivalence and Dialysis Discontinuation,” General Hospital Psychiatry Vol. 18

1996, p. 431.

105

the struggle for health and even life. But another self wants to be encouraged to

persist. . ..vacillate between a desire to persevere and a longing to succumb.”198

A Case of Ambivalence Regarding An End of Life Decision

I will now turn to the case of Mr. X; a speciﬁc case of ambivalence about an end
of life decision. This case was introduced in the preface as the motivation for the
dissertation and mentioned brieﬂy throughout. During the winter of 2004 I spent most of
my time interning at a major US hospital in the bioethics department. Of all of the cases
that l was involved in, one in particular stands out. Mr. X, a youthﬁrl 70 year old man,
was involved in a head on motor vehicle accident. As a result, he lay in a hospital bed
unconscious, on a ventilator, and paralyzed from the neck down. Mr. X’s wife and adult
son and daughter all agreed that there was no way that Mr. X would want to live this way.
In fact, they pointed to Mr. X’s advance directive to afﬁrm this View. Shortly thereafter,
Mr. X regained consciousness. To everyone’s surprise, Mr. X asserted that he wanted to
remain on the ventilator; he wanted to stay alive. The following day, we went in to talk
with Mr. X and he asserted that he did not want to remain on the ventilator; he wanted to
die. The family, the physicians, and the ethicists talked to Mr. X to try to ascertain what
he really wanted. These conversations spanned over weeks, but Mr. X continued to
alternate between asserting that he did want to remain on the ventilator and that he did
not want to remain on the ventilator. Mr. X was paralyzingly ambivalent: his conﬂict
prevented him from forming a will; from taking a position on which desire (desire to live
vs. desire to avoid suffering and poor quality of life) he wanted to have and to have be

effective in action.

 

'98 Schneider, The Practice of Autonomy: Patients, Doctors, and Medical Decisions, p. 89.

106

Resolving Mr. X’s Ambivalence in Ways that Promote Autonomy

There are various possible causes of Mr. X’s ambivalence. In talking with Mr. X
we may discover that he suffers from a basic epistemological deﬁcit; that he does not
really understand his prognosis. Or, in talking with him, we may discover that he suffers
from a more complex epistemological deﬁcit; that he has never really thought about these
issues and how the various courses of actions conﬂict or cohere with his other desires,
values, and religious beliefs. If Mr. X’s ambivalence is due to the fact that he does not
really understand what his progrosis is or what his life with that prognosis will be like,
then we should provide him with information that we think would help him resolve his
ambivalence. There is very helpful work being done now in the ﬁeld of psychology and
decision making about the various ways to present information to patients in ways that
they can understand it or and that it is salient to them, and also avoids known
misconceptions and biases such as failing to realize that 3/ 100 is equivalent to 30/1000.199

If Mr. X’s ambivalence is due to the fact that he has not really thought about these
issues and how the various courses of actions conﬂict or cohere with his other desires,
values, and religious beliefs (or even about what his other desires, values, and religious
beliefs are); then we should help him to think about these things and about how each
course of action does or does not cohere with his endorsed values and desires. For
example, we can talk with Mr. X about how much he values independence, about his
religious views, about his views on what makes life valuable, about his family’s opinions.
We can help him discover the extent to which the different courses of action uphold or

conﬂict with his values and beliefs. Another way to approach the conversation is to focus

107

not so much on which course of action can support more of his values, but on which
course of action he feels will bring out the best in him. Harry Frankfurt has written that
“. . .the decision of what to identify yourself with is not made on the basis of an evaluation
of the various object, it is not which is more or less valuable or important than the other;
the primary basis for the decision lies in what the person feels he an live most fully as,
what will bring out the best in him, what will enable him to realize most completely the
capacities for an invigorating and robust emotional and intellectual and volitional life.”200
Now of course, we are concerned about people who are making end of life decisions, and
Frankfurt is referring to deciding how to live life, but perhaps we can apply this line of
thought to dying, and suggest to the patient that he base his decision on how he wants to
die, recognizing that dying is part of living.

Filling epistemological gaps may in some cases be effective in resolving
ambivalence, but there are many reasons to think that it won’t always work. First of all,
much of medicine is uncertain. There may be great limitations to how much information
we can actually provide a patient. Second, the decisions involved in end of life situations
are just not the sort of decisions that can be made cognitively by receiving information
and reﬂecting on values and ways of living and dying. They are very much experientially
based. Often, persons have never experienced anything like this before. I do not know
what it is like to die, for my body to be medicalized; I do not know how my family will

be affected by all of this, etc. As Schneider writes, “Data take patients only so far. Often

“values” take them little farther. An AIDS patient commented, “It was time to make a

 

‘99 For examples of this see the work being done by Peter Ubel MD. and his group at the Center for
Behavior and Decision Sciences in Medicine.
20° “A Discussion With Harry Frankfurt,” Ethical Perspectives, p. 27.

108

choice. All of the questions in the world, answered and unanswered, couldn’t help

measaaZOl

If informing the patient of the consequences of each course of action and helping
her reﬂect on how each course of action coheres with her values or represents ways of
living and dying does not resolve the patient’s paralyzing ambivalence then we must
attempt a different method of resolution. Return to the AIDS patient’s quote: “It was
time to make a choice. All of the questions in the world, answered and unanswered,
couldn’t help me.” When a patient reaches this point and expresses this sentiment, we
should invite the patient to make a leap in one direction or the other. In Chapter Four I
argued that we should look at the courses of action that the person is torn between, judge
which one we think best coheres with her other preferences and values, and invite her to
take a leap in that direction. This is exactly what happened with Mr. X. After months of
ambivalence and vacillation, a physician said to him, “Mr. X, how about we send you
home with a hospice nurse?” Mr. X hesitantly nodded his head, was sent home, and died

days later.

In Chapter Four, I spent a good deal of time arguing that the invited leap is indeed
a method of resolution that promotes autonomy, so I will not re-hash those arguments
here. I encourage the reader to refer back to those arguments, which address the concern
that inviting this leap for Mr. X is not an empowerment of his autonomy, but rather a
form of manipulation. What I will do is offer two suggestions for helping the patient to
take this invited leap. Both of Mr. X’s options were frightening to him; especially the

course of action of stopping treatment and dying. If that is the direction that physicians

 

20' Schneider, The Practice of Autonomy: Patients, Doctors, and Medical Decisions, p. 73

109

and family members invite Mr. X to leap towards, then they can assure Mr. X that they
will be there to support him and to soften the blow. The most concrete way to do this is
through the services of hospice care. Hospice care can reduce feelings of ﬂight and
helplessness, and also reduce pain and suffering. The following two quotes are excerpts
ﬁ'om stories of persons who have received hospice care. They illustrate how hospice care
gave them a secure, comfortable place to fall once they had made the decision to let go;
which makes letting go much less frightening.202

“With the guidance of the hospice staff, we were able to change medications, food, and

drink to meet her needs. . .Being able to call a knowledgeable nurse for advice when my

Mother was either anxious or in pain, and then be given the instructions and resources to

administer medication or try a different comfort measure took away the helpless
feelings...”

“He had one special nurse that had the 8:00-4:00 shift and she fussed over him and
spoiled him rotten always combing his hair etc. She persuaded him to go in the hospital
bed that was set up in the family room. At this point we knew time was slowly running
out. . .that night he quietly passed away. Hospice gave him the quality of life for his last
days that he would not have had without them.”

The second suggestion for helping a patient take the invited leap is a type of
therapy called logotherapy. Lo gotherapy is a type of therapy inspired by existentialist
philosophy, created by Viktor Frank]. It has three basic tenets: (1) Life has meaning
under all circumstances, even the most miserable ones, (2) Our main motivation for
living is our will to ﬁnd meaning in life, and (3) We have ﬁeedom to ﬁnd meaning in
what we do, and what we experience, or at least in the stand we take when faced with a
situation of unchangeable suffering.203 I suggest that this method of therapy may help

patients achieve the sense of freedom and conﬁdence that they need in order to take the

 

202 www.hospicefoundation.org

110

invited leap in one direction or the other. In fact, an exploratory study on the use of
logotherapy on dying patients found that patients reported a greater sense of freedom
ﬁom the therapy. 20" Moreover, a rabbi reported that logotherapy helped a particular
dying patient that he was dealing with: "Much of the theological discussion made little
impression upon her, but attitudinal values invited her curiosity... She resolved then and
there if she could not avoid the inescapable suffering, she would determine the manner

and mode in which she would meet the illness. She became a tower of strength ....”2°5

Resolving Mr. X ’s Ambivalence in Ways that Do NOT Promote Autonomy

Unfortunately, ambivalent patients are in danger of being treated
patemalistically., They are confused and out of their element, and it is much easier for
busy physicians to behave in ways that do not promote autonomy than it is for them to
take the time to help the patient discover what he really wants. There are two methods
that may resolve a patient such as Mr. X’s ambivalence, but do not promote his
autonomy. The ﬁrst is to just let Mr. X’s ﬁrst order desires, pressures, and fears (things
that he just ﬁnds himself having, but not necessarily endorsing) take over and lead him to
one decision or another. When a patient is suffering, they enter "a state of severe distress
associated with events that threaten the intactness of the person."206 They begin to feel
themselves coming undone, they feel the situation and the fear and suffering taking over.
“. . .decisions may be warped by fear, by panic, by a passing preference for short-term

comfort, by pain, by bitterness, by guilt, by depression, by despair. . .fear, especially,

 

20’ http://www.logotherapyinstituteorg, For more on Frankl’s work, see his most well-known book, Man 's
Search for Meaning.

20’ Zuehlke TE, Watkins IT “The Use of Psychotherapy with Dying Patients: An Exploratory Study,”
Journal of Clinical Psychology Vol. 31 1975, pp. 729-732.

205 Kenneth Woodroffe, “Logotherapy (ﬁrst lecture),” http://www.jcagpc.or2/~ivuzo/Logotherapvl .htm.

Ill

torments patients. ...Fear shatters the mind’s clarity. . . ”207 If Mr. X decided to have the
ventilator removed simply out of extreme anxiety and if it is the case that Mr. X generally
does not endorse acting out of fear, then he is not metaphysically autonomous with
respect to that decision. As Frankfurt puts it, “A person’s will may be overpowered and
violated by forces, such as those of anxiety or addiction, that are generated within him
but that are nonetheless not in the fullest sense his own. They are forces with which he
does not identify. . "”208 For physicians to simply let Mr. X’s unendorsed ﬁrst order fear
to take over and be the basis of his decisions without any further conversations with Mr.
X would be to let him resolve his ambivalence in a way that does not promote autonomy.
Now, someone might object that at this point in Mr. X’s life, being autonomous
does not matter quite as much as his well-being. Even if Mr. X is acting out of
unendorsed ﬁrst order fear, it might be that he is making a decision that his family and
the physicians think is best for him. So why not go along with his decision when the
alternative is to prolong his ambivalence, and the suffering? Why should the imperative
be to make Mr. X live free no matter the costs?209 In response, I would reply that we are
not making him live free—we’re giving him the chance to live free. We are not forcing
him to act in a way that results in him being autonomous; we are giving him a chance to
act in a way that results in him being autonomous. In Chapter One, I reviewed the
arguments for why autonomy is valuable and something to strive for. Let me brieﬂy
repeat that it has been argued that letting people we act freely and choose for themselves

is what is most likely to enable people to secure their own well-being, and because

 

20” E.J. Cassell, “The Nature of Suffering and the Goals of Medicine,” New England Journal of Medicine
1982 306211, pp. 639-645.

207 Schneider, The Practice of Autonomy: Patients, Doctors, and Medical Decisions, p. 63.

208 Frankfurt, “Rationality and the Unthinkable,” p. 183.

112

exercising their own autonomy/making their own choices (even if they later regret them,
or they are not in their best interests) is part of well-being itself.

The second method that might resolve Mr. X’s paralyzing ambivalence, but
would not promote his autonomy would be to choose a course of action for Mr. X and
force him to do it, or to choose a course of action for Mr. X and implement it without its
knowledge. For example, we may judge that the option of discontinuing the ventilator is
the one that best coheres with Mr. X’s other desires, goals, and values. So, we decide to
remove Mr. X ﬁ'om the ventilator despite him crying in protest. Or, alternatively, we
sedate him and withdraw the ventilator without him knowing. As I argued in Chapter
Four, not only might this method not actually work to resolve the ambivalence and hence
restore metaphysical and bioethical autonomy (Mr. X may still feel torn despite the fact
that we have decided for him and are taking steps to remove the ventilator); but it fails to
distinguish between autonomy as an ideal and autonomy as a right. As I argued in
Chapter One, one of the purposes of metaphysical autonomy theories is to develop an
ideal of the person as someone who acts on desires that she identiﬁes with, or that cohere
with her other values, and avoids acting on ones that she repudiates; whereas bioethical
autonomy focuses on a person’s right to make her own decisions. It might be true that
forcing Mr. X off of the ventilator (or removing him without his knowledge) moves him
closer to that ideal metaphysically autonomous person (because, for example, that is the
course of action that best coheres with his other preferences and values). But the reason
that this seems ethically egregious to us is because it interferes with his right to make his

own decisions.

 

209 I thank Tom Tomlinson for raising this objection.

113

Before moving on, I want to brieﬂy mention how I think that it might be possible
for the method of inviting a leap to be used in a way that does not promote autonomy.
And that is if the leap is invited too early (assuming it is not the case that the person
generally endorses early leaping as a method for making decisions). If the leap is offered
too early (i.e. before attempts have been made to ﬁll basic and complex epistemological
gaps), then it is more likely that a person will act based on their ﬁrst order desires and not
based on what they really want. An analogous example can be found in a very strange
place: the movie “A Christmas Story.” The young child Ralphie really wants a BB gun
for Christmas, however, when he reaches Santa’s lap he has a moment of paralysis and
cannot communicate to Santa what he really wants for Christmas. Santa waits about 10
seconds and then invites Ralphie to ask for a football. Ralphie slowly nods his head,
“Yes, a football.” Until. . .”Football? Football? What's a football? A football? Oh no,
what was I doing? Wake up, Stupid! Wake up! No! No! I want an Ofﬁcial Red Ryder
Carbine-Action Two-Hundred-Shot Range Model Air Riﬂe!” Ralphie was invited to leap
too early and as a result almost made a mistake in communicating what he really wanted.
Choosing Not To Choose

I want to have a special section just for this method of resolving paralyzing
ambivalence, because it is so common in health care. Dying patients often do choose to
not choose. They may choose to leave the decision up to their doctors, or their family
members.210 As Simone DeBeauvoir succinctly writes about her dying mother, “Maman

did not want these intimate conversations. What she wanted to see round her bed was

 

210 This is one of the central points of Schneider’s The Practice of Autonomy: Patients, Doctors, and
Medical Decisions.

ll4

young smiling faces.”2 1’ In Chapter Four, I argued that not choosing can indeed be a way
to resolve paralyzing ambivalence that promotes autonomy. So, one way that Mr. X
could have resolved his paralyzing ambivalence is to turn his choice over to his wife. As
I discussed in Chapter Four, there is, however, a difference between Mr. X viewing this
as an escape from choice and Mr. X recognizing that his choosing not to choose and to
relinquish the decision to his wife is indeed a choice. If Mr. X is viewing not choosing as
an escape, then even though he is acting autonomously, he is practicing what Sartre
would call “bad faith.” But if he sees himself as constructing himself in front of others as
someone who chooses to let his physicians or his family make decisions about the end of
his life and accepts the consequences and judgments put on him by this choice, then he is
not only acting autonomously but is also practicing good faith.
Remaining Issues

There are a couple of larger issues to be addressed about ambivalence regarding
end of life decisions: (I) What if the person resolves their paralyzing ambivalence by
making a decision that does not match her life narrative? (2) What are we to say about
cases where a patient dies before resolving her paralyzing ambivalence, and (3) What are
we to say about cases of residual ambivalence regarding end of life decisions? Let me

begin with addressing the ﬁrst issue. Consider the following case:212

 

2” Simone DeBeauvoir, A Very Easy Death, p. 90.
2'2 Atul Gawande, “Whose Body Is It, Anyway?” in Complications: A Surgeon ’s Notes on An Imperfect
Science (New York: Henry Holt and Company, 2002), pp. 208-210, 212-215.

115

Mr. Lazaroff had seen his doctor about a backache. The doctor initially found nothing suspicious, but
three months later the pain had worsened and a scan revealed extensive cancer; multiple tumors in
Lazarofl’s liver, bowel, and up and down his spine. A biopsy revealed it was an untreatable cancer.
Lazaroff was only in his early sixties, a longtime city administrator with the hardened manner of a
man who had lost his wife a few years earlier and learned to live alone. His condition deteriorated
rapidly. In a matter of months, he lost more than ﬁfty pounds. As the tumors in his abdomen grew, his
belly, scrotum, and legs ﬁlled up with ﬂuid. The pain and debility eventually made it impossible for
him to keep working. His thirty-something son moved in to care for him. Lazaroff went on around-
the-clock morphine to control his pain. His doctors told him that he might have only weeks to live.
Lazaroff wasn’t ready to hear it, though. He still talked about the day he’d go back to work. Then he
took several bad falls; his legs had become unaccountably weak He also became incontinent. He
went back to his oncologist. A scan showed that a metastasis was compressing his thoracic spinal
cord. The oncologist admitted him to the hospital and tried a round of radiation, but it had no effect.
Indeed, he became unable to move his right leg; his lower body was becoming paralyzed. He had two
options left. He could undergo spinal surgery. It wouldn’t cure him--surgery or not, he had at the
most a few months left-but it offered a last-ditch chance of halting the progression of spinal-cord
damage and possibly restoring some strength to his legs and sphincters. The risks, however, were
severe. The doctors would have to go in through his chest and collapse his lung just to get at his
spine. He’d face a long, difﬁcult, and painful recovery. And given his ﬁ'ail condition his chances of
surviving the procedure and getting back home were slim. The alternative was to do nothing. He’d go
home and continue with hospice care, which would keep him comfortable and help him maintain a
measure of control over his life. The immobility and incontinence would certainly worsen. But it was
his best chance of dying peaceﬁrlly, in his own bed, and being able to say good-bye to his loved ones.
Lazaroff wanted surgery. . . .Lazarofl’s son David arrived... .Outside the room, David told Dr.
Gawande that he wasn’t sure this was the right move. His mother had spent a long time in intensive
care on a ventilator before dying of emphysema, and since then his father had often said that he did
not want anything like that to happen to him. But now he was adamant about doing “everything”. . .
Although the operation was a technical success, Lazaroff soon developed many severe complications
and his condition quickly worsened. He died exactly the way he hadn't wanted to die: strapped down
and sedated, tubes in every natural oriﬁce and in several new ones, and on a ventilator.

Mr. Lazaroff had lived his life as an independent person who consistently insisted that he

did not want to die the way his wife did—in a hospital and on a ventilator. Imagine that

there was a period of paralyzing ambivalence before Mr. Lazaroff made his decision in

favor of the surgery. So, Mr. Lazaroff resolved his paralyzing ambivalence, but he did so

in favor of a decision that let us say does not seem consistent with the way that he had

lived his whole life—independent, sticking with his principles, etc. There are a couple of

responses to this concern. First, I would argue that what matters is why Mr. X has made

the decision that he has. Someone who resolves their paralyzing ambivalence by making

a decision that surprises us should certainly be cause for concern and investigation. For

116

example, is Mr. Lazaroff making the surprising decision out of fear (which I have argued
would reduce his metaphysical autonomy with respect to that choice) that perhaps we
could attempt to alleviate so that his decision is more truly his own? Does Mr. Lazaroff
really understand his prognosis? There is reason to think that he is in denial and does not
really appreciate his prognosis, and understanding is a necessary condition for decision
making capacity and bioethical autonomy. Perhaps we could attempt to reduce his denial
and help him understand his prognosis in a way that it is salient to him. But assuming
that these concerns are not present or have been addressed, people do change their minds,
especially after experiencing new situations and ways of living. Which leads me to my
second response to the concern about a decision that does not match a life narrative; and
that is that the very idea of a coherent life narrative is problematic. Problematic in two
ways: descriptively and normatively. Descriptively, I ﬁnd it to be quite inaccurate. As
much as I might try, I fail to see much coherence in myself or others. As Nancy Johnson
et al. write: “Such a deﬁnition accords with our ﬁnding that the construction of an
overarching story within which details and events are organized neatly to form a coherent
and meaningful structure is seldom realized in the context of end-of-life narratives. The
narratives we describe, in contrast, are perhaps better understood as narratives-in-the-
making or narrative fragments. They are often incomplete and rarely “neat.” The notion
of constructing an “overarching story” is itself an idealized experience that narrators
strive for because of its promise to bring desperately desired coherence, meaning, and a

conﬁdent basis for action.”213 The normative idea that an overarching story is something

 

2'3 Nancy Johnson, Deborah Cook, Mita Giacomni, and Dennis Willms. “Towards A “Good” Death: End-
Of-Life Narratives Constructed In An Intensive Care Unit” in Culture, Medicine and Psychiatry Vol. 24
2000, pp. 275-295.

117

that should be present is also problematic. As I argued in Chapter Five, there is not a
necessary connection between living consistently and living well.
Dying With Paralyzing Ambivalence Regarding End of Life Decisions

The second issue is what to say about a patient who dies before resolving her
paralyzing ambivalence. There would be two facts about this case: the person dies in a
non-autonomous state, and if she is still conscious enough she dies in a non-peaceful
state. The latter fact is easier to see as being important. For most of us, the inability to
decide which of our desires we want to be effective in action when we have to is
internally tormenting. Especially when it is a decision that has major consequences, or
that the person sees as important or self-deﬁning. Of course, not all persons will see end
of life choices as particularly important or self-deﬁning; but many will. Cultural
inﬂuences cause us to see certain of our choices as being particularly important, or
deﬁning (e. g. choices about marriage, children, religion—in particular, a woman’s choice
about children is often seen as more important or deﬁning by much of society than her
choice about education). Like marriage and children, death is portrayed by our culture as
a major life event, and so many take the details of that event to be important and even
deﬁning. Simone DeBeauvoir writes about her dying mother: “I did not particularly want
to see Maman again before her death; but I could not bear the idea that she should not see
me again. Why attribute such importance to a moment since there would be no memory?
There would not be any atonement either. For myself I understood, to the innermost ﬁbre
of my being, that the absolute could be enclosed within the last moments of a dying
person.”2”. Thus, a patient who is paralyzingly ambivalent about an end of life choice is

likely experiencing a lot of internal turmoil, and so to die without the paralyzing

118

ambivalence resolved is to die in the midst of turmoil. But, as I mentioned, there is
another fact about the patient who dies before her paralyzing ambivalence is resolved:
she dies in a non-autonomous state. But, what is so bad about this, one might ask. After
all, in Chapter One I explained the traditional arguments for the value of metaphysical
autonomy: Kant’s argument that it is what makes us persons, and Mill’s argument that it
contributes to well-being. Kant’s claim is certainly meant to refer to our general ability
to exercise autonomy; he did not mean that if there is a particular instance where a person
does not exercise her autonomy then she is no longer a person. So, it is hard to see what
would be wrong with dying in a non-autonomous state; it is only one particular instance
of a failure to exercise autonomy. And Mill’s point about autonomy contributing to well-
being dose not really apply either, for the person is about to die—there is no future well-
being to be concerned about. Thus, I do not think that there is anything bad about dying
in a non-autonomous state such as paralyzing ambivalence per se.
Dying With Residual Ambivalence Regarding End of Life Decisions

If a patient does resolve her paralyzing ambivalence regarding the end of life
decision that she is struggling with, she will most likely be left with residual ambivalence
with respect to that decision; she may still feel moved towards the other course of action.
As I argued in Chapter Three, residual ambivalence does not take away metaphysical
autonomy or bioethical autonomy. In Chapter Five, I argued that it is often present to
some extent, and that it has its beneﬁts. We should take care to assure the dying patient
that the residual ambivalence that they may be experiencing is normal. As I explained in
Chapter Five, there are however drawbacks to residual ambivalence. The one most

relevant here is the psychological stress caused by residual ambivalence. Patients at the

 

2” Simone DeBeauvoir, A Very Easy Death, p. 62.

119

end of their lives are often already under a great deal of stress, and so minimizing
residual ambivalence may be desirable. The opposite of ambivalence is
wholeheartedness, but as I argued in Chapter Five, there are different types of
wholeheartedness that a person could strive for. The type of wholeheartedness that I
called mature wholeheartedness may be particularly valuable to patients near the end of
life. In mature wholeheartedness, although the person still feels pulled towards the other
course of action (the one that she did not decide on or will), she comes to accept or feel
comfortable with what is lost by willing what she wills. Imagine a 55 year old Oregon
(NB: Oregon is the only state in the US. where physician assisted suicide is legal)
woman who is terminally ill with cancer and is ambivalent about when she should take
the lethal substance that her physician has prescribed to her. Imagine that she resolves
her paralyzing ambivalence and decides to take it tomorrow. Even though she has ﬁnally
exercised her will and made a decision, she is still left with a good amount of residual
ambivalence about the decision. For her to resolve her residual ambivalence by
becoming maturely wholehearted would mean that she is less distressed by her decision;
she is more comfortable with, or at peace with the extra days lost by willing what she
wills. Patients who are able to achieve mature wholeheartedness will be more at peace
with their end of life decisions.

In closing, in this dissertation I hope to have illustrated the way that a particular
phenomenon—paralyzing ambivalence—blocks a person from exercising her will, and
hence takes away metaphysical and bioethical autonomy. I have proposed methods of
resolution that are compatible with restoring and respecting the person’s metaphysical

and bioethical autonomy. I have also dealt with an area where paralyzing ambivalence is

120

common: end of life decision making. I hope to have given clinicians tools to respond to
ambivalence about end of life decision making in a way that respects patient autonomy
and avoids medical paternalism. I also hope to have given dying persons themselves the
tools to respond to their own ambivalence about end of life decisions. More generally, I
hope to have made a valuable connection between the concept of autonomy that is used
in the theoretical philosophy literature (“metaphysical autonomy”) and the concept of
autonomy that is used in bioethics (“bioethical autonomy”). Metaphysical autonomy
centers around a person examining her ﬁrst order desires and deciding which of those she
wills and bioethical autonomy focuses on allowing a person to exercise her will with
respect to her medical decisions so long as she understands the consequences of doing so.
The valuable connection between the two types of autonomy is that a richer respect for
bioethical autonomy involves not just helping patients understand their choices, but also
helping them discover what they will—helping them think about their ﬁrst order desires
in light of their other desires and values. Autonomy theories have much more to
contribute to bioethics than simply warding off medical paternalism; they can even help

us bring peace to patients at the end of their lives.

121

BIBLIOGRAPHY

Arpaly, N. (2003). Unprincipled Virtue: An Inquiry into Moral Agency. New York,
Oxford University Press.

Arpaly, N. (2005). Responsibility, Applied Ethics, and Complex Autonomy Theories.
Personal Autonomy: New Essays on Personal Autonomy and Its Role in
Contemporary Moral Philosophy. J. S. Taylor. Cambridge, Cambridge University
Press: 162-180.

Aeschylus. (2008). Agamemnon. LeClue22.

Bauer, M. (2003). The Field Guide to Psychiatric Assessment and Treatment.
Philadelphia, Lippincott/Williams and Wilkins.

Beauchamp, T and Childress, J. (1994). Principles of Biomedical Ethics. New York,
Oxford University Press.

Beauchamp, T. (2005). Who Deserves Autonomy, and Whose Autonomy Deserves
Respect? Personal Autonomy: New Essays on Persogil Autonomyand Its Role in
Contemporary Moral Philosophy. J. S. Taylor. Cambridge, Cambridge University
Press: 310-329.

Benson, P. (1990). “Feminist Second Thoughts about Free Agency.” Hypatia 5: 47-64.

Benson, P. (1991). "Autonomy and Oppressive Socialization." Social Theory and
Practice 17(3): 385-408.

Bratman, M. (1999). Identiﬁcation, Decision, and Treating as a Reason. Faces of
Intention: Selected Essays on Intention and Agency. M. Bratman. New York,
Cambridge University Press: 185-206.

Bratman, M. (2000). “Reﬂection, Plamring, and Temporally Extended Agency.” The
Philosophical Review 109(1): 3 5-61 .

Bratman, M. (2002). Hierarchy, Circularity, and Double Reduction. Contours of Agency:
EssaLs on Themes from Harry Frankfurt. S. Buss. Cambridge MA, Bradford
Book/MIT Press: 65-90.

Bratman, M. (2003). “Autonomy and Hierarchy.” Social Philosophy and Policy. 2(2):
156-176.

Bratman, M. (2005). Planning Agency, Autonomous Agency. Personal Autonomy: New
Essays on Personal Autonomy and Its Role in Contemporary Mogral Philosoply. J.
S. Taylor. Cambridge, Cambridge University Press: 35-57.

Buchanan, A and Brock, D. (1989). Deciding For Others: The Ethics of Surrogate

122

Decision Making. New York, Cambridge University Press.
Buss, S. (1994). “Autonomy Reconsidered.” Midwest Studies in Philosophy 19: 95-121.

Calhoun, C. (1995). "Standing for Something." The Journal of Philosophy 92(5): 235-
260.

Camus, A. (1989). The Stranger. New York, Vintage International.

Cassell, E. (1982). “The Nature of Suffering and the Goals of Medicine.” New England
Journ_al of Medicine 306(11): 639-645.

Chekhov, A. (1991). The Cherry Orch_aLd_. New York, Dover Publications.
Chen, X. (2003). Beinwd Authenticity. New York, Rodopi.

Christman, J. (1991). "Autonomy and Personal History." Canach Journal of Philosophy
21(1): 1-24.

Christman, J. (1993). “Defending Historical Autonomy: A Reply to Professor Mele.”
Canadian J oumal of Philosophy 23(2): 281-289.

Christman, J. (1995). Feminism and Autonomy. Mgging Questions: Feminist Ethics in
Every@ Life. D. Bushnell. Lanham, Md, Rowman & Littleﬁeld.

Christman, J. (2001). “Liberalism, Autonomy, and Self-Transformation.” Social Ther
and Practice 27(2): 185-206.

Cohen, G. (1996). Reason, Humanity, and the Moral Law. The Sources of Nonnatim C.
Korsgaard. New York, Cambridge University Press.

Cohen, L. et al. (1996). “Ambivalence and Dialysis Discontinuation.” General Hospital
Psychiat_ry 18(6): 431-43 5.

Cuypers, S. (1998). “Harry Frankfurt on the Will, Autonomy, and Necessity.” Ethical
Perspectives 5(1): 44-52.

DeBeauvoir, S. (1985). A Very Easy Death. New York, Pantheon.

Double, R. (1992). “Two Types of Autonomy Accounts.” Canadian Journal of
Philosophy 22(1): 65-80.

Dworkin, G. (1970). “Acting Freely.” Nous 4: 367-383.

Dworkin, G. (1976). “Autonomy and Behavior Control.” Hastings Center Raport 6: 23-
28.

123

Dworkin, G. (1988). The Theory and Practice of Autonomy. New York, Cambridge
University Press.

Ekstrom, L. W. (1993). “A Coherence Theory of Autonomy.” Philosophy and
Phenomenological Resea_rc_h_ 53(3): 599-616.

Ekstrom, L. W. (2005). Autonomy and Personal Integration. Persorﬂ Autonomy: New
Essg/s on Personal Autonomy and Its Role in Contemporary Moral Philosophy. J.
S. Taylor. Cambridge, Cambridge University Press: 143-161.

Fisher, J. M., Ed. (1993). The Metaphysics of Death. Stanford, Stanford University Press.
Feinberg, J. (1986). Harm to Self. New York, Oxford University Press.

Frankfurt, H. (1971). "Freedom of the Will and the Concept of a Person." Journal of
Philosophy 68: 5-20.

Frankfurt, H. (1976). Identiﬁcation and Extemality. The Importance of What We Care
About: Philosophical Essays, H. Frankfurt. New York, Cambridge University
Press: 58-69.

Frankfurt, H. (1987). Identiﬁcation and Wholeheartedness. The Importance of What We
Care About: Philosophrﬁil Essays. H. Frankfurt. New York, Cambridge
University Press: 159-177.

Frankfurt, H. (1988). Rationality and the Unthinkable. The Importance of What We Care
About: Philosophical Essays. H. Frankfurt. New York, Cambridge University
Press: 177-190.

Frankﬁnt, H. (1982). The Importance of What We Care About. The Importance of What
We Care About: Philosghical Essays. H. Frankfurt. New York, Cambridge
University Press: 80-94.

Frankﬁrrt, H. (1992). The Faintest Passion. Necessity. Volition, and Love. H. Frankfurt.
New York, Cambridge University Press: 95-108.

Frankfurt, H. (1994). Autonomy, Necessity, and Love. Necessity, Volition, and Love. H.
Frankfurt. New York, Cambridge University Press: 129-142.

Frankfurt, H. (1998). “Discussion With Harry Frankfurt.” Ethical Perspectives 5(1): 30-
35.

Frankﬁrrt, H. (2004). The Reasons of Love. Princeton, Princeton University Press.

Frankfurt, H. (2006). Taking Ourselves Seriously. Taking Ourselves Seriouslyand

124

Getting It Rigm. H. Frankfurt. Stanford, Stanford University Press: 1-27.

Frankfurt, H. (2006). Getting It Right. Taking Ourselves SerioushLand Getting Right.
H. Frankﬁnt. Stanford, Stanford University Press: 27-54.

Friedman, M. (2003). Autonomy, Gender, Politics. Oxford, Oxford University Press.

Gawande, A. (2002). Whose Body Is It, Anyway? Complication_s: A Surgeon’s Notes on
An Imperfect Science. New York, Henry Holt and Company.

Greenspan, P. (1980). A Case of Mixed Feelings: Ambivalence and the Logic of
Emotion. Explaining Emotions. A.O. Rorty. Berkeley, University of California
Press: ?-?.

Harrist, S. (2006). “A Phenomenological Investigation of the Experience of
Ambivalence.” Journal of Phenomenological Psychology 37(1): 85-114.

Haworth, L. (1991). “Review: Dworkin on Autonomy.” Ethics 102(1): 129-139.

Hill Jr, T. E. (1992). The Kantian Conception of Autonomy. Dignity and Practical
Reason in Kant's Moral Theory. T. E. Hill Jr. Ithaca, Cornell University Press:
76-96.

Johnson, N. et al. (2000). “Towards A “Good” Death: End-Of-Life Narratives
Constructed In An Intensive Care Unit” Culture. Medicine and Psychim 24(3):
275-295.

Kierkegaard, S. (1983). Fear and Trembling. m and Trembling/Repetition:
Kierkegaard's Writings, E. Hong and H. Hong. Princeton, Princeton University
Press.

Kierkegaard, S. (2000). Repetition. The Essential Kierkegaard. E. Hong and H. Hong.
Princeton, Princeton University Press.

Koch, P. (1987). "Emotional Ambivalence." Philosophy and Phenomenological Research
48(2): 257-279.

Korsgaard, C. (1996). The Sources of Norrnativity. New York, Cambridge University
Press.

Kristinsson, S. (2000). “The Limits of Neutrality: Toward a Weakly Substantive Account
of Autonomy.” Canadian Jomal of Philosophy 30(2): 257-286.

Kunkel, B. (2005). Indecision. New York, Random House Trade Paperbacks.

Lewis, CS. (2001). Mere Christianity. San Francisco, Harper Collins Publishers.

125

Lugones, M. (1987). "Playfulness, World-traveling, and Loving Perception." Hypatia
2(2): 3-19.

May, T. (1994). “The Concept of Autonomy.” American Philosophical Quarterly. 31(2):
133-145.

McFall, L. (1987). "Integrity." Ethics 98: 5-20.

Mele, A. (2001). Autonomous Agents: From Self-Control to Autonomy. New York,
Oxford University Press.

Meyers, D. (2000). Intersectional Identity and the Authentic Self. Relational Autonomy:
Feminist Perspectives on Autonomy, Agency, and the Social Self. N. S. Catriona
MacKenzie. New York, Oxford University Press: 151-180.

Meyers, D. (2004). Gender In the Mirror: Cultugrl Imagery and Women’s Agency New
York, Oxford University Press.

Mill, IS. (2008). On Libm. bnpublishing.com.

Mullin, A. (1995). "Selves, Diverse and Divided: Can Feminists Have Diversity Without
Multiplicity?" Hypatia 10(4): 1-31.

Orwell, G. (1961). 1984. New American Library; Reissue edition.

Oshana, M. (1998). “Personal Autonomy and Society.” Journal of Social Philosgrhy 29:
81-102

Petty, R. (2006). “Implicit Ambivalence from Attitude Change: An Exploration of the
PAST Model.” Journal of Personality and Social Psycholggy. 90(1): 21-41.

Sartre, J. P. (1948). Anti-Semite and Jew. New York, Schocken Books.

Sartre, J. P. (1956). “Existentialism is A Humanism.” Existentialism from Dostoevsky to
Sartre. New York, World Publishing Company.

Schneider, C. (1999). The Practice of Autonomy: Patients. Doctors. and Medical
Decisions. New York, Oxford University Press.

Schneider P. L. and Bramstedt K. A. (2006). “When Psychiatry and Bioethics Disagree

About Patient Decision Making Capacity.” Journal of Medical Ethics 32(2): 90-
93.

Schwartz, A. W. (2004). "Minimal Rationality and Self-Transforrnation: Christrnan on
Autonomy." Social Theory and Practice 30(2): 215-228.

126

Stoljar, N. (2000). “Autonomy and the Feminist Intuition.” Relational Autonomy:
Feminist Perspectives on Autonomy, Agency, and the Social Self. N. S. Catriona
MacKenzie. New York, Oxford University Press: 94-111.

Taylor, J. S. (2000). "The Theory of Autonomy. " Humane Studies Review 12(3).
Taylor, J. 8., Ed. (2005). Personal Autonomy: New Essays on Personal Autonomy and Its
Role in Contemporary Moral Philosophy. Cambridge, Cambridge University

Press.

Tormey, J. (1983). "The Ambivalent Self." Philosophic Exchange: Annual Proceedings
14: 109-120.

Waugh, E. (2003). Declineﬁ and Fall. Penguin Classics; New Ed edition.

Weigert, A. (1991). Mixed Emotions: Certain Steps Toward Understanding
Ambivalence. New York, State University of New York Press.

Wittgenstein, L. (1973). Philosophical Investigations. Penguin Classics; New Ed edition.

Wolf, S. (1987). Sanity and the Metaphysics of Responsibility. Responsibility, Character,
and the Emotitﬂ. F. Schoeman. New York, Cambridge University Press: 46-62.

Wolf, S. (1990). Freedom within Reason. New York: Oxford University Press.

Woodroffe, K. “Logotherapy (ﬁrst lecture)?’
Imp://www.ica.apc.org/~iyuzo/Logotherapyl .htrn.

Zuehlke, T. E. and Watkins, J. T. (1975). “The Use of Psychotherapy with Dying
Patients: An Exploratory Study,” Journal of Clinical Psychology 31(4): 729-732.

127

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

mpppppp