AUTONOMY

The concept of autonomy in moral philosophy and bioethics recognizes the human capacity for self-determination, and puts forward a principle that the autonomy of persons ought to be respected. At this level of generality, there is not much with which to take issue; a full account of autonomy must further define self-determination and state how and to what extent autonomy should be respected. Autonomy as a capacity of persons must be distinguished from autonomy as a property of actions and decisions, for a person with the capacity for autonomy may act nonautonomously on particular occasions, for example, a person who is coerced to do something. Autonomy as a fundamental value and a basic right is part of the moral and political theory of liberal individualism. According to this view, autonomous individuals are the ultimate source of value: The basis for an action, social practice, or government policy to be right or good is in the values, preferences, or choices of autonomous individuals. In social philosophy, individual autonomy as a basic value and a fundamental right is in tension with community values, such as caring for others, promoting the good of society, and preserving and enhancing the moral practices of society. In clinical bioethics, the right to autonomy of individual patients is in tension with healthcare professionals' obligations to benefit patients. These conflicts will be examined in what follows.

Autonomy as Capacity

There are three elements to the psychological capacity of autonomy: agency, independence, and rationality. Agency is awareness of oneself as having desires and intentions and of acting on them. (Desire includes inclinations, aversions, wants, and similar terms.) When people have a desire for some state of affairs, they form an intention to do what they believe will bring about the desired state of affairs; further, they want their desire to determine their action (Benn; Haworth).

The capacity for agency distinguishes persons from inanimate objects and from nonhuman animals. Inanimate objects can be affected by objects and conditions external to them, as can persons, but unlike persons, inanimate objects cannot be said to act on desires. Nonhuman animals have desires, but there is no (noncontroversial) reason to believe that they have the capacity for self-consciousness that is manifest in having an awareness of desires and wanting them to be effective in action. Agency does not imply that persons are never influenced by external forces or that persons never act impulsively. It is an account of how persons are able to act and not how they always act.

Independence is the absence of influences that so control what a person does that it cannot be said that he or she wants to do it. This may seem a feature of an autonomous action rather than an element of psychological capacity. However, there are cases in which a person's course of life is under constant threat of violence from others, and the person acts always to avoid harm: war, poverty, abusive relationships, police states. When the whole of a person's beliefs, plans, self-image, and ways of relating to others are the result of unrelenting coercion and manipulation, then that person has little or no capacity for autonomy.

Autonomy also requires that persons have an adequate range of options. Coercion and manipulation limit options, but options are also limited by social and physical environments. If a person's options are numerous and noncoerced but are trivial in relation to what is valued by the person, then there is no capacity for autonomy in a significant sense (Raz). This would be the case in a totalitarian, caste, or slave society where a combination of coercion and ideology suppress the aspirations and real options of a segment of the members of the society. A full account of the conception of autonomy must distinguish external influences that defeat autonomy from external influences that are consistent with being autonomous. The former includes coercion and manipulation, and the latter includes persuasion and the normal limitations of physical and social environments.

The third element of the capacity for autonomy is means-end rationality, or rational decision making. In addition to the self-consciousness of agency, the capacity for rational decision making requires a person: (1) whose beliefs are subject to standards of truth and evidence; (2) with ability to recognize commitments and to act on them;(3) who can construct and evaluate alternative decisions;(4) whose changes in beliefs and values can change decisions and actions; and (5) whose beliefs and values yield rankings of action commitments. Another way to understand rationality as an element of the capacity for autonomy is as the capacity for reflection on desires. A rational person can have a desire for or fear of something, such as a desire for food or a fear of surgery, and also have the wish that he or she not have that desire or not be moved by that fear (Dworkin, 1976, 1988; Childress). Persons who lack the psychological capacity for rational decision making are those who are severely mentally ill—paranoiacs, compulsive neurotics, schizophrenics, and psychopaths. Such persons have the capacity for agency, that is, they are aware of acting on their desires, but they fail to meet one or more of the above conditions. For example, a paranoid patient who persists in a delusion that the healthcare professionals are Martians attempting to capture him is unable to adjust beliefs and actions to a reality confirmed by evidence (Benn).

Principle of Respect for Autonomy

Principles that support autonomy can be directed at the everyday relationships and encounters between persons; at the constitution, laws, and regulations of a nation-state; and at the policies of institutions such as hospitals, insurance companies, schools, and corporations. What ought to be done to respect autonomy will not be the same at all these levels and will be a function of a broad social ideology.

The minimal content for a principle of respect for autonomy is that persons ought to have independence, that is, be free from coercion and other similar interferences. John Stuart Mill made this the main principle in On Liberty (1947): No one should interfere with the liberty of action of another except to prevent harm to others. This obligation not to coerce others is defensible as an obligation binding on individuals, private organizations, and governments. Mill defended his principle of liberty, not because he believed that there is a fundamental right to autonomy nor that autonomy is valuable in itself, but because the recognition of liberty is supported by the principle of utility. This principle is that an action or policy is right to the extent that it promotes the greater happiness for the greater number. However, securing negative liberty does not establish autonomy as fundamental in moral theory. Other philosophers have gone further than Mill in their defense of autonomy.

The most widely quoted principle of respect for autonomy is one of Immanuel Kant's versions of the categorical imperative: "Treat others and oneself, never merely as a means, but always at the same time as an end in himself" (p.101). This is frequently expressed as treating others as persons, and its distinctive Kantian claim is that others should be treated as rational beings who have their own ends. A further explanation of this principle is that persons should be seen as having interests in two senses. First, interests in those things that are a benefit to nearly everyone, for example, being free of pain, not being killed, being saved from dying. A physician can treat a patient without that person's consent and still protect these interests. Second, autonomous persons "take an interest" in things, that is, have preferences, projects, and plans. Acting only with concern to serve interests in the first sense, as is sometimes alleged against uses of the principle of utility, is not sufficient for respecting another's autonomy; we must also discover and take into account the individual's values and objectives (Benn). For example, a physician may believe that a surgical procedure is an effective treatment to relieve the pain of a patient's ulcer, but the patient may have a greater aversion to the risks of surgery than the physician does, and would prefer a restricted diet and medication. To not solicit, or to ignore, the patient's preferences in this matter would not respect his or her autonomy.

Autonomy, Rights, and Liberty

The concept of rights presupposes that right-holders are beings who have the capacity for autonomy, who make choices and can use discretion to exercise a right or not. Basic liberties in a liberal democracy are protected by constitutional and other legal rights. The idea of a right has three elements: the right-holder (the person who has the right); the object of the right (the activity or thing that the right-holder has a right to); and the duty-bearer (the person or institution who must do what the right requires). Negative rights are rights not to be interfered with; for example, everyone has the right not to be given medical treatment without consent, and all healthcare providers must respect this right. Positive rights are rights that a person be provided with something—for example, the right of all senior citizens in the United States to Medicare payment for healthcare, a right that is binding on government agencies and healthcare providers.

Recognizing the negative right to autonomy imposes on everyone the obligation not to coerce or otherwise interfere with the action of another. This protection of autonomy is not as costly to social institutions as recognizing positive rights to autonomy. If there is a positive right to X , this means that someone is under an obligation to provide X to the right-holder(s). For example, if every citizen has a fundamental positive right to the best-quality medical care, then the state must provide full access to medical care to all citizens. While there cannot be a positive right to autonomy per se—for autonomy as capacity is not something that can simply be given to persons who do not have it—there can be rights to other things that are required for, or supportive of, autonomy. Among them are rights to a decent minimum of healthcare, education, a decent standard of living, political participation, freedom of inquiry and expression, and equal opportunity to compete for positions in society. These goods contribute to autonomy in two ways: First, they make possible the development of the capacity for autonomy; second, they make autonomy meaningful by establishing the personal and social powers and range of options for autonomously chosen projects and plans. Discrimination against minorities and women decreases their autonomy by explicitly excluding them from desirable positions in society and by implicitly agreeing to the limited range of options offered to minorities and women.

Autonomy as an Ideal

There is no sharp line separating accounts of autonomy as an ideal from autonomy as an actual capacity of persons. Autonomy can be described as a high level of self-determination that few persons will actually achieve, and yet it can still be regarded as a capacity for all persons, if it is believed that all persons under suitable conditions could acquire it and use it to direct their lives. Views that describe autonomy at a level that nearly all normal adult persons can and do exercise are views of autonomy as capacity, and views that describe it at a higher level are accounts of autonomy as an ideal.

Autonomy as an ideal will center on a person's use of the capacity for deliberation and reflection. The person who realizes the ideal of autonomy is, first, one who is consciously aware of having the capacity, someone who believes that he or she can use it to shape his or her life. Second, the autonomous person will make particular decisions with a sense of control—creating and evaluating options. That person will also reflect on how values, preferences, attitudes, and beliefs received in the socialization process function in his or her own decision making, examine the kind of person this makes him or her, consider alternatives, and make a commitment to accept or try to alter who he or she is. This is of course a matter of degree; like every virtue, it can be realized well and thoroughly or in some small measure. The ideal of autonomy does not require individuals to make conscious, deliberated decisions before every action. A person who has accepted a set of preferences, beliefs, and attitudes can respond without much thinking to common situations that fall into recognized patterns.

Autonomy of Actions

In a clinical setting, it is often important to determine whether a patient's decision regarding treatment, or the decision of a proxy in the case of an incompetent patient, is autonomous. A person who has the capacity for autonomy may, for a variety of reasons, not act autonomously on a particular occasion. Determining whether a particular action or decision is autonomous is a matter of how the three elements of the capacity for autonomy (agency, independence, and rationality) are involved in the process of deciding. The autonomy of actions is a matter of degree because independence and rationality are matters of degree, though agency is not.

Ruth Faden and her colleagues describe the three elements of autonomy as intentionality, freedom from controlling influence, and understanding. They point out that controlling influences and understanding can be seen on two independent continua. An action can be performed within the range of full understanding to full ignorance, and within the range of completely uncontrolled to completely controlled.

Bruce Miller views the autonomy of actions and decisions on four levels: (1) as free action (agency and independence); (2) as authenticity (the decision is consistent with what is known about the person's values, preferences, and plans); (3) as effective deliberation (rationality); and (4) as moral reflection (deliberation about one's values, preferences, and plans). The decision of a patient may be autonomous at one or more, but not all levels. For example, a patient who accepts a recommended treatment without reflecting much about the decision, acted autonomously at the level of free action, and perhaps authenticity, but not at the levels of rationality and moral reflection.

The legal concept of competence is closely related to the concept of autonomy. A competent person is one who has the capacity for autonomy, and a competent decision is one that is autonomously made.

David Jackson and Stuart Youngner present six cases of decision making in an intensive-care unit that "illustrate specific situations in which superficial preoccupation with the issues of patient autonomy and death with dignity could have led to inappropriate clinical and ethical decisions …"(p. 407). In one of the cases, a patient with multiple sclerosis appeared to autonomously refuse further lifesaving treatment following a suicide attempt. However, psychiatric evaluation showed that the patient had become depressed and withdrawn at the time his wife and sons began spending time with his mother-in-law who had been diagnosed with inoperable cancer.

Jay Katz has said that insufficient attention has been given to the unconscious and irrational motivations of behavior. It is not only patients' motivations that should be examined, but physicians' as well, for example, their denial of uncertainty. Whether a patient's decision to consent to or refuse treatment is autonomous depends on more than the patient's statement of decision and reasons. Physicians and patients must engage in conversations; physicians are obligated to facilitate patients' opportunities for reflection to prevent ill-considered decisions, and patients are obligated to participate in the process of thinking about their choices. The U.S. President's Commission (1982) echoes this view in its discussion of the importance of communication between patient and health professional to attain shared decision making based on mutual trust.

Privacy, Informed Consent, and Paternalism

Autonomy as a fundamental right is used to justify rights to privacy, confidentiality, refusal of treatment, informed consent, and a decent minimum of healthcare. The legal right to privacy has two components. The right to control information about oneself is protected in medicine as the patient's right to confidentiality of information gained by health professionals. The right not to be interfered with and to make one's own decisions is protected in medicine as a competent patient's right to refuse recommended treatment and as the obligation of health professionals to obtain a patient's informed consent to treatment. Informed consent requires that a patient be informed of a recommended treatment and of the options for treatment and their likely consequences, and that the patient give express permission for a treatment (often in writing). The right to autonomy also requires that patients be told the truth about their medical status and prognosis, that their questions be answered, and that they receive assistance from healthcare providers in making rational decisions. Meaningful exercise of the right to autonomy in living requires that individuals possess physical and psychological capacities within the normal, human range. So the positive right to autonomy supports a right to a level of healthcare that will return and maintain a person to the normal range of functioning. This includes acute care, for example, repair of a broken bone; chronic care, for example, treatment of diabetes or heart disease; and supportive care for permanent disability, for example, wheelchairs for paraplegics.

Paternalism in healthcare is treating a patient against his or her wishes on the grounds that the healthcare provider is professionally obligated to provide care that will benefit patients, and that the healthcare provider knows better than the patient what is good for the patient. When paternalism is justified, it overrides patient autonomy, at least partially. An example of justified paternalism could be when a physician does not accede to a patient's refusal of emergency treatment because the patient believes he or she will surely die.

Criticisms of Autonomy

Some authors (Clements and Sider; Callahan; Thomasma) have criticized the centrality of autonomy in medical decision making. Their argument states that the primary obligation of healthcare providers is to maintain and restore health. There are two aspects to this claim. First, if patient autonomy is given primacy over the obligations of health professionals, physicians and other providers may violate their obligation to maintain and restore the health of patients; for example, a patient may refuse a treatment that will save his or her life or prevent a serious illness. These conflicts between autonomy and patient benefit have often been decided by courts, usually in the form of a request by a terminally ill patient's family member, or other agent, that life-preserving treatments such as respirators be withdrawn, a request denied by physicians who cite their obligation to preserve life.

A second aspect of the criticism of autonomy recognizes the centrality of patients' values and wishes in cases of deciding whether to forgo life-preserving treatment for a terminally ill patient, but other sorts of medical-care decisions depend less on respecting patients' rights to autonomy and more on the value of restoring and maintaining the capacity for living a meaningful life. In this sort of case, autonomy is secondary to principles of beneficence, compassion, and caring.

Defenders of autonomy can make several replies to this critique. (1) Some of the attacks on autonomy wrongly assume that it is simply a principle of negative freedom, that is, the right not to be interfered with. (2) The claim of the centrality of patient autonomy in medicine does not imply that it is the only value. The principles of beneficence or nonmaleficence may, in some circumstances, justify paternalism. (3) Autonomy cannot be ignored in medical decision making. Knowing what will be most beneficial for a patient often requires input from the patient on values, objectives, and preferences. This is true not only in morally difficult situations that call for a decision about preserving the life of a terminally ill patient, but in less dramatic cases as well, for example, whether a patient should have surgery for a condition that causes minor discomfort and dysfunction but will not develop into something more threatening to health, or whether the patient should simply "live with" the condition. In cases of acute and severe injury or illness where there is clearly a best treatment that will almost certainly restore the patient to health, it can usually be safely assumed that whatever else the patient values, he or she will value the restoration of health, and hence, discussion of the relative value of options and their consequences is not required to respect the autonomy of the patient.

Criticisms of autonomy have also been launched from a broader, communitarian perspective (MacIntyre; Sandel; Callahan). Communitarians charge that the political theory of liberal individualism states that individuals are fully self-determining and that rights to autonomy are the primary or sole standard for individual behavior, institutional practices, and government policy. Communitarians object to liberal individualism on several grounds. First, the socialization process determines, or shapes, the values and preferences of individuals, hence, the idea of autonomously chosen values is factually incorrect. Second, an individual's actions, desires, and objectives are comprehensible only within the context of social conventions and institutions. For example, a person cannot report that he or she is thinking about depositing a check without the conventions of language and the institution of banking. Third, the view that an autonomous individual chooses his or her own values, preferences, and desires presupposes a self that does the choosing. This self will have to have a core of values with which to choose, in which case either there are values not autonomously chosen, or it is inexplicable how individuals come to have a set of values. Communitarians also claim that liberal individualism regards persons as separate from others in the sense that individuals have no obligations to others or society that are not voluntarily assumed, other than the obligation to respect the individual rights of others. A society that respects only the autonomy rights of all its members is not morally complete. A good society must recognize obligations to help others; its members must have virtues such as compassion, caring, and love, and they should recognize a commitment to society to maintain social practices and institutions that establish and promote these obligations and virtues (Callahan).

There may be theories of autonomy that are susceptible to these criticisms, but the fundamental value of autonomy can be defended without embracing such versions of liberal individualism (Sher; Taylor, 1985). The conceptions of autonomy presented above recognize that persons are social beings whose values and preferences are shaped by society and that the capacity for autonomy is itself socially determined. Being autonomous requires language and reason, and these abilities are not possible without socially given practices and standards. Reflecting on socially given values and preferences and either accepting them as one's own or changing them in some measure, which is a feature of autonomous persons, cannot be done unless there is a social environment that encourages autonomy. A free society makes autonomy possible.

However, a society in which no one does more or less than respect everyone else's liberal rights, in which there is no caring, love, or friendship and no neighborhood associations, political parties, or civic groups, is not one we would want, though it may be a liberal society (Gutmann). On the other hand, a society organized to promote civic virtues and obligations such as beneficence, caring, and compassion, but which does not recognize a right of individuals to be different, to make their own decisions about matters of importance to them or to find a style of life that makes them happy, is also not one we would want. Love and care can be stifling if they do not recognize an individual's own view of what his or her good is. Finally, a defensible theory of the nature and value of individual autonomy will fall between radical individualism and extreme collectivism. It must explain the obligations to create and maintain social and political institutions that support the exercise and flourishing of autonomy. It must explain how the exercise of autonomy depends upon the opportunity range and values given in the traditions and structure of society. It will also recognize other fundamental values and explain their place in decision making.

In the early period of contemporary medical ethics, much attention was on medical paternalism in cases of life-and-death decision making for terminally ill patients and on what can be called "medical opportunism" in research on human subjects. Critics of these practices brought the rights of patients and subjects to the forefront of medical ethics. In a climate of concern for allocation of healthcare resources and other issues of social policy, autonomy appears less frequently in medical ethics literature than do moral concepts such as justice, fairness, equality, economic efficiency, and cost-containment. This shift in issues should not lead to the view that autonomy has lost its importance in moral and social theory and in bioethics.

bruce l. miller (1995)

bibliography revised

SEE ALSO: Beneficence; Coercion; Conscience; Ethics: Social and Political Theories; Freedom and Free Will; Human Dignity; Human Rights; Informed Consent; Justice; Professional-Patient Relationship; Research Policy: Subjects; Sexism

BIBLIOGRAPHY

Benn, Stanley I. 1988. A Theory of Freedom. Cambridge, Eng.: Cambridge University Press.

Berg, Jessica; Appelbaum, Paul S.; Lidz, Charles W.; and Parker, Lisa S., eds. 2001. Informed Consent: Legal Theory and Clinical Practice, 2nd edition. New York: Oxford University Press.

Callahan, Daniel. 1984. "Autonomy: A Moral Good, Not a Moral Obsession." Hastings Center Report 14(5): 40–42.

Childress, James F. 1990. "The Place of Autonomy in Bioethics." Hastings Center Report 20(1): 12–17.

Clements, Colleen D., and Sider, Roger C. 1983. "Medical Ethics' Assault on Medical Values." Journal of the American Medical Association 250(15): 2011–2015.

Davis J. K. 2000. "The Concept of Precedent Autonomy." Bioethics 16(2): 114–133.

Donchin, A. 2001. "Understanding Autonomy Relationally: Toward a Reconfiguration of Bioethical Principles." Journal of Medicine and Philosophy 26(4): 365–386.

Dworkin, Gerald. 1976. "Autonomy and Behavior Control." Hastings Center Report 6(1): 23–28.

Dworkin, Gerald. 1988. The Theory and Practice of Autonomy. Cambridge, Eng.: Cambridge University Press.

Ells, C. 2001. "Shifting the Autonomy Debate to Theory as Ideology." Journal of Medicine and Philosophy 26(4): 417–430.

Faden, Ruth R.; Beauchamp, Tom L.; and King, Nancy M. P. 1986. A History and Theory of Informed Consent. New York: Oxford University Press.

Gaylin W. 1996. "Worshiping Autonomy." Hastings Center Report 26(6): 43–45

Gutmann, Amy. 1985. "Communitarian Critics of Liberalism." Philosophy and Public Affairs 14(3): 308–322.

Haworth, Lawrence. 1986. Autonomy: An Essay in Philosophical Psychology and Ethics. New Haven, CT: Yale University Press.

Kant, Immanuel. 1956. Groundwork of the Metaphysics of Morals, tr. and ed. Herbert James Paton. New York: Harper & Row.

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

Jackson, David L., and Youngner, Stuart. 1979. "Patient Autonomy and 'Death with Dignity.'" New England Journal of Medicine 301(8): 404–408.

MacIntyre, Alasdair C. 1981. After Virtue: A Study in Moral Theory. Notre Dame, IN: University of Notre Dame Press.

Mill, John Stuart. 1947. On Liberty, ed. Aubrey Castell. New York: Appleton-Century-Crofts.

Miller, Bruce L. 1981. "Autonomy and the Refusal of Lifesaving Treatment." Hastings Center Report 11(4): 22–28.

Pantilat, S. Z. 1996. "Patient-Physician Communication: Respect for Culture, Religion, and Autonomy." Journal of the American Medical Association 275(2): 107–110.

Raz, Joseph. 1986. The Morality of Freedom. Oxford: Clarendon Press.

Sandel, Michael J. 1982. Liberalism and the Limits of Justice. Cambridge, Eng.: Cambridge University Press.

Schneider, Carl E. 1998. The Practice of Autonomy: Patients, Doctors, and Medical Decisions. New York: Oxford University Press.

Sher, George. 1989. "Three Grades of Social Involvement." Philosophy and Public Affairs 18(2): 133–157.

Taylor, Charles. 1985. "Atomism." In Philosophical Papers, vol. 2 of Philosophy and the Human Sciences. Cambridge, Eng.: Cambridge University Press.

Taylor, Charles. 1991. The Ethics of Authenticity. Cambridge, MA: Harvard University Press.

Thomasma, David C. 1984. "Freedom, Dependency, and the Care of the Very Old." Journal of the American Geriatrics Society 32(12): 906–914.

U.S. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. 1982. Making Health Care Decisions: A Report on the Ethical and Legal Implications of Informed Consent in the Patient-Practitioner Relationship. Washington, D.C.: Author.

Wear, Stephen; Engelhardt, H. Tristam; and Wildes, Kevin W., eds. Informed Consent: Patient Autonomy and Clinician Beneficence within Health Care, 2nd edition. (Clinical Medical Ethics Series). Washington, D.C.: Georgetown University Press.