The word autonomy derives from the two Greek words auto (“self”) and nomos (“law”). The term was originally applied to political entities in ancient Greece: An autonomous city-state, such as Athens, was one that laid down its own laws rather than having them imposed on it by some outside authority (e.g., by the Persians). Similarly, and as elaborated by the famous German philosopher Immanuel Kant, an autonomous person is someone who determines his or her own maxims or rules for action and acts in accordance with them. So an autonomous person, like an autonomous city-state or nation-state, is self-legislating.

There are two sets of distinctions to be kept in mind. The first distinction is between rationality and morality. An autonomous agent is a rational agent. However, arguably, being rational is not a sufficient condition for autonomy. For perhaps an autonomous agent is also a moral agent.

The second distinction pertains to sources of potential domination. An autonomous agent is one whose decisions are not externally imposed; he or she is not dominated by external forces or other persons. Autonomy thus stands in some tension with security, at least insofar as security concerns justify granting the police and other security agencies powers, for instance, to ban speakers who incite violence and to detain suspected terrorists who might, nevertheless, turn out to be innocent. However, an autonomous person is also possessed of self-mastery; he or she is not dominated by internal forces (e.g., addictions).

Rational Agency

Some have suggested that an autonomous person is both rational and moral. So what is it to be a rational person? Evidently a rational person is possessed of a continuing, rationally integrated structure of mental attitudes, such as intentions, beliefs, and desires. Moreover, the attitudes in question, notably beliefs, are evidence based. In short, the mental attitudes of a rational person are both rationally coherent and based on evidence.

Second, a rational person’s actions and dispositions to action are based on such coherent and evidence-based attitudes. So the actions are rational in light of the person’s mental attitudes (which are themselves rational).

Third, for a person’s attitudes and actions to be rational in this sense, the person must surely engage in both practical (action oriented) and theoretical (knowledge oriented) reasoning that makes use of objectively valid procedures, such as deriving valid conclusions from evidence and selecting means on the basis of their efficacy with respect to relevant ends.

Fourth, the concept of a rational person or being needs to be relativized to empirical circumstances, including inherent properties of the particular kind of rational beings in question. And it is possible that there are rational persons who are not human beings (e.g., Martians or creatures from some far-flung and as yet undiscovered planet). If so, then such nonhuman rational beings might not have all the inherent properties that human beings have. For example, human beings, but not necessarily other rational beings, have emotions, are highly social, and live for a finite number of years. Naturally, a rational being will act rationally in light of such additional inherent properties (as well as contingent external features of his or her environment).

Fifth, a rational beings are able to engage in rational scrutiny of their extant higher-order attitudes, such as beliefs about their own beliefs. If, for example, a rational person is engaged in self-deception (and, as a consequence, has false beliefs about his or her own motives), then, at least in principle, such a person can come to recognize and eliminate this self-deception.

Sixth, evidently, rationality in the sense in question admits of degrees; some people, for example, are better than others at drawing true conclusions from the evidence presented to them.

Moral Agency

Someone can be rational, up to a point, without necessarily being moral. Consider, for example, a highly intelligent psychopath. Such a person may well pursue his or her goals efficiently and effectively and make sophisticated, evidence-based judgments in doing so. So evidently psychopaths can be highly rational. However, psychopaths do not care about other people and are happy to do them great harm if it suits their own purposes. Moreover, psychopaths, even if they recognize the constraints of morality and pay lip service to them, do not feel the moral force of moral values and principles. In short, psychopaths can be rational and yet are not moral agents. So rationality and morality seem to be different, albeit related, concepts.

It is worth noting the distinction between nonrational and irrational agents, and between nonmoral and immoral agents. A nonrational agent cannot make judgments or inferences. An irrational agent has the capacity to make such judgments and inferences but has some significant deficit in his or her rationality, and thus makes a significant number of false judgments and/or invalid inferences, or often fails to act on the results of his or her practical reasoning. Similarly, a nonmoral agent lacks the capacity to make moral judgments and act on them; an immoral agent, by contrast, is merely (significantly) deficient in his or her moral judgment making or often fails to act on his or her correct moral judgments. That said, sometimes it is not clear whether we should think of a person as nonmoral (or nonrational) or as immoral (or irrational).

Autonomous Agency

Given that a fully human life involves responsiveness to moral reasons, an autonomous human being will thus be both rational and moral. Understood in the way outlined herein, rationality and morality imply independence and self-mastery. Someone who is dominated by the overriding desire to please an authority figure, and who acts only in accordance with that aim, will not count as autonomous. Similarly, the autonomous human being not only must be able to make good judgments about what to believe and how to act but must also be capable of acting in conformity with those judgments. Drug addicts, for example, may know perfectly well that it is unwise to keep feeding their addiction, but they may find themselves unable to act on that knowledge; the drug addict’s lack of self-mastery in respect of his or her desire for the drug means that the person lacks autonomy, at least in this area of his or her life.

To say that an autonomous human being is independent and possesses self-mastery does not, of course, imply that autonomy is incompatible with all forms of constraint. The autonomous person cannot infringe the laws of physics or the laws of logic. The fact that a human agent cannot hope to fly when jumping off a tall building or cannot both walk and not walk at the same time does not undermine his or her autonomy. Moreover, an autonomous person can choose to comply with the law without compromising his or her autonomy. Specifically, when human beings choose to comply with laws because these laws enshrine their moral beliefs and principles, they may well be acting autonomously, the laws in question being in effect self-imposed. Morally justified laws needed to protect individual and collective security are a case in point.

Autonomous human beings are ones who decide for themselves what is important and valuable to them, and they possess the capacity to make reason-based choices on the basis of recognizing, assessing, and responding to relevant considerations, including nonmoral facts and moral principles. When we call an act autonomous, we mean that it is something done by such a person on the basis of such a response.

Moreover, autonomy is in part constituted by various moral features, including freedom of thought and individual privacy. None of us, presumably, is completely autonomous, since we all fall short of full rationality, perfect morality, and absolute self-mastery, for example. Since these qualities vary from person to person, some people are more autonomous than others. Moreover, someone might be autonomous in one area of his or her life but not another. Nevertheless, we achieve the status of an autonomous human being—someone who is entitled to decide for oneself how one wishes to live—when we are sufficiently autonomous. Furthermore, autonomy can be undermined if one or more of its constitutive moral features are compromised. For instance, violations of privacy, such as ongoing intrusive surveillance, can undermine autonomy. Autonomous agents have a right and a need to control external access to their private lives.

Seumas Miller

See also Morality

Further Readings

Alexandra, Andrew and Seumas Miller. Ethics in Practice: Moral Theory and the Professions. Sydney, Australia: University of New South Wales Press, 2009.

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

Frankfurt, Harry. “Freedom of the Will and the Concept of a Person.” Journal of Philosophy, v.68/1 (1971).

Kant, Immanuel (trans. Mary J. Gregor). Groundwork of the Metaphysic of Morals. Cambridge, England: Cambridge University Press, 1998.

Kleinig, John, et al. Security and Privacy: Global Standards for Ethical Identity Management in Contemporary Liberal Democratic States. Canberra, Australia: ANU Press, 2011.

Mele, Alfred. Autonomous Agents: From Self-Control to Autonomy. New York, NY: Oxford University Press, 1995.

Miller, Seumas. “Individual Autonomy and Sociality.” In F. Schmitt (ed.), Socialising Metaphysics: Nature of Social Reality. Lanham: Rowman & Littlefield, 2003.

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