Paternalists maintain that restricting the autonomy of persons is justified if these persons would be likely to cause serious harm to themselves or fail to secure an important benefit for themselves. The main ethical issue is whether paternalistic interventions are morally justified, and if so, under what conditions. In bioethics, rightful authority in the patient–physician relationship and public-health interventions have been the focus of the discussion. For health policy, paternalism is central to questions concerning the government's role in promoting healthy lifestyles and preventing self-caused injury and illness.
Many actions, rules, and laws are commonly justified by appeal to some paternalistic principle. Examples include laws that protect drivers by requiring seat belts; restrictions on the availability of drugs; rules prohibiting a healthy subject of biomedical research from voluntarily undergoing a high-risk procedure; overriding adult refusals of treatment; disclosing confidential information about a patient to protect the patient's health; involuntary commitment to hospitals; interventions to prevent suicides; and denial of an innovative therapy to someone who wishes to receive it. Laws are the usual vehicle for translating paternalistic beliefs into public policy, but individual actions and institutional policies can also have paternalistic roots.
Early History in Ethical Theory
In an eighteenth-century discussion, the philosopher Immanuel Kant denounced paternalistic government ("imperium paternale") for its benevolent cancellations of the freedoms of its subjects (pp. 58–59). However, it was the nineteenth-century English philosopher John Stuart Mill who presented the first systematic attack on paternalism, a term he avoided, in his 1859 monograph On Liberty:
The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because in the opinions of others, to do so would be wise, or even right. These are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him …. In the part which merely concerns himself his independence is, of right, absolute. (p. 223)
Mill thus articulated a principle that properly restricted social control over individual liberty, regardless of whether such control is political, religious, or of some other type. He defended his principle with the utilitarian argument that granting people liberty rather than subjecting them to paternalism produces the best possible conditions for social progress and for the development of individual character and talent. Independent of his commitment to utilitarianism, Mill's On Liberty has played a more important role in discussion of paternalism than any treatise in ethical theory.
Neither Mill nor Kant anticipated that a paternal model of justified intervention into the affairs of competent adults might be extended to interventions with adult persons who, like children, have only a restricted or compromised capacity to choose autonomously. Yet this latter and broader model has become the most widely defended account of paternalism.
Definitions of Paternalism
The word paternalism refers loosely to acts of treating adults as a benevolent father treats his children, but the term has been given both a narrow and a broad meaning in ethical theory. In the narrow sense, paternalism refers to acts or practices that restrict the autonomy or liberty of individuals without their explicit consent; justification for such actions is either the prevention of some harm they stand to do to themselves, or the production of some benefit for them that they would not otherwise secure. This conception of paternalism leads to the following definition: Paternalism is the intentional limitation of the autonomy of one person by another, where the person who limits autonomy justifies the action exclusively by the goal of helping the person whose autonomy is limited (Dworkin, 1992; Beauchamp and McCullough). Following this definition, an act of paternalism overrides the value of respect for autonomy on some grounds of beneficence. Paternalism seizes decision-making authority by preventing persons from making or implementing their own decisions.
Many writers object to this analysis of paternalism because it does not comprehend the meaning of the term as it has descended from common usage and venerable legal precedent, where the notion is linked to guardianship, surrogate decision making, and government intervention to protect the vulnerable. The root sense of paternalism in ordinary language ("government as by a benevolent father") is joined with the law's wide-ranging use of terms such as parens patriae to produce a broad meaning that includes interventions into both autonomous and nonautonomous actions. Those who follow this broad vision recommend the following definition: Paternalism is the intentional overriding of one person's known preferences by another person, where the person who overrides justifies the action by the goal of benefiting the person whose will is overridden. Under this second definition, if a person's stated preferences do not derive from a substantially autonomous choice, overriding his or her preferences can still be paternalistic. The only essential condition of paternalism is beneficent treatment that overrides a known preference; a condition of substantial autonomy is not essential (VanDeVeer; Kleinig).
Defenders of the first definition argue that there are compelling reasons for resisting this second definition. First, paternalism originates in ethical theory as an issue about the valid limitation of freedom and autonomy. To include cases involving persons who lack substantial autonomy, such as drug addicts or the mentally disabled, broadens the term in a way that obscures the central issue, which is how, whether, and when liberty or autonomy can be justifiably limited. Second, the legal concept of parens patriae powers has its own subtleties and complexities. Courts do not apply this notion across the same range of thought and conduct that paternalistic literature treats as problematic. To incorporate a marginal legal doctrine together with the vagueness of ordinary language might prove more confusing than instructive in the end.
These two definitions are currently contested in literature. However, defenders of these two definitions need not disagree on all controversies about the meaning of paternalism. For example, it has sometimes been said that the term paternalism is inherently pejorative because it implies that authorities may treat adults such as hospital patients as if they were children lacking considered preferences of their own; therefore, they reason, the term is tainted by illegitimate authoritarianism or repressive dominance (Feinberg, 1980, 1986; Sherwin). Proponents of the above two definitions are free either to accept or to resist this interpretation. For example, they can both resist this pejorative meaning by arguing that paternalism suggests nothing beyond an analogy to respectable parental benevolence, in which parents act in the best interests of their children for good reason.
Weak (Soft) Paternalism and Strong (Hard) Paternalism
Joel Feinberg's distinction between weak and strong paternalism has profoundly affected literature on the subject. Although he switched to the language of "soft" and "hard" paternalism in his later work (1986), the terms "weak" and "strong" seem to have more deeply influenced the bioethics literature and will be used here.
In weak paternalism, one "has the right to prevent self-regarding harmful conduct only when it is substantially nonvoluntary or when temporary intervention is necessary to establish whether it is voluntary or not" (Feinberg, 1971,p. 113). This type of paternalism confines permissible limitations of autonomy to substantially nonautonomous (or nonvoluntary) behaviors. For example, it is permissible to pick up injured, partially incoherent victims of automobile accidents who refuse ambulance service and to admit against their will mentally ill persons who are dangerous to themselves. In strong paternalism, however, it is proper to protect or benefit a person by autonomy-limiting measures even if the person's contrary choices are autonomous. This paternalism supports interventions that protect competent adults against their will; that is, it controls or restricts substantially autonomous behaviors. For example, refusing to release a competent hospital patient who will die outside the hospital but requests the release knowing the consequences is an act of strong paternalism.
Weak paternalism is built on conditions of compromised ability or dysfunctional incompetence. When conduct that affects only the actor is restricted, some degree of autonomy may be present in the restricted actor, but the action must be substantially nonautonomous. Conditions that can significantly compromise the ability to act autonomously include the influence of psychotropic drugs, painful labor while delivering a child, and a blow to the head that affects memory and judgment. In medical situations, a patient's illness can be so devastating that it affects decisionmaking capacity. As the patient becomes weaker, less aware, or less alert, his or her dependence on the physician increases. A member of the medical profession who over-turns the preferences of a substantially nonautonomous patient in the interests of the person's medical welfare acts paternalistically and justifiably by the standards of weak paternalism. For this reason, weak paternalism has been widely accepted in law, medicine, and moral philosophy as an appropriate basis for intervention.
Strong paternalism, by contrast, supports some interventions intended to benefit a person whose choices and actions are informed and autonomous. Strong paternalism usurps autonomy by either restricting the information available to a person or overriding the person's informed and voluntary choices. These choices may not be fully autonomous or voluntary, but in order to qualify as strong paternalism the choices of the beneficiary of paternalistic intervention must be substantially autonomous or voluntary. For example, a strong paternalist would prevent a patient capable of autonomous choice from receiving diagnostic information that might lead to suicide. Unlike weak paternalism, strong paternalism does not require any conditions of compromised ability, dysfunctional incompetence, or encumbrance as the basis of intervention (although strong paternalists of course accept the justifiability of weak paternalistic interventions as well).
Justification of Paternalism and Antipaternalism
Defenders of paternalism in ethical theory have paid more attention to the justifying grounds for paternalism than to the type of paternalism justified. Some justifications range widely and defend both strong and weak paternalism. Typically, however, a condition in the argument states or hints that only weak paternalism is justified, although strong paternalism is the most controversial and interesting type of paternalism and may be the only type worth the effort of justification.
JUSTIFIED PATERNALISM. Defenders of paternalism often appeal to either a principle of rational consent or a principle of welfare or beneficence in order to justify their position. In one prominent justification, Gerald Dworkin argues that paternalism should be regarded as a form of "social insurance policy" that fully rational persons would take out for their protection (1972, p. 65). That is, paternalism is justified under conditions to which an impartial rational agent would consent if he or she were to appreciate the possibility of being tempted at times to make decisions to commit acts that are potentially dangerous and irreversible. The agent might at other times be driven to do something that would be considered too risky if he or she could objectively assess the situation—for example, smoking or drinking so heavily that health and life are endangered. A paternalistic health policy would remove or severely restrict the availability of tobacco and alcohol. In other cases, persons might not sufficiently understand or appreciate the dangers of their conduct, or might distort information about their circumstances. Seat-belt laws and motorcycle-helmet laws have often been enacted on this paternalistic basis.
Dworkin argues that a paternalistic act that denies a person an immediate liberty may paradoxically protect deep autonomy (i.e., the person's deeper values and preferences about the principles and standards on which he or she ought to act). A physician might lie to a patient, for example, in order to prevent a suicide, if the physician knows that the patient really wants to live and will later calm down and not commit suicide, although the patient is presently in no position to appreciate this fact. Dworkin argues that rational consent (consent that would be given) is the only acceptable way to express the conditions of justified paternalism. Many philosophers subsequently agreed with this thesis and made some form of consent a necessary condition of justified paternalism. However, justifications on bases other than consent have also been attempted (Dworkin, 1972; see VanDeVeer; and Kleinig).
A justification based on consent may do more to obscure than to clarify the issues. If the paternalist's objective is to protect or improve the welfare of another, then intervention can be justified by harm-avoidance or benefitproduction, as is the case in the justification of parental actions that override the wishes of their children. Children are treated paternalistically not because they will subsequently consent or would have consented were they rational, but because they will have better lives. This justification rests on providing for their welfare, not on respecting their autonomy.
THE JUSTIFICATION OF ANTIPATERNALISM. Some believe that paternalism is never justified. Mill supported this position, but with the important qualification that we are justified in restricting a person's liberty temporarily in order to ensure that the person is acting intentionally with adequate knowledge of the consequences of the action; once warned and informed, the person must be allowed to choose whatever course he or she desires. One need not be a follower of Mill's utilitarianism to defend this antipaternalism. For example, it can be defended by appeal to principles of respect for autonomy and privacy. Perhaps the most widely shared reason antipaternalists oppose (strong) paternalism is that it interferes with the authority of the individual, insults autonomous agents, and fails to treat them as moral equals (Childress).
The antipaternalist permits an initial, temporary infringement of liberty and privacy in the belief that persons who have a well-formed, autonomous resolution to do something harmful to themselves will have ample opportunity to perform the action after the temporary intervention has occurred. Intervention, in this limited respect, need not be a deep moral offense. Defenders of weak paternalism, however, view this qualified antipaternalism as insufficient because it disallows some highly desirable forms of intervention, such as long-term involuntary hospitalization for those in need of medical attention. Who, they reason, would not support altruistic beneficence directed at confused cardiac patients, ignorant consumers, frightened clients, and young persons who know little about the dangers of alcohol, smoking, drugs, and motorcycles? No caring and decent person would leave these individuals unprotected, and no reasonable philosopher would defend a normative thesis that permits such outcomes.
Weak paternalists thus project the appearance of steering a moderate and reasonable course between two radical and excessive extremes, strong paternalism and antipaternalism. The solution to the problem of paternalism, from their perspective, is to present the most defensible form of weak paternalism. But a severe stumbling block lies in the path of this tempting resolution of the issues: Weak paternalism has no clear substantive moral disagreement with antipaternalism, and therefore there is no reason to choose one over the other. Protection from harm caused to an individual by conditions beyond his or her knowledge and voluntary control—for example, by conditions beyond his or her self—is not an intervention that antipaternalists either criticize or disallow; they deny only the acceptability of intervention with substantially autonomous, self-caused harm. Weak paternalists too condemn such actions as an unjustifiable form of strong paternalism.
Weak paternalism, then, seems to be a defensible but noncontroversial position that virtually everyone accepts in some form. As Feinberg notes, it is "severely misleading to think of [weak paternalism] as any kind of paternalism," because weak paternalism is not "'paternalistic' at all, in any clear sense" (1986, pp. 12–14). Both weak paternalism and antipaternalism agree on the following critical claims:
- (a) It is justifiable to interfere in order to protect persons against harm from their own substantially nonautonomous decisions; and
- (b) it is unjustifiable to interfere in order to protect persons against harm from their own substantially autonomous decisions.
Weak paternalism is thus not a form of paternalism that can be distinguished in any morally important respect from antipaternalism. Weak paternalism does not seem to rest on a liberty- or autonomy-limiting principle independent of some moral principle of beneficence that supports prevention of harm to others (see Feinberg, 1971, pp. 107f., 124, and Feinberg, 1986, p. 13). Feinberg sarcastically suggests that the label "soft antipaternalism" seems to mean the same as "soft paternalism" (1986, p. 15).
The weak paternalist and the antipaternalist also join hands in opposition to the strong paternalist, who alone allows interventions that override and violate substantially autonomous actions.
THE JUSTIFICATION OF STRONG PATERNALISM. Although substantial autonomy is necessarily overridden in strong paternalism, conditions can be specified by a strong paternalist to restrict severely the range of justifiable interventions. For example, the strong paternalist might maintain that interventions are justified only if: no acceptable alternative to the paternalistic action exists; a person is at risk of serious harm; risks to the person that are introduced by the paternalistic action itself are not substantial; projected benefits to the person outweigh risks to the person; and any infringement of the principle of respect for autonomy is minimal.
Strong paternalism, so interpreted, will stand or fall on the strength of the argument that major welfare interests under some specifiable conditions legitimately override relatively minor autonomy interests. Many cases can be found that fit this model. For example, when healthy persons with no heart disease volunteered as subjects in a research study to have an artificial heart transplanted at the University of Utah, it was entirely reasonable that a review committee declared that the risk relative to benefit for a healthy subject is morally unacceptable and that they should not be allowed to undergo the procedure (Beauchamp and Childress).
Issues of Paternalism in Bioethics
Many examples of controversial paternalistic justifications are found in bioethics. Only a few general topics are treated here.
OVERRIDING REFUSALS OF TREATMENT. It is sometimes controversial whether procedures should be withheld or withdrawn even when the patient refuses the procedures. Justifications for overruling a patient's refusal of therapy need not be paternalistic, but they often are paternalistic because their objective is to prevent harm that would be caused by the patient's refusal. The issue is not whether a physician actually knows what is best for the patient, but whether the patient has a right to refuse treatment even if the refusal is harmful and the treatment beneficial.
Persons of questionable competence who refuse therapy present delicate moral problems and difficult conceptual issues about whether interventions are paternalistic. For example, do schizophrenic patients have a right to refuse a therapy for dehydration if a physician determines it to be safe and efficacious, and would an intervention after a refusal be paternalistic? Similarly, do children who understand what is being done to them have a right to refuse therapies when their parents and physicians judge these therapies to be essential, and are such interventions paternalistic?
OVERRIDING REQUESTS FOR TREATMENT. Patients or their legal representatives occasionally request medical procedures that physicians believe are harmful, ineffective, or futile. The physician may then refuse to act on these requests for paternalistic reasons. If the requests by patients are incompatible with accepted standards of care or conflict with the physician's conscientious beliefs about standards of care, a physician's refusal to comply may be justified for these apparently nonpaternalistic reasons of appropriate physician conduct. Nonetheless, the interventions are pater-nalistic whenever the primary ground of noncompliance with the request is that the treatment is not in the patient's best interests. Moreover, setting professional standards of practice is itself often a paternalistic attempt to protect patients' interests, and as such may be either justified or unjustified paternalism (Childress; Brett and McCullough). The same argument can be applied to drug policies of a government agency that refuses to accept requests for experimental therapies on grounds of risk to patients.
PARTIAL DISCLOSURES TO PREVENT HARM. Physicians and families often argue that a particularly devastating diagnosis or prognosis should not be disclosed to a patient. The concern is that bad news might adversely affect the patient's health or lead the patient to commit suicide. If the patient asks for the information or expects a truthful disclosure, it is paternalistic to withhold the truth. Physicians also occasionally make difficult medical decisions without consulting the parents of seriously ill newborns. These actions too are paternalistic if the objective is to prevent anguish to the parents. Other examples extend beyond serious patient illness. For example, genetic counselors sometimes use potential marital conflict for a patient as a reason not to disclose a condition such as nonpaternity, thereby depriving a patient of information generated in part by materials the patient provided.
In a much-quoted article on medical ethics, L. J. Henderson claimed that "the best physicians" use the following as their primary guide: "So far as possible, 'Do No Harm.' You can do harm by the process that is quaintly called telling the truth. You can do harm by lying…. But try to do as little harm as possible, not only in treatment with drugs, or with the knife, but also in treatment with words"(p. 823). The premise that some information may legitimately be withheld or disclosed only to the family for the patient's good is a clear instance of this rule and an equally clear case of paternalism. Why the family, rather than the competent patient, is given the information without the patient's prior permission is itself an important issue concerning paternalistic medical practices.
INVOKING THE THERAPEUTIC PRIVILEGE. Therapeutic privilege is a legally recognized privilege of the physician to withhold information from a patient if disclosure would cause serious deterioration in the physical, psychological, or emotional condition of the patient. This privilege has long been used in clinical settings to justify not obtaining consent and has elicited a particularly furious exchange over whether autonomy rights can be validly overridden for paternalistic reasons.
The courts have yet to develop a standard for appropriate use of the therapeutic privilege that renders it coherent with requirements of informed consent. If stated broadly, physicians can withhold information when disclosure would cause any countertherapeutic deterioration, however slight, in the physical, psychological, or emotional condition of the patient. If stated narrowly, physicians can withhold information if and only if the patient's knowledge of the information would have serious health-related consequences—for example, by jeopardizing the success of the treatment or harming the patient by critically impairing relevant decisionmaking processes. Confusion has also surrounded appropriate measures of rationality, psychological damage, and emotional stability under the standard of therapeutic privilege. Loose standards can permit physicians to climb to safety over a straw bridge of speculation about the psychological consequences of information, and this threat of abuse has made the therapeutic privilege highly controversial.
HEALTH POLICY FOR EXCESSIVE RISK. Antipaternalists argue that paternalistic standards for policy would authorize too much intervention. Paternalism could in principle prohibit smoking, drinking, and hazardous recreational activities such as hang gliding, mountain climbing, and white-water rafting, making such activities subject to criminal sanctions. Careful defenses of paternalism would disallow these extreme interventions, and at best antipaternalist arguments establish only a rebuttable presumption against paternalistic intervention. Nonetheless, antipaternalists are convinced that an unacceptable latitude of judgment would remain in contexts in which power is subject to abuse. Strong paternalism suggests that it would be permissible and perhaps obligatory to restrain and punish those who violate paternalistic rules. If so, antipaternalists argue, the state would be permitted to coerce morally heroic or valiant citizens if they act in a manner "harmful" to themselves. More generally, the state would be empowered to take away from persons the right to make decisions about their lives whenever officials view risks as excessive.
GOVERNMENT AGENCY RESTRICTIONS. Some government bureaus can be viewed, at least in part, as paternalistic guardians. For example, the Food and Drug Administration (FDA) in the United States is chartered to restrict persons from purchasing foods, drugs, and medical devices that are unsafe or inefficacious. A controversial decision by the FDA in 1992 to severely restrict the use of silicone-gel breast implants exemplifies paternalistic controversies that have beset the FDA. Women had elected implants for over thirty years, either to augment their breast size or to reconstruct their breasts following mastectomies. Over two million women in the United States had had these implants (three million worldwide) when, in April 1992, the FDA restricted the use of silicone-gel breast implants until additional studies could be conducted to establish their safety. Concerns centered on the implants' longevity, rate of rupture, and link with various diseases. Those who defended complete prohibition contended that no woman should be allowed to take a risk of unknown but potentially serious magnitude because her consent could not be informed. The FDA defended a restrictive policy, rather than prohibition, holding that patients with breast cancer and others have a legitimate need for breast reconstruction. The FDA distinguished sharply between reconstruction candidates and augmentation candidates, arguing that the favorable risk–benefit ratio is confined to reconstruction candidates (Kessler).
Critics of this decision charge that the government's decision is inappropriately paternalistic, especially in contrast to the more permissive public decisions reached in European countries. These critics argue that subjective benefits for many women outweigh the identified risks, and opinion surveys indicate that 90 percent of women receiving the implants are satisfied with the results (see Parker). Critics argue that the only defensible policy is to permit the continuing use of silicone-gel breast implants while requiring adequate disclosure of information about risks. Raising the level of disclosure standards is, from this perspective, more appropriate than raising the level of paternalistic restraints on choice.
THE MODEL OF PATERNAL AUTHORITY. The term paternalism has often been criticized as sexist and in need of correction to parentalism. However, some feminists in bioethics as well as some critics of paternalistic medical practices have argued that this usage is a rare case in which gendered language should be retained on grounds that an appropriate link is made between the privileges of a father in a patriarchical family and the privileges of physicians in an authoritarian medical system. The thesis is that just as hierarchical arrangements have long been the norm in the family, so paternalism has been the norm in medicine; to appreciate the need to revise authority structures in the family should similarly point to the need to revise the model of rightful authority in medicine (see Sherwin).
This criticism extends beyond analysis of the meaning of paternalism. It assumes the persistence among physicians, male and female, of the belief that a paternal model of authority is requisite in clinical practice because of compromised reasoning abilities in patients, the essential need for technical information in medical decision making, and the needs many patients have for an authority figure as healer. Susan Sherwin (1992) and other writers in bioethics have argued for replacing this traditional paternalistic model with a radically different model of the physician–patient relationship, such as a model based on friendship or on contract.
However, those who support justified paternalism in medicine believe that paternalism, properly understood, fits coherently with our normal expectations of altruistic beneficence and fiduciary responsibility in professional healthcare relationships. Their model is that of a dedicated professional who possesses superior knowledge, experience, and skills and who seeks to further a patient's best interest. Whether pieces of these two starkly different models can be joined consistently is a matter of widespread controversy in bioethics.
SUICIDE INTERVENTION. Many views about reporting, preventing, or intervening in suicide are paternalistic. Because of the extreme and irreversible effects of suicide, some defenders of intervention believe that a principle of respect for life creates an obligation to prevent suicide that overrides obligations based on the principle of respect for autonomy. A weaker account relies on Mill's strategy: Intervention is justified to establish autonomy in the person; but after it is determined that the person's decisions are substantially autonomous, further intervention would be unjustified. (Kleinig discusses several other paternalistic arguments for suicide intervention.)
Both this weaker account and stronger accounts have been defended on grounds that others do sometimes know our best interests with more insight and foresight than we do. It is often difficult to know how much ability persons have to act autonomously or how much insight they have into their "best interests." The stronger account is also defended on grounds that many suicidal persons are under intense strain or the influence of drugs or alcohol, clinically depressed, destabilized by a crisis, or simply wish to end their pain, and that these persons can be helped with their problems by health professionals. Another defense is that failure to intervene symbolically communicates to potential suicides an absence of communal concern and diminishes a feeling of communal responsibility. Finally, some argue that it is a justified form of paternalism for friends and healthcare professionals to infringe confidentiality by reporting suicide threats to those who may be in a position to help prevent the acts. Some even defend a paternalistic obligation to report suicide threats (Bloch).
INVOLUNTARY INSTITUTIONALIZATION. Finally, a vast literature surrounds the involuntary hospitalization of persons who have never harmed others or themselves but are thought to stand in danger of inflicting such harm or of being vulnerable to harm by others. A major part of the contemporary rationale for use of police powers for the emergency detention and civil commitment of those dangerous to themselves is a paternalism supported by the knowledge that treatment has often helped persons over a momentary crisis. These interventions can involve a double paternalism: a paternalistic justification for commitment and a paternalistic justification for forced therapy (e.g., psychotherapy) after commitment.
Bioethics in the 1970s and 1980s exhibited a strong tendency to reject paternalism as an unjustified tampering with autonomy. However, from the mid-1980s through the mid-1990s many voices began to be heard that were more sympathetic to various paternalistic appeals. Paternalism seems likely to continue to be a viewpoint that will gain or lose adherents as the issues and larger social context shift. We may never again see the concentrated flurry of scholarly interest in this subject that was exhibited from the mid-1970s to the mid-1980s, but paternalism is not likely to be an issue that will soon disappear.
tom l. beauchamp (1995)
Beauchamp, Tom L., and McCullough, Laurence B. 1984. Medical Ethics: The Moral Responsibilities of Physicians. Englewood Cliffs, NJ: Prentice-Hall.
Bloch, Kate E. 1987. "The Role of Law in Suicide Prevention: Beyond Civil Commitment—a Bystander Duty to Report Suicide Threats." Stanford Law Review 39(4): 929–953.
Brett, Allan S., and McCullough, Laurence B. 1986. "When Patients Request Specific Interventions: Defining the Limits of the Physician's Obligation." New England Journal of Medicine 315(21): 1347–1351.
Buchanan, Allen. 1978. "Medical Paternalism." Philosophy and Public Affairs 7(4): 371–390.
Dworkin, Gerald. 1972. "Paternalism." Monist 56(1): 64–84.
Dworkin, Gerald. 1992. "Paternalism." In Encyclopedia of Ethics, pp. 939–942, ed. Lawrence C. Becker. New York: Garland.
Feinberg, Joel. 1971. "Legal Paternalism." Canadian Journal of Philosophy 1: 105–124, rev. in Social Philosophy. Englewood Cliffs, NJ: Prentice-Hall 1973. Recast in Harm to Self (below).
Feinberg, Joel. 1980. "The Child's Right to an Open Future." In Whose Child? Children's Rights, Parental Authority, and State Power, pp. 124–153. Edited by William Aiken and Hugh LaFollette. Totowa, NJ: Rowman and Littlefield.
Feinberg, Joel. 1986. Harm to Self. Vol. III, The Moral Limits of the Criminal Law. New York: Oxford University Press.
Kant, Immanuel. 1974. (1793). On the Old Saw: That May Be Right in Theory But It Won't Work in Practice, tr. E. B. Ashton. Philadelphia: University of Pennsylvania Press.
Kessler, David A. 1992. "Special Report: The Basis of the FDA's Decision on Breast Implants." New England Journal of Medicine 326(25): 1713–1715.
Kleinig, John. 1983. Paternalism. Totowa, NJ: Rowman & Allanheld.
Mill, John Stuart. 1977. (1859). On Liberty. In Collected Works of John Stuart Mill, vol. 18. Toronto: University of Toronto Press.
Parker, Lisa S. 1993. "Social Justice, Federal Paternalism, and Feminism: Breast Implants in the Cultural Context of Female Beauty." Kennedy Institute of Ethics Journal 3(1): 57–76.
Pellegrino, Edmund D., and Thomasma, David C. 1988. For the Patient's Good: The Restoration of Beneficence in Health Care. New York: Oxford University Press.
Purdy, Laura M. 1992. In Their Best Interest: The Case against Equal Rights for Children. Ithaca, NY: Cornell University Press.
Sartorius, Rolf, ed. 1983. Paternalism. Minneapolis: University of Minnesota Press.
Sherwin, Susan. 1992. No Longer Patient: Feminist Ethics and Health Care. Philadelphia: Temple University Press.
VanDeVeer, Donald. 1986. Paternalistic Intervention: The Moral Bounds on Benevolence. Princeton, NJ: Princeton University Press.
The derivation of the term “paternalism” from a Latin-English kinship term suggests its root meaning: a type of behavior by a superior toward an inferior resembling that of a male parent to his child—in most cases, a son. The precise forms of this behavior vary from society to society because the culture of kinship varies, and also because the nature of the tasks performed in paternalistic relationships vary. There are two general functions of paternalism: (1) the transmission of goods and services across gaps between generations and status levels and (2) the provision of means of access to resources to persons ordinarily excluded from such means. While economic phenomena probably are more commonly associated with paternalism than are any others, almost any kind of social content can appear: political, intellectual, religious, or marital.
Within different types of paternalistic systems, the following three basic ideas and two modes of motivated action can be found. First, since a “child” is defenseless and lacks property, he requires assistance and support. Second, since a “child” is not fully aware of his role and therefore not fully responsible, he requires guidance. These first two ideas underlie the benevolent mode of paternalistic action, where the superior person’s actions are dominantly supportive of the inferior. The third idea holds that since a “child” is ignorant, he can be deceived, or treated in such a way as to serve the interests of the “adult,” without becoming aware of this. This third idea obviously leads toward exploitative paternalism.
The theoretical relevance of paternalism was first noted by Max Weber, who developed the concept of patrimonialism ([1906-1924] 1946, p. 297). Patrimonial relations were defined by Weber as consisting of those existing between a boss, employer, feudal lord, or other similar figure and his band of henchmen, who give loyalty and obedience in return for protection. Weber pointed to the great historical depth of this type of relationship. It is well to remember that paternalistic systems contribute clear-cut solutions to the problem of complementarity of roles within a hierarchical structure: status differences within a hierarchy can be organized by complementary services and functions so that the necessary tasks are performed smoothly and efficiently.
In the present article we shall be concerned, first, with economic paternalism in plantation and industrial contexts. Second, we shall examine certain patron-client relations in local communities. In many cases both economic paternalism and patron-client systems are modeled on roles and terminology of folk kinship systems. The selection of cases in these two contexts is by no means exhaustive, since paternalism can become a model for relationships in any hierarchical setting. However, the cases selected for description represent the majority of types.
Plantation systems in both the New World and the Old World have received considerable attention from social scientists interested in paternalism. These observers agree that plantation paternalism was in large part a projection of social patterns found elsewhere in the national society. For example, Siegel (1955, p. 405) noted that everywhere in rural Brazil structural relationships similar to specific plantation forms are to be found: “There is the set of social and economic patterns which determine the relationship of the vaqueiro,the northeastern cowboy, to the ranch owner; the patterns linking the guarimpero to the merchant and diamond buyer, and the camarada system of the plateau region of southern Brazil” (a camarada is an agricultural worker who enters into paternalistic agreements with a farmer). The fact that plantation paternalism is a symptom of more widespread social patterns highlights the general problem of change which these systems present to the social scientist. Paternalism in any setting must be considered from at least two points of view: (1) the specific causes which call it forth in a particular setting and (2) the extent to which its presence is related to traditional forms in the society at large.
One cause of paternalism in plantation systems is the character of the labor force. Greaves (1959) and Stein (1955) see paternalism as the result of the existence of a poor and uneducated aboriginal labor force, which led plantation owners to supply many personal services in order to induce laborers to remain or, in the exploitative mode, in order to hold them as virtual slaves (see Alexander 1962, on northeast Brazil, where a slave-owning economy influenced the later plantation system). The case of Argentina is instructive in that paternalism in both the plantation system and industry has been much less marked than in other Latin American countries, because the labor force was drawn from better-educated European immigrants lacking traditions of dependence (Alexander 1962).
Also relevant to the discussion of causes is the economic format of the plantation: rural and relatively isolated, thus requiring that both managers and workers live at the site of work. In this situation daily relations between management and labor take intimate forms, and the possibility of contractual, functionally specific relations is less than in the case of an urban factory. This situation underlines a feature found in many paternalisms: the superior or boss is a kind of “gatekeeper”; access to rewards and necessities can be obtained only through him, and hence he can demand more than just adequate performance. The fact that the labor force is often derived from peasant or tribal communities plays a role insofar as these people are accustomed to authoritarian and dominancesubmission patterns based on kinship. However, Wolf (1959) saw the personalized relationships of the plantation as a “ritual pantomime of dependence”: not a genuine personalization, but forms used by the manager to soften the rigor of an inherently exploitative labor system.
In agricultural tenancy, paternalistic relations are generally less well developed than on the plantation. However, in parts of Japan (Beardsley et al. 1959) and in Lebanon (Sayigh 1958) the generally low educational level of the tenants, plus low levels of productivity, encouraged certain forms of paternalism.
Change in labor paternalism
While there is no detailed survey specifically aimed at studying change, the cumulative evidence suggests that paternalism on plantations has been in a process of decline, especially since World War n. Adams (1959) points out that in Guatemala the rise of labor unions since 1944, coupled with the national government’s labor security and land reform measures, had the effect of dissolving the old patronmozo relationship between plantation owner and worker. Specifically, a hostile relationship developed between workers and employer, promoting a willingness on the part of the former to look to labor unions and the government for the supports formerly furnished, at the price of low wages and semibondage, by the plantation owner. This case illustrates a phenomenon noted also in the case of Japan (Bennett & Ishino 1963, chapter 5): when external agencies began to supply unemployment relief and medical care to boss-organized dock workers, the workers deserted the boss systems and signed up with the public employment offices. In Hawaii the deeply rooted paternalism of the sugar and pineapple plantations has been giving way under the impact of unionization and the Americanization of Japanese and Filipino laborers (Norbeck 1959, p. 148). However, it should not be assumed from these cases that all paternalistic systems automatically vanish when other forms of labor security are introduced. In the cases noted, there was evidence that these particular organizations had strong exploitative tendencies and that workers rejected them because the price paid for security was no longer bearable. The process was also facilitated by changes in the national cultural climate toward more rational, universalistic values.
In other cases paternalistic plantation systems have survived or even have been created under modern conditions because the workers needed these kinds of relationships and because the price of protection was not excessive. Ford (1955), in a report on Peru, notes that sugar plantation owners and managers would willingly give up paternalistic practices, as demanded by labor leaders, but cannot, because the workers expect this kind of treatment. Somewhat similarly, in Indonesia the Dutch introduced a paternalistic system on the sugar plantations not only because the labor force consisted mainly of small farmers removed from the land and, hence, devoid of any important support other than wage work, but also because the workers expected this type of treatment, having received it from wealthy Muslim employers under the basic obligation of charity in their religion (Hawkins 1962; Geertz 1965). However, Hawkins (1962, p. 84) also notes that in spite of these expectations, workers on the plantations had “a strong feeling that work was something the Dutch imposed upon them.” Thus, one might reasonably expect that as labor organization and government intervention continue to develop in the formerly colonial countries, paternalistic relations in the plantations will give way to a more rationalized employer-employee relationship. The idea of individual freedom of contract, coupled with nationalistic feelings, may be sufficiently strong to dissolve the traditional sentiments. Moreover, there is no denying that labor paternalism, however benevolent, has an element of bondage; the worker simply is not free to move from job to job. Whether or not his personal welfare is better served under a freer, but less paternally secure, system is another issue.
At least three points of view have emerged on the question of change in labor paternalism (see Odaka 1964). The first, or “Marxist,” position sees paternalism as a traditional residue which will pass when the final vestiges of “feudal” society have been removed. The second views paternalism as a necessary, and often humane, system of relationships which organizes modern economic effort in ways acceptable to the local cultural traditions; hence, it need not disappear, since modern economic effort can be organized in ways other than those common in the Western countries (e.g., C. S. Belshaw 1955; I960; Abegglen 1958). The third sees paternalism as a response to particular socioeconomic conditions associated with scarce capital, needs for concentration of capital in a few hands, and general low levels of living (e.g., Bennett & Ishino 1963). Thus, in this third view, paternalism may appear or recur in any country during any historical period when these conditions are present.
The factory usually differs from the plantation insofar as it is located in a town and does not require the isolated, in-service residence of both management and labor. This feature may be less important in cases where the town is particularly small and/or the factory is the sole source of wage labor. Moreover, the factory, except in a few cases where the organization is a lineal descendant of “preindustriar systems (as in the case of provincial Japanese industry), was usually established on the concept of the labor contract; hence, the feature of bondage or indenture, so prominent in the case of the plantation, was generally of lesser significance. Finally, labor organization was introduced earlier and was more influential in the factory system, in any given case, than on plantations, because it stems from an urban-industrial tradition.
However, as in the case of plantations, paternalistic relations in factory systems have appeared wherever the national social structure contains such patterns to an appreciable degree. In Iran, Vreeland (1957, p. 166) notes, “the employeremployee relationship . . . derives from the ancient patterns of work organization . . . the traditional authority pattern found in the Iranian family and government. The worker . . . still commonly looks to his employer as a son looks to his father, for direction, for protection, for control/’ In Brazil the cotton-textile factories built housing for their workers, established company stores, built churches on factory grounds, and financed the saints’ festivals (Stein 1955). In Japan the nineteenth-century spinning mills did similar things and even more—the plant owner became a foster father to the young girls indentured to him and even arranged marriages. In modern Japanese factories such extreme practices have withered, but a similar sense of responsibility often pervades the relationships between labor and management (Abegglen 1958). However, there is considerable evidence that the rise of vigorous labor unions in postwar Japan is contributing to a decline of this type of hierarchical mutualism.
Industrial paternalism was common in the United States in the nineteenth and early twentieth centuries, and it continues in many parts of Europe today. The Pullman Company case is perhaps the classic example of American paternalism: the company created a total Utopian world for its workers, including housing, but forbade unions and any sort of bargaining. In spite of the many fringe benefits, the Pullman workers mounted one of the most massive strikes in the history of the American labor movement. The Pullman and similar cases demonstrate the inherent incompatibility of paternalism with a national social system which stresses individualism and free contract. Again the evidence suggests that paternalistic systems will wither once these values become part of the national culture. Paternalism endures if its participants believe in it and if they are getting something out of it at a bearable price.
A case with transitional features is that of Puerto Rico (Gregory 1960). Here the modern framework of unions and rationalized industry coexists with traditional paternalistic practices. Interestingly, the most effectively unionized sectors are the older ones, while the many new factories often tend to have the most paternalistic systems. This is explained by the fact that the new industries draw their labor from a rural population that is familiar with Hispanic patronage practices. These immigrants, unaccustomed to wage labor and lost in the teeming cities, look to the employer for comfort and protection. The existence of a benevolent, paternalistic governor also encouraged these sentiments, and government paternalism in Puerto Rico has often been accused by intellectuals of fostering traditionalism and delaying individual responsibility. However, these are transitional aspects of a society undergoing rapid development; they perhaps bear out the point that a little paternalism may be a good thing in warding off the shock of rapid change—although it creates patterns of dependent behavior which must be changed in later stages of development.
In Brazil industrial paternalism was developed after World War n, as a measure designed to help the workers adapt to a new industrial economy and to prevent them from turning to the unions and the Communist party (Alexander 1962, p. 178). The industrial federations pooled their resources to establish two organizations which provided medical care, schools, libraries, low-cost grocery stores, insurance, and many other services for workers in various industries. In Argentina, as noted, paternalism was generally of less importance because of the relative sophistication of the labor force. However, after 1943 the expansion of industry brought large numbers of rural people and relatively uneducated urban proletarians into the factories; hence, the Per on government found an opportunity to develop paternalistic assistance. The consequence was the descamisados movement, the “shirtless ones” who formed the bastion of Peron’s mass support. Thus paternalism, on a national industrial scale, can have major political significance (as in the Pullman case).
In Turkey the state enterprises have provided considerable aid to workers in the areas of housing, medical care, food, and clothing. The 1936 labor code required private employers to provide similar, although not so elaborate, benefits. Critics claim that “a benevolent, though somewhat stifling paternalism dominates the personnel policies of the state enterprises’’ (Rosen 1962, p. 264). These state and legalized private paternalistic practices, based on folk paternalism rooted in the Turkish social structure, provide an example somewhat comparable to that of Puerto Rico. Folk paternalism has not become translated into institutionalized forms in so efficient a manner in all cases.
The general features of both plantation and industrial paternalism may be summarized. In both, the managerial element assumes responsibilities for workers over and beyond the basic contractual provisions for wages and routine working conditions. These responsibilities include various benefits, depending upon local customs: financial protections and guarantees concerning the payment of medical expenses for the worker and his family; basic services and supplies, such as clothing and food; the practice of carrying workers on the payroll in periods of business decline; housing; religious facilities; wage differentials, depending upon size of family or length of service; and many others. Motives for extending such services are various and can take a benevolent form as well as calculating and exploitative modes. However, workers in all cases are expected to remain with the employer and to exhibit loyalty in return for these “fatherly” manifestations. Political support for the owner-manager class, in the event this class enters politics, is anticipated and used.
We turn now to a brief discussion of paternalistic systems of a different sort: those which involve not an entire labor force or group but a relationship between two individuals, the “patron” and the “client” (sometimes more than one client). These patron-client systems are extremely widespread, since they usually do not require institutionalized recognition. They appear in such culturally disparate situations as among ranchers and cowboys in the North American West (writer’s own research, unpublished); in the Japanese lumbering and forestry industries (Bennett & Ishino 1963); in Greek villages (Campbell 1963); and between castes in India (Kolenda 1963). Many of these systems take on quasi-kinship attributes.
Kenny’s description of the Spanish village patron may serve as a general model of the Mediterranean and Latin patron-client systems: “I define patron as .. . someone who is regarded (and who regards himself) at once as a protector, a guide, a model to copy, and an intermediary to deal with someone else or something else more powerful than oneself, whether or not such power is imaginary or real in a single context or in all, and whether or not the advantages to be gained from his patronage are material or intangible” (1960, p. 15). In Mediterranean and Latin countries (both Old World and New World), the individual will seek out such patrons in order to receive certain benefits and protection; above all, the patron serves as an intermediary who can deal with the official and professional world. The specific methods of acquiring such benefactors differ from country to country, but their basic functions appear remarkably constant. It is important to note that these functions are entirely desirable from the point of view of the client; that is, on the whole, patronclient systems, as a variant of paternalism, lack exploitative or clear-cut dominance-submission features. In return for the favors granted, the client is of course expected to perform certain services, but generally these appear to be well within the limits of a permissible price.
These functions are underlined in the reports on Greek patron-client systems by Friedl (1962) and Campbell (1963), who both point out that these systems actually result in a more equitable distribution of facilities and privileges which otherwise would tend to be concentrated in educated and wealthy elites. The systems, moreover, serve a vital function in keeping the relatively anonymous and isolated village population in touch with the institutions of the national society.
This Mediterranean—Latin style of patron—client system is found in other parts of the world, notably certain parts of the Far East (see Dewey 1962, on Java). However, in these Oriental cases, including Japan, there is a much stronger element of feudalderived noblesse oblige; that is, the patron and client have rather clearly understood, and often ritualized, obligations vis-a-vis each other. The local prestige of either can be injured if these obligations are voided or ignored. In the Japanese case, Bennett and Ishino (1963) reported on patron-client systems in the Japanese forestry industry, where a timber dealer would become a patron to a group of lumbermen (nakama), who, in return for his guarantee of steady employment and support even in economically depressed periods, would agree to cut trees for him only and to work around his house and grounds. The dealer’s prestige in the community was dependent upon his continued observance of the agreement, and the lumbermen would suffer a loss of prestige and work if they left his employ without notification. Bennett and Ishino contrasted this type of patron-client system with another paternalistic arrangement in the same community, where a big lumber entrepreneur had organized elements of the industry from dealers and lumbermen to factories, running the whole enterprise in a domineering manner. In Japan the latter type of arrangement, verging on the exploitative mode of the employer-employee systems described earlier, is often referred to as anoyabun-kobun system (oya: parent; ko: child;bun: status).
A patron-client system of considerable antiquity and pervasiveness is found in India, where it may be called the jajman-kamin relationship. Thejajman is the patron figure, a person of a higher caste than the kamin, or client. The latter performs numerous services for the jajman, who pays thekamin in kind, with products of the agricultural economy, clothing, or other articles. Kamins may also perform informal services for their jajmans,such as communication to lower castes. In the very large number of analyses and critiques of analyses of this system, there are few which attempt to handle it as a variant of patron-client systems (see Kolenda 1963; Pocock 1962 for summaries of the literature). Some emphasize the economic distribution feature, while others emphasize role theory, power relations among castes, “feudal” relationships, and the like.
Reviewing the evidence, it appears that paternalistic systems have two main currents. One, stemming from purely instrumental institutions and goals with hierarchical structures, moves between exploitative and benevolent modes, with the former possibly somewhat more prominent. The other stems from the genuine needs of inferiors and superiors in hierarchical social systems for mutual support; in this category we have the patron-client systems described, and also a large variety of other types, including professor-graduate student, master-apprentice, and priest-neophyte. In all cases, persisting feudal or caste traditions can have a powerful influence on paternalistic arrangements.
The question of trends toward disappearance or persistence of paternalistic relations is difficult to answer. In some contexts economic modernization and social democratization definitely encourage their decline; in other cases rapid change creates conditions exposing a proletariat to insecurity, which can create new forms of paternalistic dependency. The survival or creation of a paternalistic system depends on needs and on the existing social organizational patterns and traditions. The memories of suitable forms of paternalism, and the requisite familial models, may survive indefinitely, even in societies with highly rationalized economic and political institutions, and thus be available for use if the need arises.
John W. Bennett
Abegglen, James C. 1958 The Japanese Factory: Aspects of Its Social Organization. Glencoe, 111.: Free Press.
Adams, Richard N. 1959 On the Relations Between Plantation and “Creole Cultures.” Pages 73-79 in Seminar on Plantation Systems of the New World, San Juan, P.R., 1957, Plantation Systems of the New World: Papers and Discussion Summaries of the Seminar Held in San Juan, Puerto Rico. Washington: Pan American Union.
Alexander, Robert J. 1962 Brazil, Argentina, and Chile. Pages 138-186 in Walter Galenson (editor),Labor in Developing Economies. Berkeley: Univ. of California Press.
Allen, George C. (1946) 1963 A Short Economic History of Modern Japan, 1867-1937; With a Supplementary Chapter on Economic Recovery and Expansion, 1945-1960. Rev. ed. New York: Praeger.
Beardsley, Richard K.; Hall, John W.; and Ward, Robert E. 1959 Village Japan. Univ. of Chicago Press.
Belshaw, Cyril S. 1955 The Cultural Milieu of the Entreprenuer: A Critical Essay. Explorations in Entrepreneurial History 7:146-163.
Belshaw, Cyril S. 1960 Adaptation of Personnel Policies in Social Context. Pages 92-108 in Wilbert E. Moore and Arnold S. Feldman (editors), Labor Commitment and Social Change in Developing Areas. New York: Social Science Research Council.
Belshaw, Horace 1947 Industry and Agrarian Reform. Far Eastern Survey 16:153-156.
Bennett, John W.; and Ishino, Iwao 1963 Paternalism in the Japanese Economy: Anthropological Studies of Oyabun-Kobun Patterns. Minneapolis: Univ. of Minnesota Press.
Campbell, J. K. 1963 The Kindred in a Greek Mountain Community. Pages 73-96 in Julian Pitt-Rivers (editor), Mediterranean Countrymen: Essays in the Social Anthropology of the Mediterranean. Paris: Mouton.
Dewey, Alice G. 1962 Peasant Marketing in Java. New York: Free Press.
Ford, Thomas R. 1955 Man and Land in Peru. Gainesville: Univ. of Florida Press.
Freyre, Gilberto (1933) 1956 The Masters and the Slaves (Casa-grande & senzala): A Study in the Development of Brazilian Civilization. 2d English-language ed., rev. Translated from the 4th definitive Brazilian edition. New York: Knopf. → First published in Portuguese. An abridged edition was published in 1964.
Friedl, Ernestine (1962) 1965 Vasilika: A Village in Modern Greece. New York: Holt.
Galenson, Walter 1962 Introduction. Pages 1-10 in Walter Galenson (editor), Labor in Developing Economies. Berkeley: Univ. of California Press.
Geertz, Clifford 1965 The Social History of an Indonesian Town. Cambridge, Mass.: M.I.T. Press.
Greaves, Ida C 1959 Plantations in World Economy. Pages 13-23 in Seminar on Plantation Systems of the New World, San Juan, P.R., 1957, Plantation Systems of the New World: Papers and Discussion Summaries of the Seminar Held in San Juan, Puerto Rico. Washington: Pan American Union.
Gregory, Peter 1960 The Labor Market in Puerto Rico. Pages 136-172 in Wilbert E. Moore and Arnold S. Feldman (editors), Labor Commitment and Social Change in Developing Areas. New York: Social Science Research Council.
Hawkins, Everett D. 1962 Indonesia. Pages 71-137 in Walter Galenson (editor), Labor in Developing Economies. Berkeley: Univ. of California Press.
Kenny, Michael 1960 Patterns of Patronage in Spain. Anthropological Quarterly 33:14-23.
Kolenda, Pauline M. 1963 Toward a Model of the Hindu Jajmani System. Human Organization 22:11-31.
Levine, Solomon B. 1958 Industrial Relations in Postwar Japan. Urbana: Univ. of Illinois Press.
Moore, Wilbert E.; and Feldman, Arnold S. (editors) 1960 Labor Commitment and Social Change in Developing Areas. New York: Social Science Research Council.
Norbeck, Edward 1959 Pineapple Town: Hawaii. Berkeley: Univ. of California Press.
Odaka, K. 1964 Traditionalism and Democracy in Japanese Industry. Volume 3, pages 39-49 in World Congress of Sociology, Fifth, Transactions. London: International Sociological Association.
Pocock, David F. 1962 Notes on Jajmani Relationships. Contributions to Indian Sociology 6:78-95.
Rosen, S. H. 1962 Turkey. Pages 251-295 in Walter Galenson (editor), Labor in Developing Economies,Berkeley: Univ. of California Press.
Sayigh, Yusif A. 1957 Lebanon: Special Economic Problems Arising From a Special Structure. Middle East Economic Papers :60-88.
Sayigh, Yusif A. 1958 Management-Labor Relations in Selected Arab Countries: Major Aspects and Determinants. International Labor Review 77:519-537.
Siegel, Bernard J. 1955 Social Structure and Economic Change in Brazil. Pages 388-411 in Simon Kuznets, Wilbert E. Moore, and Joseph J. Spengler (editors),Economic Growth: Brazil, India, Japan. Durham, N.C.: Duke Univ. Press.
Stein, Stanley J. 1955 The Brazilian Cotton Textile Industry: 1850-1950. Pages 430-447 in Simon Kuznets, Wilbert E. Moore, and Joseph J. Spengler (editors), Economic Growth: Brazil, India, Japan. Durham, N.C.: Duke Univ. Press.
Sufrin, Sidney C; and Wagner, Frank E. 1961 A Brief Annotated Bibliography of Labor in Emerging Societies. Syracuse, N.Y.: Syracuse Univ., Maxwell School of Citizenship and Public Affairs, Center for Overseas Operations.
Van Den Berghe, Pierre L. 1958 The Dynamics of Racial Prejudice: An Ideal-type Dichotomy. Social Forces 37:138-141.
Van Den Berghe, Pierre L. 1964 Caneville: The Social Structure of a South American Town. Middletown, Conn.: Wesleyan Univ. Press.
Van Nieuwenhuijze, Christoffel A. O. 1963 Crosscultural Studies. The Hague: Mouton. → See especially Chapter 4, “Post-colonial Paternalism.” VREELAND, HERBERT H. (editor) 1957 Iran. New Haven: Human Relations Area Files.
Weatherford, W. D. 1962 Pakistan. Pages 11-70 in Walter Galenson (editor), Labor in Developing Economies. Berkeley: Univ. of California Press.
Weber, Max (1906-1924) 1946 From Max Weber: Essays in Sociology. Translated and edited by Hans H. Gerth and C. Wright Mills. New York: Oxford Univ. Press.
Whitehill, Arthur W.; and Takezawa, Shinichi 1961 Cultural Values in Management-Worker Relations: Japan; Ginu in Transition. Research Paper No. 5. Chapel Hill: Univ. of North Carolina, School of Business Administration.
Wolf, Eric R. 1959 Specific Aspects of Plantation Systems in the New World. Pages 136-146 in Seminar on Plantation Systems of the New World, San Juan, P.R., 1957, Plantation Systems of the New World: Papers and Discussion Summaries of the Seminar Held in San Juan, Puerto Rico. Washington: Pan American Union.
The term paternalism has long been in currency among moral and political philosophers, but its circulation became much wider, and its definitions much more precise, following the widely read debate over "the legal enforcement of morality" between Patrick Devlin (The Enforcement of Morals, 1965) and H. L. A. Hart (Law, Liberty, and Morality, 1963). Hart had endorsed the liberal doctrine of J. S. Mill, that the only legitimate reason for state interference with the liberty of one person is to prevent him from harming other persons. Mill was especially emphatic in denying that the actor's "own good, either physical or moral," is ever an adequate reason for interference or criminal prohibition (, 1985, p. 9). What Mill denied in this passage is precisely what came to be called "legal paternalism" in the writings of his followers, including Hart nearly a century later. Thus, paternalism was regarded as a thoroughly unacceptable view by nineteenth-century liberals.
Physical and Moral
In his exchange with Devlin, however, Hart conceded that a certain amount of physical paternalism could be accepted by twentieth-century liberals, here departing from Mill who, he wrote, "carried his protests against paternalism to lengths that may now appear to us as fantastic" (Hart 1963, p. 32). He cited, for example, Mill's criticism of restrictions on the sale of drugs. Devlin then responded by drawing a distinction between "physical paternalism," which protects people from physical harm that could be caused by their own voluntary conduct, and "moral paternalism," which offers similar protection against "moral harm" of the actor's own causing. Devlin could see no consistent way in which the physical paternalist like Hart could avoid commitment to moral paternalism, for if it is the prevention of harm that justifies prohibition in the one case, why not use state power to prevent an equal amount of harm, though of a different kind, in the other case? Similarly, Devlin concluded, there is no relevant difference between criminalization meant to prevent moral harm and criminal prohibitions meant to "enforce the moral law as such." The view that "enforcement of morality," quite apart from harm prevention, is a valid reason for criminal prohibitions is widely called "legal moralism." It is anathema to liberals.
One way in which liberals sometimes defend themselves from Devlin's argument is by maintaining that Devlin's moves from physical to moral harm and from preventing moral harm to "enforcing the moral law" do not follow logically. One liberal critic, Joel Feinberg (1986), even goes so far as to deny, in the teeth of the immense combined authority of Plato and Aristotle to the contrary, that "moral harm" is a coherent concept.
Hard and Soft
A distinction is commonly made between hard (or strong) paternalism and soft (or weak) paternalism. Hard paternalism justifies the forcible prevention of some dangerous but self-regarding activities even when those activities are done in a fully voluntary (i.e., free and informed) way. Soft paternalism, on the other hand, permits individuals or the state to prevent self-regarding dangerous behavior only when it is substantially nonvoluntary or when temporary intervention is necessary to establish whether it is voluntary or not.
Most soft paternalists are liberals strongly opposed to paternalism. Most of them, when they think of the paternalism they oppose, think of what is here called hard paternalism. Therefore they would prefer to go by the name of soft antipaternalists. The term hard antipaternalism could be reserved for the totally uncompromising liberal who would oppose interference even with some choices known to be involuntary, and with temporary compulsory intervention that is only for the purpose of determining whether the intended conduct truly is voluntary, and even with the imposition of compulsory education about risks or state-administered tests to assess the dangerous actor's understanding of the risks, with licenses required for self-regarding dangerous behavior, like mountain climbing. Clarity would be improved if philosophers would speak of paternalism only when what is meant is hard paternalism, justifying prohibition even of wholly voluntary self-regarding conduct, when dangerous. Then soft and hard antipaternalism would be the names of a moderate and extreme liberalism, respectively.
The controversy over paternalism in the criminal law is genuine and difficult. Those who are strongly opposed to paternalism find it not only mistaken but arrogant and demeaning. It is very difficult to reconcile it with even a minimal conception of personal autonomy (rightful self-government) when it proclaims that state officials may rightfully intervene even against my protests to "correct" my choices, and this on the ground that they know what is good for me better than I do myself. But if we reject paternalism altogether, we seem to fly in the face both of common sense and of long-established customs and laws. The state, for example, does not accept "consent" as a justification for mayhem or homicide. Similarly, the law of contracts will not validate certain agreements even though they are voluntary on both sides—when, for example, they are usurious or bigamous. One would be hard put to accept these traditional state-created disabilities without abandoning one's opposition to paternalism. But if we continue our adherence to paternalism, we may discover that in other areas paternalism justifies too much, the flat-out prohibition, for example, of whiskey, cigarettes, and fried foods, which tend to be bad for people too, whether they know it or not.
Writers on medical ethics confront paternalism at every turn, often in human contexts that are less familiar to those whose interest is primarily focused on criminal law. Those characteristic social situations have led to some forms of ethical analysis supplementary to those that prevail among the critics and defenders of "legal paternalism." For example, not all of the moral problems raised by paternalism in medical settings are problems for legislators drafting mandatory rules or other governmental officials such as judges or police officers. Moreover, paternalism is not exclusively a criterion for the legitimacy of coercion. Sometimes what is at issue is some other practice that normally has high moral costs, most notably deception rather than coercion, as in false but comforting statements to frightened patients or the unacknowledged or mendacious use of placebos. Sometimes a medical provider may have to decide whether to tell a "white lie" to his patient, not for the sake of her health, but rather as a way of preventing her from experiencing intense despair in her final hours about a matter having no direct connection with medical treatment. In a hypothetical case invented by C. M. Culver and B. Gert (1976, p. 46), a woman on her deathbed asks her physician how her son is doing, and the doctor replies that he is doing well even though he knows that "the son has just been killed trying to escape from jail after having been indicted [a fact unknown to his mother] for multiple rape and murder." An opponent of (hard) paternalism would probably consider the doctor's mendacity to be a violation of the patient's autonomy. A medical paternalist would probably argue that the truthful alternative in this case would be cruel to the point of indecency. They might both be right.
Pros and Cons
Problems involving paternalism in medical contexts are quite diverse. They include not only truth-telling cases but also suicide attempts, requests for euthanasia, and the use of human volunteers in dangerous experiments. The paternalist position in these conflicts is that protecting volunteers or patients from harm and promoting their benefit should take precedence over respecting their autonomy by permitting them to act freely on their well-informed choices in matters that are almost exclusively self-regarding.
T. L. Beauchamp (1977) and Beauchamp and J. F. Childress (1979) in their influential works rejected hard paternalism nearly categorically, emphasizing that to overturn the deliberate choices of adult human beings that affect only them, or only them clearly and directly, is to deny that their lives really belong to them. The apparent exceptions—cases in which commonsense morality would seem to justify interference with the patient's voluntary choice—invariably turn out to be cases in which that choice is not fully voluntary after all; that is, the patient or volunteer subject had not been adequately informed about the risks he would be accepting, or he was not perfectly free of coercive influences, or some other condition, such as infancy, drug intoxication, high fever, rage, or depression, had diminished his capacity to act rationally. To restrict his liberties in such circumstances, or to motivate him by telling him a lie, would be to interfere with actions that are not fully voluntary in the first place. To interfere with dangerous self-regarding but less-than-voluntary behavior can be justified by soft paternalism (that is by soft and hard antipaternalism). Another example illustrates the point. "If we see a normally calm person who we know has been experimenting with hard drugs, go into a sudden frenzy, and seize a butcher knife with the clear intention of cutting his own throat, then [of course!] we have the right to interfere. In so doing we will not be interfering with his real self or blocking his real will.… His drug-deluded self is not his 'real self,' and his frenzied desire is not his 'real choice,' so we may defend him against these internal threats to his autonomous self, which is quite another thing than throttling that autonomous self with external coercion" (Feinberg 1986, p. 14). Interference on this ground is no more paternalistic than interference designed to protect an individual from an attack by some berserk assailant. Paternalists have been quick to point out, however, that this example, and others like it, hardly fit the more usual examples of risky choice making.
Writing from the practical point of view, and a philosophical position more friendly to paternalism, Culver and Gert (1982), in response to Beauchamp, point out that many crucial questions remain for the soft antipaternalist analysis. Most of these stem from the vagueness of the distinction between voluntary and nonvoluntary. Culver and Gert remind us that voluntariness is usually a matter of degree with no conveniently placed bright lines to guide us. In this respect it resembles the concept of harm (which is also crucially involved in hard paternalists' calculations) and the degree of violation of a moral rule, like that forbidding telling lies, or that condemning coercion, and even the degree to which the overruled choices of, say, a patient, are purely self-regarding—another essential variable.
Culver and Gert, however, do not endorse the hard paternalistic position without limit. Rather, they hold that some (hard) paternalistic interventions are justified, and some are not, but reject the unqualified antipaternalism of Beauchamp and Childress, which denies that (hard) paternalistic prohibitions and interferences are ever justified, and the unqualified paternalism of many utilitarian writers, which holds that all paternalistic behavior is justified, except that which will be counterproductive in the long run.
Arneson, Richard. "Paternalism, Utility and Fairness." Revue Internationale de Philosophie 170: 409–423.
Beauchamp, T. L. "Paternalism and Bio-behavioral Control." Monist 60 (1977): 62–80.
Brennan, Samantha. "Paternalism and Rights." Canadian Journal of Philosophy 24 (1994).
Brock, D. "Paternalism and Promoting the Good." In Paternalism, edited by R. Sartorius. Minneapolis: University of Minnesota Press, 1983. This is one of the leading statements of a qualified utilitarian theory of paternalism.
Culver, C. M., and B. Gert. "Paternalistic Behavior." Philosophy and Public Affairs 6 (1976): 45–47.
Devlin, P. The Enforcement of Morals. London: Oxford University Press, 1965.
Dworkin, G. "Paternalism." Monist 56 (1972): 64–84. An influential early article that helped shape twenty years of discussion.
Dworkin, G. "Paternalism: Some Second Thoughts." In Paternalism, edited by R. Sartorius, 105–113. Minneapolis: University of Minnesota Press, 1983.
Dworkin, Gerald. "Moral Paternalism." Law and Philosophy. 2005.
Dworkin, Gerald. "Paternalism." The Stanford Encyclopedia of Philosophy (Winter 2002), edited by Edward N. Zalta.
Dworkin, Ronald. "Equality and The Good Life." In Sovereign Virtue, 268–270. Cambridge, MA: Harvard University Press, 2000.
Faden, R. R., and T. L. Beauchamp. A History and Theory of Informed Consent. New York: Oxford University Press, 1986. The definitive work on its subject.
Feinberg, J. The Moral Limit of the Criminal Law. Vol. 3: Harm to Self. New York: Oxford University Press, 1986.
Freeman, Samuel. "Liberalism, Inalienability, and Rights of Drug Use." In Drugs and the Limits of Liberalism, edited by Pablo DeGrieff. Ithaca, NY: Cornell University Press, 1999.
Goldman, Alan. "The Refutation of Medical Paternalism." The Moral Foundations of Professional Ethics. Towata: Rowman & Littlefield.
Hart, H. L. A. Law, Liberty, and Morality. London: Oxford University Press, 1963.
Husak, Douglas N. "Legal Paternalism." In The Oxford Handbook of Practical Ethics, edited by Hugh LaFollette. New York: Oxford University Press, 2003.
Kleinig, J. Paternalism. Totowa, NJ: Rowman and Allanheld, 1983.
Mill, J. S. On Liberty (1859). Indianapolis, 1985.
Sartorius, R., ed. Paternalism. Minneapolis: University of Minnesota Press, 1983. This excellent collection includes, in addition to the selections by Brock and Dworkin already cited, fifteen useful articles and a superb bibliography.
Shiffrin, Seana Valentine. "Paternalism, Unconscionability Doctrine, and Accommodation." Philosophy and Public Affairs 29 (Summer 2000): 205–250.
Sunstein, Cass, and Richard Thaler. "Libertarian Paternalism Is Not an Oxymoron." University of Chicago Law Review 70 (Fall 2003): 1166.
VanDe Veer, D. Paternalistic Intervention: The Moral Bounds of Intervention. Princeton, NJ: Princeton University Press, 1986.
Joel Feinberg (1996)
Bibliography updated by Gerald Dworkin (2005)
The practice of paternalism has a long history in American business, and it has been both maligned and extolled. Many companies have taken the opportunity to highlight their “fatherly” actions, claiming to be family businesses that take care of their employees through extra benefits and care that their workers would not be able to afford otherwise. Others have complained that paternalistic methods used by businesses form unhealthy environments, creating company towns that are fully dependent on companies to provide too many aspects of their living conditions. Paternalism, according to skeptics, fosters a top-down style of management that hinders creative thought among employees and encourages harmful promises or inconsistent behavior. However, a new style of paternalism has become more common in recent years as companies have begun exploring ways in which to develop employee innovation and social resources by helping their workers live healthier, fuller lives.
In the late 2000s, paternalism often involves offering employees benefit plan advice (instead of only a limited number of available plans) and compensation packages that provide a number of personalized extras. These include fitness gym memberships and other healthy living options. By supporting employees in these socially conscious ways, companies are able to maximize employee retention and maintain favorable, efficient workplaces.
DEVELOPMENT OF PATERNALISM IN AMERICA
Ronald Sims, in his 2003 book Ethics and Corporate Social Responsibility, identifies three different stages of development that occurred in the progression of paternalism, as companies explored the meaning of social responsibility and employee welfare in the twentieth century of United States business.
The first stage was profit maximizing management, during which business executives decided that their primary goal, for both society and success, was to encourage profit and growth. The American economy was seen as dependent on the wealth of companies, and anything done to encourage that wealth was seen by managers as acceptable. This was before the era of legislation concerning child labor and gender inequality, so some of the accepted methods—while profitable for businesses— were harmful to society. Sims writes that this era ended at the time of the Great Depression (1930s) when the success of business and business practices was called into question.
The second stage was trustee management . After the Great Depression, wealth was more spread out, and fewer privately-owned companies existed. Instead, companies were under the control of stockholders and subject to many outside groups such as suppliers, creditors, and of course customers. Government involvement also increased, leading to legislation that established fair labor principles and welfare systems; such legislation also encouraged the protection of employees. Corporate wealth started to be distributed for the purposes of social responsibility, and corporate authority moved away from one or two top executives to corporate boards and trustees. Employee benefits became common in this stage. Unfortunately, several paternistic problems also arose as employees became dependent on their companies, while companies struggled to not misuse their paternistic authority.
The third stage is the quality-of-life movement , where corporate social responsibility became a popular topic. Companies began inspecting ways that they could profit by their relationships with employees and employees' relationships throughout society. Social capital, the collective ideas and connections of the society, became important to businesses. Focus moved from purely economic goals to wider initiatives involving care of the environment, employee physical health, employee mental health, and employee emotional security. This is the current state of business paternalism, which is growing into a more complex style of management as organizations realize the marketing and image benefits that can be earned through social responsibility, along with the rewards of employee energy and innovation.
PROBLEMS WITH PATERNALISM
The negative effects of paternalism developed largely in the second stage, as companies began to assume that employees did not have enough knowledge or available information to make wise decisions. Therefore, management took the burden of these decisions away from employees, locking some organizations into dangerous, dependent relationships that could be exploited. However, as electronic communication became more common, the excuse of employee ignorance no longer applied, and companies were able to offer their workers a number of choices, providing them with the necessary education to make wise decisions concerning health plans and benefit options.
One of the most popular current ideas under the heading of “New Paternalism” is the idea of “nudging,” or giving employees—and others—small psychological nudges to make the right decisions, rather than forcing them to make certain choices in any direct way. These nudges are designed to help people realize the benefits to choosing health-conscious options without limiting their freedoms. Government nudges include “sin” and “fat taxes” to support beneficial lifestyles. A company interested in “nudging” their employees might offer gym memberships as a part of an attractive and flexible health benefit plan, encouraging workers to maintain physical well-being without requiring it.
Aronoff, Craig E., and John L. Ward. “The High Cost of Paternalism.” Nation's Business. May 1993.
Dowling, John Malcolm, and Yap Chin-Fang. Modern Developments in Behavioral Economics. Hackensack, NJ: World Scientific, 2007.
Goldstein, Evan R. “The New Paternalism.” The Chronicle Review, 2008. Available from: http://chronicle.com/temp/reprint.php?id=pwq4w52rk7wg916xkfflm6r43x0h2d5s.
Sims, Ronald R. Ethics and Corporate Social Resposibility. Westport, CT: Greenwood Publishing Group, 2003.
Winning, Ethan A. “Pitfalls of Paternalism.” Ewin.com, 2008. Available from: http://www.ewin.com/articles/paternal.htm.
Consideration of paternalism involves the interactions of two principles of medical ethics—beneficence and respect for autonomy. Historically, beneficence has long retained primacy in medical ethics, and physicians have been able to rely almost exclusively on their own judgement about their patients' needs for treatment, information, and consultation. However, medicine has increasingly been confronted—especially since around 1970—with a different kind of need, namely the patient's asserted need to make an independent judgment.
The central problem in these discussions is whether the principle of respect for autonomy, which gives primary decision-making authority to patients, should have priority in medical practice over the principle of beneficence, which gives authority to providers to implement sound principles of health care. Resolving this issue requires coming to terms with the problem of paternalism. Medical paternalism poses significant moral questions because it holds that beneficence can legitimately take precedence over respect for autonomy. From this perspective, a professional is like a parent dealing with dependent, and often ignorant and fearful, patients.
For example, suppose an incurable cancer is found in a sixty-nine-year-old man. Based on a long relationship, the man's physician knows that the patient has a history of psychiatric illness and is emotionally fragile. When the patient blurts out, "Am I OK? I don't have cancer, do I?" the physician answers, "You're as good as you were ten years ago," knowing that the response is a paternalistic lie, but also believing it justified in protecting the health and well-being of the patient.
Some leading ethicists maintain that paternalistic interventions are seldom justified, because the right of the patient to act autonomously almost always outweighs obligations of beneficence toward the autonomous patient. In the practical world, it is important to seek a balance between the demands of both beneficence and respect for autonomy. It is useful to note that this balance may be seen differently in other cultures, where there may be a stronger tilt toward beneficence.
John H. Bryant
(see also: Autonomy; Beneficence; Equity and Resource Allocation; Ethics of Public Health )
Veatch, R. M. (1989). Medical Ethics. Boston: Jones and Bartlett Publishers.
A wide variety of social relationships have been described and analysed as characteristically paternalist, including those between husbands and wives, master and slave, employer and employee. The relationship between certain factory owners and their employees, for example during the early phase of industrialization in the West, has often been viewed in this way. The former group exerted almost unrestrained power over the latter. However, as a tactic for securing social control, the early mill-owners attempted to convert power relations into moral ones; or, in the terminology of Max Weber, to translate domination into traditional authority. This was to be achieved by the institutionalization of such practices as periodic gift-giving, charitable religious and educational activity, provision of company housing and insurance schemes, and support for company-affiliated voluntary associations and clubs. One of the most systematic studies of this form of paternalism, which examines employer domination and operative responses in the Northern textile mills of Victorian England, is Patrick Joyce's Work, Society and Politics (1980).
The suggestion is often put that paternalism, practised in this way, is a device for managing and legitimating overtly and potentially disruptive hierarchical and exploitative relationships: it serves the interests of men rather than those of women, the ruling class rather than the proletariat, or of White masters as opposed to Black slaves. However, it has proved difficult to demonstrate empirically that the ritualistic (usually deferential) responses of subordinates to the paternalistic strategies of their superiors indicate identification with or approval of the status quo, rather than merely an external and calculated management of impressions (or what has been called ‘the necessary pose of the powerless’).
pa·ter·nal·ism / pəˈtərnlˌizəm/ • n. the policy or practice on the part of people in positions of authority of restricting the freedom and responsibilities of those subordinate to them in the subordinates' supposed best interest: the arrogance and paternalism that underlies cradle-to-grave employment contracts. DERIVATIVES: pa·ter·nal·ist n. & adj. pa·ter·nal·is·tic / -ˌtərnlˈistik/ adj. pa·ter·nal·is·ti·cal·ly / -ˌtərnlˈistik(ə)lē/ adv.