Introduction

In August of 2021, Nigerian doctors went on a protracted two-month strike over emoluments even as COVID-19 cases surged (Saric, 2021). The Associated Press reported that Kenya’s 5,000 public-sector doctors resumed work on Friday, March 17, 2017, after a 100-day strike. The report confirmed that the 100-day doctors’ strike created a health crisis for millions of Kenyans (Kaguthi, Nduba, and Adam, 2020). In July 2015, doctors in the West African nation of Ghana went on strike. Data compiled by the Health Insurance Service Providers Association of Ghana (HISPAG) revealed that about 500 people died needlessly nationwide within a month of the declaration of the strike (Ghanaweb, 2015). Junior doctors in England went on their first strike in four decades, disrupting treatment for thousands of patients in the National Health Service (Toynbee et al. 2016). The 2016 strike led to the postponement of operations, and appointments were canceled. David Cameron, the Prime Minister at the time, warned that the labor action would create “real difficulties for patients, and potentially worse.” Russo, Xu, and McIsaac (2019) documented 70 health workers’ strikes in 23 low-income countries between 2009 and 2018, accounting for 875 days of strike. These examples confirm a worrying trend: the willingness of physicians these days to resort to industrial action to press for their legitimate demands, something that used to be uncommon in medical history (Thomson and Salman, 2006).

Industrial action (IA) in the medical professions puts patients in vulnerable situations (Isangula, 2012). Vulnerability can be defined as the extent to which an agent is susceptible to harm or abuse (Formosa, 2014). Whether viewed as a universal expression of the human condition (Nussbaum 1992; Rendtorff, 2002), or more narrowly as groups that are more susceptible to harm than others (Formosa, 2014), patients are a vulnerable group and contemporary ethical theories speak to our individual, and collective moral responsibilities to the vulnerable (Kant, 1996; Rendtorff, 2002; Annas, 1993; White 2011). As Rendtorff (2002, p. 237) observes, “the protection of the vulnerable is considered as the bridging factor between moral strategies in a pluralistic society, therefore respect for vulnerability is essential to policymaking in the modern welfare state” (p.237). In clinical care, there is agreement that vulnerable patients require special protection (Haugen, 2010).

Questions of vulnerability, and moral agency has generated contentious debates in moral philosophy (e.g., Strawson, 1962; Rudy-Hiller, 2018, Goodin, 1985a). Therefore, both vulnerability and moral responsibility are at the heart of the core issues in this paper. Talbert (2019) argues that a theory of moral responsibility needs to include the following: (1) defining what moral responsibility as a construct means (2) what qualifies an individual as a moral agent (3) the conditions that must be met for us to hold someone responsible for failure to perform a morally significant thing and (4) actions, omissions and nonactions for which we can link responsibility to.

In this paper, we explore issues at the nexus of vulnerability, moral agency, and responsibility. We proceed as follows. Firstly, we examine theories of moral responsibility. Knowing the conditions under which an individual becomes morally responsible for someone, and therefore worthy of blame or praise is important. Secondly, we review theories of vulnerability. Whether patients are indeed in a vulnerable position is important for determining when we can hold physicians morally responsible when they withhold their services during a strike. We address both issues by drawing on theories of vulnerability, moral responsibility, and ethics. Thirdly, we explore the implications of both vulnerability and moral agency for moral obligations as they apply to physicians’ industrial actions and the limits of those obligations. To determine whether physicians are morally responsible for the vulnerable and would be violating some moral code for which one would be justified to blame them requires us to explore the question of moral responsibility and vulnerability. Finally, we present the theoretical and practical implications of the paper.

The key presumption here is that IA by physicians, and those in the medical professions in general, raise moral and ethical issues and that patients are in a vulnerable position giving rise to issues of moral responsibility and obligations to the vulnerable. We note that doctors have the legal right to strike when, and where the law allows them to, and there are times when such strikes may be in the long-term interest of patients. Indeed, it would be immoral to prevent physicians from using legally mandated tools to achieve their legitimate ends. Together, the opposing rights set up conflicting moral claims and values. We argue that theories of vulnerability, ethics, and moral responsibility may present us with ideas to resolve some of the competing claims and determine the nature of agential responsibility, including its limits, to the vulnerable.

Industrial action in the medical professions: causes and debates

IA occurs when employers, often the government as an employer in health services (for example developing countries, and places like the UK and Hong Kong) cannot agree on labor issues. Specifically, strikes are often a response to an impasse in collective bargaining (Grosskop et al. 1985; Isangula, 2012). As Chima (2013) observes, strikes put physicians in a difficult moral dilemma. As people who have taken a professional oath of duty of care based on “the ethical principles of respect for autonomy, justice and beneficence” strikes create conflict as “doctors struggle with their role as ordinary employees who are rightfully entitled to a just wage for just work versus their moral obligations to patients and society” (Chima, 2013, p 2.). Indeed, in the interest of fairness and justice, it would be unfair to deny physicians the right to strike (Ogunbanjo and Knapp van Bagaert, 2009). The fact is no matter what justification doctors may have for going on strike, the result may be patient suffering and it would seem prudent to use the immediate outcome, not the motivation and long-term impact, as the yardstick for evaluation moral responsibility because the impact of withholding medical care is often immediate and severe as the cases in Nigeria, Kenya, UK, Ghana, and elsewhere have shown.

Physicians go on strike for two main reasons. First, strikes may be in support of the demand for higher salaries and emoluments (Isangula, 2012). Striking, for improved salaries, may be in the self-interest of physicians as individuals. Thompson and Salman (2006) suggest that physicians may increasingly be employed on wage contracts with employers and may join unions, sometimes using strikes as a bargaining tool. The non-payment of wages was listed as one reason for the Nigerian strike. Dr. Okhuaihesuyi, the President of the Nigerian Doctors Association (NARD) noted that his group called the strike over unpaid salaries, COVID-19 hazard allowances, and the benefits to families of doctors who died of COVID-19 (Nwachukwu, 2021). Beyond that, the President of NARD seemed to justify the strike action by invoking the poor state of healthcare in the country. In justifying what appears to be the social good, the President decried the “deplorable state” of Nigeria’s hospitals and appealed directly for public sympathy. He was quoted as saying: “We appeal to Nigerians to bear with us.”

Second, physicians may strike to improve healthcare delivery, thereby acting in the patient’s long-term interest (Ogunbanjo and Van Bogaert, 2009). Such a strike may be justified because it could in the long run generate benefits for a larger number of people. Both the strike in Nigeria and Hong Kong (Li and Ng, 2021) may have a social dimension as both actions could be deemed necessary for the greater good. The fact is the average citizen in developing countries such as Nigeria, Kenya and Ghana fit a subgroup of the vulnerable because of their economic and social conditions (Formosa, 2014) and failure to act in the interest of the greater good in the face of such dire situations may be unethical, and cause moral distress to physicians (Jameton, 2013). Moral distress is a situation people face when institutional constraints prevent them from doing their work. In the case of Israel, physicians went on strike in 2011when negotiations between the Israeli Medical Association (IMA) and the government collapsed. The IMA demanded that hospital capacities should be increased, 1000 additional doctors hired, and salaries doubled (Weil et al. 2013). Indeed, Ogunbanjo and Knapp van Bogaert (2009) argue that it would amount to the neglect of the vulnerable when doctors do not protest substandard facilities in hospitals, and if it takes a strike to change things, then it would all be worth it. While this calculative, utility-based argument may be contradicted by other views, calls for involving the public in healthcare issues are consistent with several theories discussed in this paper. Indeed, the American Medical Association (AMA) Code of Ethics (1.2.10) argues that “physicians have an ethical responsibility to seek change when they believe the requirements of law or policy are contrary to the best interests of patients.”

Ongoing research in the medical and bioethical field has addressed the implications of physician industrial action in both developing and developed countries (Waithaka et al. 2020; Abbasi, 2014; Selemego, 2014). The emerging literature seems to diverge along two paths. First, are those who argue that under certain conditions, such strikes may be justified. For example, Selemego (2014) develops six criteria, using the “justum bellum” (just war) hypothesis (Singer, 2004) to justify the fact that under certain specific conditions, strikes, as much as wars, may be justified. The just war hypothesis may be a pragmatic one; however, it fails to account for the position of the patient who is in a similar position as a non-combatant in a war. Those justifying strikes for any reason imply that denying doctors the right to strike would be immoral (Chima, 2013).

Second, is the strand of research, generally, more sympathetic to the patient, that argues strikes are unjustified because they cause harm, and because doctors have a fiduciary responsibility to patients (Abbasi, 2014; Chima, 2013). For example, Mawere (2010), using African communalism and utilitarian ethics, argues that doctors owe the community of which they are a part, a sacred duty to care for patients, and strikes are unjustified as they harm the community and violate the communal duty of care. Similar arguments have been made about the implicit “social contract between doctors and patients” (Ogunbanjo et al. 2009; Abassi, 2014). In other cases, moral responsibility based on religion has been used to make the point that striking for any reason would be violating fundamental moral laws of caring for the ill (Rosner, 1993). Sachdev (1986) suggests that strikes have negative effects on patients and that physicians may be violating their ethics when they strike. While both views have some validity, a more helpful approach may be exploring whether patients are in a vulnerable position to start with, and secondly, the conditions that must exist for us to hold physicians as moral agents, therefore worthy of blame or praise and the limits of that responsibility as physicians cannot be expected to bear a more than reasonable responsibility for others.

Research gaps and objectives

The emerging literature on this topic has increased our understanding of the reasons for, and consequences of strikes in the medical professions (see, for example, Essex and Weldon 2021; Brecher, 1985; Fiester, 2004). Despite the increasing research on the subject, gaps remain in our understanding of the issues and calls have been made for further exploration of the moral implications of doctor’s strikes (Abbasi, 2014). This study responds to that, and similar calls, and aims to fill some of those lacunae. First, most of the existing studies have tended to be fairly atheoretical. Considering that the patient-doctor relationship raises ethical questions, exploring theories of moral responsibility, vulnerability and obligations would be useful to our understanding of the issues. Second, issues of vulnerability of patients are largely assumed away. That may be a serious omission, given that knowing whether an individual is in a vulnerable position or not is a condition precedent for ascribing blame or praise and for activating our reactions towards such agents (Strawson, 1962). Third, determining whether doctors are worthy of condemnation when patients suffer the consequences of their industrial action requires annotation of the links between moral responsibility, vulnerability, and moral obligations. Finally, are the nature of moral responsibility and moral obligations that arise in the presence of vulnerability, including its limits. Greater clarity on issues of responsibility and obligation would be helpful as not everyone agrees that doctors have any special obligations to patients (Brecher, 1985). As Weissman (2019, p. 263) notes, “the vulnerability of a patient gives rise to special obligations to provide aid, but the extent of our obligations to those vulnerable is not always clear.“ Addressing these issues requires us to explore issues at the nexus of moral responsibility, vulnerability, and moral obligation.

Nature of doctor-patient relationship

In most of the world when public health facilities are concerned, the relationship between the patient and physician is unique because there is no direct relationship between the doctor and patient, as the government is an important intermediary. Isangula (2012) proposed a tripartite relationship between doctors, the government, and patients. The government provides the infrastructure and resources for doctors to be able to treat patients. Doctors have service contracts with the government which hires and pay them. Doctors in turn treat patients whose relationship here is direct with the government. Sachdev (1986) argues that the care of the patient, therefore, is the joint responsibility between the government and doctors as the nature of care doctors render to a patient is dependent on how well the government supports the doctors with facilities. The example of Nigeria is again relevant here as the doctor’s association placed responsibility for the strike on the government’s failure to provide a satisfactory work environment.

Moral responsibility and moral agency

Moral responsibility and agency have generated vigorous debates in Western philosophical thought with some of these debates going as far back as Homer’s time (Talbert, 2014; Feldman and Skow, 2020). At the heart of those debates is the central idea of when and under what conditions is an individual either blameworthy or praiseworthy for an action or inaction. Not all actions are blameworthy. For example, some researchers identify what they call morally neutral acts as those actions where a person can be morally responsible but not blameworthy (King, 2012). Morally obligatory actions are those that people should do, and moral supererogatory acts are activities that are especially praiseworthy or even heroic. An act is morally wrong if it violates our basic humanity, for example, murder. Moral acts (sometimes called human acts) are acts that an agent deliberately chooses through the exercise of their free will, using their judgment and conscience (Alyssa, 2015).

Moral Agency

Knowing whether a physician qualifies to be a moral agent or not is important and theories of moral agency may help answer that question. According to Behdadi (2021, p. 226) “a moral agent is an entity considered to be able to do wrong (or right) and typically taken to be morally responsible for actions, omissions, beliefs, and/or character traits.” Implicit in this definition is the fact that such an agent can be held responsible for their behavior. Behdadi (2021) identifies two main streams of thought on when an agent can be deemed to be a moral agent.

First are the capacity-focused approaches to determining moral agency. This approach to moral agency rests on a key factor: a moral agent is deemed to have certain personal mental capacities. Amongst others, such a person becomes a moral agent only “if it can (morally) understand, reflect on, and evaluate potential or actual actions, omissions, or character traits of oneself and others” (Behdadi, 2021, p. 226). MacIntyre (1999) similarly argues that moral agency depends on specific intrapersonal capacities or capabilities. Those who argue that animals for example are incapable of becoming moral agents base their arguments on the fact that animals may be incapable of possessing these advanced mental capacities (Rowlands, 2012). To MacIntyre (1999), a moral agent would only be liable to be held responsible for their actions when the act for which we are holding them responsible was done intentionally, that the individual is aware of the consequences of their action, and it is reasonable for them to predict the consequences of their action. MacIntyre (1999) acknowledges that it is not always easy for individuals to meet these three criteria. For example, a judge who releases a suspect arrested for a minor traffic offense who goes on to murder someone on the same day of their bail could not have reasonably known such an individual would commit murder. The workaround, according to MacIntyre is that we hold individuals responsible as moral agents when an act violates societally determined standards of behavior. Violation of such standards exposes individuals to be held responsible and thereby subject to be blameworthy. To MacIntyre (1999), being a moral agent means there is no separation between one’s values and character (qua persona) and how one acts in their official role. In addition, such individuals must act as rational agents capable of making critical judgments and realize that they are accountable for their behavior as rational individuals (qua persona) as well as in their official roles.

Second, is the practice-focused approach to moral agency. While the capacity-focused approach to moral agency relies on the capacities of an individual, the practice-focused approach relies on “moral behavioral patterns and social interactions” (Behdadi, 2021, p. 228). Practice-based approaches to moral agency are based on the idea that we form social expectations and develop dispositions that we internalize and our reactions to violations of these norms invite certain kinds of reactions: either blame or praise. Anyone who we determine is a participant in this moral responsibility regime, is therefore deserving of blame or praise. Expectations about the social practice in which an agent is situated are the key element of this framework. Moral agency arises when others associate, attribute or hold expectations of moral responsibility of the social practice with which a person is associated. The practice-based approach is based on Strawson’s (1962) “reactive attitudes” view. According to this view, there is a pattern of holding people responsible (we have some everyday reactions) based on what we think normal people in social life and relationships should do. This “participant reactive” view assumes that people are in some personal relationships with others and naturally expect that such people would act based on goodwill. We, therefore, react with praise, or blame if we determine that an individual has violated expectations, which we normally expect from them in their social practice. We excuse the behavior and responsibility of a moral agent when we believe that an act that violates our natural expectations was justified or done accidentally, or that the agent did not have all the knowledge to act. An extension of the practice-based approach to moral agency is McKenna’s (2012) “moral responsibility exchange” framework. This framework of moral agency is based on an agent’s capacity to engage in moral conversations, using specific language, and those unable to understand that sort of language are excused from being moral agents.

The nature of moral responsibility: theoretical perspectives

Holding someone responsible for a behavior confers some expectation of moral agency on that individual. Feldman and Skow (2020) argue that we often give blameworthiness greater emphasis because we associate blame with liability and the corresponding sanction. After all, everyday ideas of moral responsibility connote a sense of obligation or duty. Feldman and Skow (2020) distinguish moral responsibility from causal responsibility as people can be morally responsible for something for which determining causality would be difficult. Moral responsibility has been a subject of intense philosophical debates for years setting up competing claims and tensions (Strawson, 1962, King 2012; Wallace, 1994; Watson, 2004). At the heart of those debates is determining when an agent is a candidate for blame or praise. We can blame or praise people for things they did or did not do (other-directed) or people can blame themselves for their actions or inaction (self-directed) in which case we would be suggesting that “someone got their just deserts: According to Feldman and Skow (2020), the typical desert claim is a claim to the effect that someone, “the deserver” deserves something, the “desert” in virtue of his or her possession of some feature, the “desert base”. Other researchers have linked moral responsibility to the desert (King, 2012; Smilansky, 1996). Linking desert and moral responsibility seems appropriate for our purposes as it links those attitudes that fit specific actions and nonactions.

Moral responsibility skeptics

At the heart of some of the greatest debates on moral responsibility is the idea of free will and whether agents have moral responsibility and are therefore subject to blame or praise or not. Incompatibilists or moral skeptics rest their primary argument on the fact that because people have no free will, we cannot assign moral responsibility. The presumption is because events are presumed to be predetermined by forces beyond our control and because human action is often a result of factors outside of our control, people cannot be morally responsible for their actions. For example, Galen Strawson’s (1994, p.13–14) Basic Argument (BA), suggests that no one is ultimately responsible (UR) for anything because people do what they do because of the way they are (causa sui). Perhaps the strongest case against moral responsibility is the BA model.

Strawson (1994) and his intellectual fellow travelers are in effect discounting the possibility that people can act otherwise. With roots in determinism and Homerian ideas of fatalism (Talbert, 2014), incompatibilists argue that because things are predetermined by forces beyond our control, and that we have no free will, no one can be truly morally held responsible. What Galen Strawson seems to suggest is that an agent needs to have control over all the conditions that lead them to choose before they are either blameworthy or praiseworthy and because that basic precondition cannot be satisfied, we can never truly hold people responsible (UR). Istvan (2011) suggests that Strawson’s basic argument may have some validity but interprets the BA in terms of some moral responsibility MC, only if the person has a “sliver” of responsibility for the action. Other researchers argue against the BA. For example, Kane (2000, p.407) argues that to be UR, one must be responsible for a crucial portion of the reason for the action happening. To Kane, accepting BA means that an agent is not morally responsible even for “intentional and consciously deliberated actions” (p. 442). Fisher (2006) argues that Strawson is merely setting up a strawman when he suggests that because agents do not have control over all conditions preceding an action, they are not morally responsible. Galen Strawson’s BA and UR may simply be adopting a too rigid requirement for moral accountability. Robson (2017) argues that “the way you are” (causa sui) requires far greater clarity to serve as an anchor for such a general principle.

Resolving this intractable philosophical issue is beyond the scope of this paper. However, we note that to accept the BA would lead us to ask if agents can ever be morally held responsible for anything. The fact is whether deserving or not, using the “just desert argument” people ascribe praise or blame based on their evaluation of an agents’ action. Strawson’s (P. F. 1962) approach to moral responsibility may present us with one tool for making sense of the conundrum as it may allow us to balance compatibilism and incompatibilism. According to Strawson (1962), we hold people responsible, not necessarily because they are, but because we hold them responsible (Fischer and Ravizza, 1993). Strawson’s (1962) “reactive attitudes” position suggests that we hold people morally responsible when we determine that their actions or non-actions violate some specific attitudes that we hold about people we are in some personal relationship with. We then express praise or anger if we determine there has been a violation (the participant’s reactive attitudes). For example, a patient who has a long-running relationship with the primary physician is denied treatment because the doctor is part of strike action, would blame the physician. In this case, the blame is irrespective of the presence or absence of the causa sui.

Strawson (1962) argues that such reactive attitudes can be suspended if the behavior was accidental or if we determine that such behavior was justified, based on the greater good. Using this condition, one could argue that doctors who go on strike to push the Government to provide better healthcare would be excused and not blameworthy. Also, some moral skeptics argue that we may not be morally responsible for an act either because we are morally impaired (Shoemaker, 2011) or because we were morally lucky, a situation in which factors beyond the agent’s control affect our evaluative judgment of their action (Levy, 2011).

Arguments for moral responsibility

Compatibilists, or those who support the idea that free will and moral responsibility are still possible, even if we accept that determinism, and therefore external forces control what we do, accept that agents have moral responsibility. The genesis of this school of thought is that one always has the choice “to do otherwise.” For example, Fara (2008) argues that people confronted with a situation can always act in ways they see fit, and so determinism may not always apply; indeed, some argue that moral responsibility does not even require the ability to do otherwise (Frankfurt, 1969).

Feldman and Skow (2020) summarize what they term forward-looking accounts of moral responsibility. At the core of this perspective is the idea that determinism does not preclude independent choice of behavior, and such accounts are largely consistent with prevailing compatibilist views. Moral responsibility has been expanded to include attribution, accountability, and lately answerability (Shoemaker, 2011, 2015). Both attributability and accountability suggest that we can hold people morally responsible because they are, and they are responsible because we hold them responsible. Attributability is based on the character of the agent. This view implies that it is an agent’s “moral personality” (Hieronymi, 2008) that enables him or her to make some evaluative judgment about whether to engage in a moral act or not. We blame the agent as a form of moral protest when we determine there has been a moral failure (Talbert, 2012). Accountability is about holding agents responsible for an act. An agent is accountable for an act when they are responsible for that event happening. Shoemaker (2012) adds answerability to attribution and accountability. Answerability responds to an agent’s evaluative judgment. Shoemaker (2012, p.631) argues that for someone to be attributable-responsible for an action, we must be able to properly attribute that action to the individual and for someone to be answerable-responsible for an action, that individual must be able, in principle to justify why they engaged in the act. According to Shoemaker (2012, p.631), the “primary targets of assessment of answerability are the reasons the agent took to justify an action”.

Ethical theories and moral responsibility

In addition to the mainly situational theories of moral responsibility and agency discussed earlier, ethical theorists have looked at moral responsibility and obligations. While consequentialists would ask what rules of behavior physicians should follow, virtue ethics would ask what kind of an individual a physician ought to be and what sort of behaviors they should learn. Answers to both questions determine each perspective’s claims on moral judgments.

Kantianism, vulnerability, and moral responsibility

Kant’s original views, like those of other philosophers, need to be viewed within the context of the times they lived in (Brender, 2004), which naturally leads to multiple interpretations of Kant’s works (White, 2011; Robinson, 2019a). However, critically viewed, Kant leaves us with guidelines for a greater understanding of moral personhood that still resonates with us and may help address some of our most intractable moral problems such as the ones we are discussing here: the notion of duty and goodwill are most important in the context of physicians strikes and duties to vulnerable patients. As Scheneewind (1998) observed, Kant’s overall focus on the human condition, the dignity of all persons, means that his views present us with resources that may be useful for addressing contemporary moral issues such as those we are discussing in this paper. Indeed, Kant (MM 6: 421–447) defines character “as the awareness of one’s obligations towards others and a commitment to pursue those obligations” (Robinson, 2017, p.134).

Kant’s core focus on human dignity, driven by goodwill, is consistent with care for the vulnerable. Resting on three key principles of autonomy, dignity, and duty, Kant’s moral philosophy addresses issues at the nexus of moral agency and duty to the vulnerable (White, 2010). First, individual autonomy holds that we make moral decisions based on our moral judgment of what we think is right, unclouded by any external influence, unless necessary (White, 2011). Kant’s second focus is on dignity, the intrinsic worth of every human being (White, 2011). Hill summarizes Kant’s concept of dignity as follows, “the root idea of dignity is simply that virtually everyone, regardless of social station, talents, accomplishments, or moral record, should be regarded with respect as a human being.” Kant is emphatic that the dignity of a person cannot be traded for another source of dignity (White, 2011). Finally, is the idea of duty. Kant is emphatic about an individual’s moral obligation to be beneficent towards those in need. To Kant (1997, 6:454) “the happiness of others is, therefore, an end that is also a duty.”

Kant differentiates what he calls perfect from imperfect duties, a distinction that may be relevant to the agency of physicians. Perfect duties are precise duties that one must do without exception, except when there are conflicts in moral obligations. In effect, perfect duties are moral maxims. Imperfect duties, to oneself and others, to the contrary, involve general attitudes that require a consideration of situational, including social factors in decision making (Robinson, 2019; White, 2011). A physician’s beneficence towards a patient is an example of an imperfect duty. In his MM (1797, 6:394), Kant equates imperfect duties with virtue, and wide obligations to others (Ibid, 6: 390, cited in Robinson, 2019, p.123).

Kant suggests that the nature of obligation under imperfect duties relies on the closeness between persons. The rationale here is that we have a better understanding of what a vulnerable person or group needs when we have a more intimate knowledge of what such a person needs and therefore a clearer idea of what would be the most effective form of beneficence, by way of duty, we should perform (Robinson, 2017). Indeed, the idea here is that the more intimate the agent is to the beneficiary and the more knowledge an agent has about a beneficiary, the higher the imperfect duty, as closeness and knowledge allows us to both know the needs of the beneficiary and the effects of our duty (Robinson, 2017). We would conclude that physicians are close to patients; have intimate knowledge of diseases and so the precariousness of a patient’s situation if left untreated and more importantly, the effects of the treatment (duty). These conditions imply that the imperfect duties of physicians to patients are enlarged, not reduced. In developing country cases such as Nigeria, physicians understand the average patient may be in dire conditions when they show up for treatment.

Kant’s moral philosophy, notably his position that we must fulfill certain duties and his advocacy for some universal moral law indicate that Kant accepts that human beings must be moral agents, who may be praiseworthy or subject to blame for their performance or non-performance of their moral obligations. At the core of Kant’s moral philosophy is his belief that everyone deserves respect and goodwill. To Kant, morals come from reason, not from any passions. Kant suggests that people are morally responsible because they are rational and can decide what they should do. Kant suggests that we have obligations or moral duty to act in a way that reflects goodwill towards others. Kant’s “supreme principle of morality” is his Categorical Imperative (CI). Kant talks of the moral law in his Metaphysics of Morals when he writes: “the universal imperative of duty [the CI] can also go as follows: act as if the maxim of your action were to become by your will a universal law of nature” (Kant, 1996 p. 209).

White (2011) notes that Kant’s CI may be a tool for deriving guidelines for moral action, as well as a reminder of why moral action is important. Kant’s concept of duty emphasizes that our beneficence must be viewed from the point of the beneficiary, not the agents because it is the happiness of the others, not just the self, that is important (White, 2014, p.110). One implication here is that the doctor must view their beneficence through the lens of the patient. The supremacy of duty in Kant’s philosophy is such that he defines virtue in terms of “the moral strength of a human being’s will in fulfilling his duty” (MM 6:405) and vice as principled immorality (MM 6:390). Kant formulates his CI with the following maxims: (1) “act only by that maxim through which you can at the same time will that it become a universal law” (G 4:421). (2) ‘So, act as to treat humanity, whether in your personal or in that of any other, in every case as an end, never as a means only”. Kant cautions us that people should never be treated as an end only. While this maxim would seem to contradict utilitarian arguments that justify a strike based on its perceived long-term benefits to the public (Li and Ng, 2021), a critical reading of Kant shows that Kant’s focus on individual autonomy does not negate his concern for the welfare of groups or society. For example, White (2011) observes that although the moral agent is individualistic under Kantian ethics, this does not mean that people should be selfish or self-centered in making moral decisions, a position Kant himself rejects (IA 495ff). To O’Neill (2003, 2004), that would be a misinterpretation of Kant because links morality to autonomy. Kant’s views on autonomy and moral decision-making have also come under criticism because autonomy has often been interpreted as a license for individuals to make moral law, a situation that can become untenable if there are conflicting bases for those individual decisions. The criticism is that autonomy may undercut morality as people can make moral choices in their self-interest. However, it is noted that this idealization of human agency “need not be equated with rugged individualism” (O’Neill, 1989, p.75) to the exclusion of the duty to others as exemplified in the concept of imperfect duties to others (MM, Kant, 1996).

John Rawls’ (1999) theory of justice, partly Kantian, offers an additional tool for moral decision-making around physician strikes. Rawls’ ethical philosophy rests on the concept of justice and fairness and may have some relevance to the issues being discussed here. Rawls focuses on the principle of equity and fairness and introduces an original position from which agents can make decisions, his “veil of ignorance” concept in which the decision-maker is unaware of who is going to benefit from their choice. Rawls develops several principles of fairness and equality and argues for the respect of people’s basic rights and the duty to help those who are worse off in society.

Utilitarianism, vulnerability, and moral responsibility

Utilitarianism has traditionally been a powerful framework for public policy, helping to protect the vulnerable in society, and may yet be a powerful framework for public health and patient welfare today (Savulescu et al. 2020). At the core, utilitarianism argues that acts that produce the best outcomes for the majority (act-utilitarianism) and rules that produce the best consequences (rule utilitarianism) are the correct moral choices to make. Utility maximization is the single measure of morality under this perspective. It is important to remember that the focus on the greatest good should not mean that the interests of the minority are ignored. Indeed, Jeremy Bentham cautions that the interest of the powerless should hold sway over that of the powerful when we calculate utility (McCloskey, 1957). Utilitarianism suggests that “we are morally obligated to abstain from inflicting harm, to actively prevent harm, to actively provide for all persons” (Edwards, 1986, p. 125).

Hare (1981) suggests that in deciding what is the moral good, decision-makers can make decisions based on intuition or in a reflective manner. While the former relies on a rule of thumb, much like Kant’s perfect duties, making decisions reflectively implies that decision-makers need to carefully consider the impact of their decision, including how it helps the vulnerable. As a consequentialist theory, this framework suggests that an agent’s moral action ought to be based “solely on their impersonal and objective evaluation of the consequences of their actions” (Da Silva, 2018). Consequentialists hold “that the rightness of an act depends solely on its ability to confer the greatest good to the greatest number of people”. Based on this, one might argue that if a strike is motivated by forcing the government to provide better facilities for the hospital, then such a strike may be justified. Strikes that happen, in principle as a means of helping improve healthcare delivery may be justified based on a utilitarian approach alone. For example, Li and Ng (2021) in justifying a strike by medical personnel in Hong Kong, and seeking public empathy noted:

We hope and believe Hong Kong citizens can understand our action, as we were striking for every single Hong Konger’s health and the public interest. It’s because of the nature of our occupation, our workplace safety has a significant influence on everyone’s safety. Do you know what I mean? Our workplace is full of health risks. We are risking our lives to protect public health. (p.8)

Using the Nigerian example for illustrative purposes, one can argue that there may be a compelling need for Nigerian doctors to strike as means of protecting the vulnerable. For example, the WHO (2021) report shows that Nigeria has the fourth-worst healthcare subsystem in the world with a physician-to-patient ratio of 1.95 doctors for every 1000 patients, no public health insurance, and health facilities collapsing (Nwachukwu, 2021). The lack of facilities places patients in vulnerable positions. Forward-looking accounts of utilitarianism avoid a simplified view of morality and instead offer “a more complex moral structure that accepts moral costs and dilemmas” (Slote, 1985, p. 168).

Virtue ethics and moral responsibility

Although both Kant and consequentialists contain treatments of virtue and character (Nussbaum, 1999), virtue ethics has emerged as a separate ethical theory that places character at the heart of its theory (Anna, 1995). Dating back to Aristotle and the Stoics, virtue ethics may be considered a tool for moral deliberation. Specifically, it provides us to ask what kind of individual a physician should be.

Aristotle in his Nichomedian Ethics (NE) lays out his philosophy on moral responsibility. To Aristotle, the goal of life is to create the conditions that lead to personal happiness and the good life (eudaimonia). Eudaimonism holds that only the virtues that promote the overall well-being of an individual are desirable. Achieving a fulfilled life requires the development of virtues (arete) such as humility (MacIntyre, 1984). Two kinds of virtues are especially important: moral (character) and intellectual virtues (reasoning). While character allows us to make the right choices in life, reasoning allows us to develop moral sensibilities so we can effectively relate to situations (MacIntyre, 1984), and agents become virtuous through practicing the virtues (Annas, 1993). Developing phronesis, or practical wisdom, a key intellectual virtue, is important as this allows us to make good moral judgments . Contemporary virtue ethics (Nussbaum, 1958), tilts away from eudaimonia because it seems to focus on the self and suggests, instead, that the virtuous person may sometimes have to sacrifice their happiness for the greater good of others and accept that the full realization of every individual human life is the goal of happiness (Annas, 1993; Hursthouse and Pettigrove, 2018).

Aristotle suggests that we praise and blame people concerning their virtues and vices (NE 1106a1-2). According to the Aristotelian theory of moral responsibility, we can assign praise or blame based on an agent’s dispositions, character, and actions. Aristotle notes that only agents who possess the capacity for decision, act voluntarily and not under any duress, and are aware of the consequences of what they are doing can be held morally responsible (Bobzein, 2011). It is fair to argue that physicians have the capacity for decision when it comes to strikes, even if their participation may not be completely voluntary. Virtue ethics seems more useful when we consider that context is an important element in moral decision-making. Virtue ethics suggest that agents need to not only consider whether an action is right or wrong; they must consider the context and the situation. According to Aristotle, “a virtuous person knows how to feel or act towards the right person, to the right extent, at the right time, with the right aim, and in the right way…. that is not for everyone, nor is it easy.” (NE 1109a26-9, cited in Da Silva, 2018, p. 6). It is important to note that virtue ethics does not completely reject broad rules because it is not easy to separate moral rules from moral reasons as moral reasons have some generality (Chappell, 2014). Virtue ethics is often criticized for a lack of guidance on how an agent is to behave. Da Silva (2018) notes that Aristotle may have provided the seed for the idea that virtuous agents are likely to make reliable moral decisions always as these virtues are acquired over time, as part of an agent’s upbringing and education. Da Silva (2018) suggests that virtuous agents use moral perception, and more importantly, empathy, “the capacity to re-experience the mental states of other people” (p.9).

Theories of vulnerability

Vulnerability has been defined in two ways. First, vulnerability has been defined broadly in terms of the human condition (“conditio humana”, Adriaanse, 2011; Naussbaum, 1992; Rendtorff, 2002). Our corporeal nature makes us vulnerable in many ways. For example, Mackenzie, Rogers, and Dodds (2014) define vulnerability as part of the human condition when they note that vulnerability implies being “fragile, to be susceptible to wounding and to suffering; this susceptibility is an ontological condition of our humanity” (p.4). To Fineman (2008, p.8) vulnerability “is a universal, inevitable, enduring aspect of the human condition.” Second, Sossaver, Schildler, and Hurst, (2019, p. 3) define clinical vulnerability of patients in terms of individual characteristics. Fineman (2008, p. 8) notes that clinicians define clinical vulnerability as both a part of the human condition and individual characteristics. Formosa (2014, p.89) provides a more explicit definition of vulnerability:

To be vulnerable is to be susceptible to harm, injury, failure, or misuse…. vulnerability implies that x is susceptible to y being inflicted by z, where y is some harm, injury, failure, or misuse, and x and z are some person, animal, object, event, or group.

These definitions of vulnerability differ from its use in theology where vulnerability is not equated with dependency, and powerlessness, but rather a quality reflective of Man’s humility (derived from the Imago Dei tradition) and something desirable (see, for example, Keenan, 2020). We may be vulnerable for several reasons. First, because we are human. In other words, our human condition makes us vulnerable (MacIntyre, 1999). Second, we may be vulnerable to those we are in a relationship with. Agents or groups are vulnerable to specific forms of harm or threat by others. As Goodin (1985) argues, “any dependency or vulnerability is arguably created, shaped, or sustained, at least in part, by existing social arrangements. None is wholly natural” (p. 191). People are vulnerable because of “the idea of asymmetrical power that characterizes relationships” (Clark, 2020, p. 1). McKenzie et al. (2014) suggest that agents are vulnerable especially when they lack the temporal ability to protect themselves or if they find themselves in an asymmetrical position of power, are dependent on someone, or are in a particular situation that renders them open to harm by others (p.6). Finally, people belonging to some subgroups may be more vulnerable than others (Formosa, 2014). Pellegrino and Thomasma (1996, p. 208) suggest that patients are in exactly that kind of situation. The authors describe the vulnerability of a patient and their loss of freedom in terms of a “wounded humanity”.

There are several forms of vulnerability. For example, researchers in bioethics have a narrower definition of vulnerability (Racine and Bracken-Roche, 2017). Their main concern seems to be situations where vulnerability is tied to a specific situation or context, and that vulnerability may be short-term. Scully (2014) calls this sort of temporary vulnerability contingent vulnerability. Other forms of vulnerability exist. Pathogenic or existential vulnerability (Weissman, 2019) occurs when actions we take to reduce the conditions of a vulnerable person end up worsening it. For example, if doctors go on strike to pressure the government to improve facilities so patients can be better treated, that strike nonetheless may end up making the patient’s situation worse as care may be postponed. MacKenzie et al. (2014) note that “both inherent and situational vulnerability may be dispositional or occurrent”. Dispositional vulnerability is conditional, which is, likely to occur, while the occurrent vulnerability is real and immediate. A patient at the hospital experiences occurrent vulnerability but our humanity puts us in a dispositional-vulnerable position. MacKenzie et al. (2014, p.9) note that occurrent vulnerability can “promote a sense of powerlessness, loss of control and agency”.

Any kind of vulnerability affects people and a higher dispositional vulnerability (maybe due to lack of adequate medical facilities say in developing countries) may lead to higher occurrent and pathogenic vulnerability for patients. Formosa (2014) argues that the idea of dispositional vulnerability plays a key role in developing Kant’s view of developing universal duties, including his injunction of fulfilling our moral duties to the vulnerable. We presume that patients are vulnerable as part of their human condition, in terms of their individual characteristics, and the context they are, in a doctor-patient relationship. More importantly, a patients’ vulnerability is both inherent, may be immediate, and existential. For example, a patient with a life-threatening infection may die unless they get immediate care, and that puts them in an immediate, and pathogenically vulnerable position. Even less serious ailments may develop into more serious cases unless treatment is quickly administered.

Vulnerability, moral responsibility, and moral obligations

In determining who bears primary responsibility for the vulnerable, MacKenzie et al. (2014), suggest that two principles may be vying for mastery. First, is the idea that vulnerability, sui generis, connotes moral obligation (Goodin, 1985; Weissman, 2019; Pellegrino and Thomasma, 1996). The idea here is that because people in vulnerable positions lack control, lose autonomy, and are dependent on other people, we must activate our moral obligations and render care. Related to patients, Pellegrino and Thomasma (1996) argue that we have a “special obligation to alleviate the harm patients face because being ill renders patients vulnerable”. The authors express the vulnerability of patients by noting that:

The patient is no longer free to make rational choices among alternatives. He lacks the knowledge and the skills necessary to effect a cure or to gain relief from pain and suffering. In many illnesses, the patient is not even free to reject medicine, as in severe trauma or other overwhelming emergencies (p. 208).

Pellegrino and Thomasma (1996, p. 209) argue that “medical professionals have a special obligation toward patients, and conditions for fulfilling this obligation extend to all of society”. Views equating vulnerability to moral obligation and duty of care, reject any notions of an agent having a voluntary duty, or the assumption that there is some reciprocal equivalence between an agent and a vulnerable person. However, Scully (2014) suggests that there may be a reciprocal relationship between dependence and vulnerability, at least as it relates to the disabled. Goodin (1985) for example, argues that dependency breeds moral obligation, and people often do not choose how they become dependent on another person. Instead, dependence is held as a trigger for vulnerability and moral obligation (Nussbaum, 2006; Goodin, 1985a). Of course, there are times when a vulnerable person is only temporarily dependent on someone else (Fineman, 2008).

The second view linking vulnerability to moral obligation and care suggests that while vulnerability does not lead to obligations, vulnerability serves as a normative trigger that alerts us to the need to render care as someone may face potential harm if we do not act. For example, bystanders routinely jump into action to save a person from a burning car, a behavior that is certainly consistent with our everyday moral intuitions. Miller (2013) extends this second view by suggesting that special kinds of needs, what the author calls “foundational needs”, or needs that if unmet would result in harm, are the classes of needs that invite moral obligations and duties as the burning car example indicates. Some patients have foundational needs, depending on the severity of their ailment, and in the context of developing nations, it is fair to say that most patients have foundational needs.

In his book Protecting the Vulnerable, Robert Goodin (1985) identifies the central premise of his argument for moral obligation to the vulnerable in the following terms: “It is dependency and vulnerability rather than voluntary acts of will which give rise to. . our most fundamental moral duties” (p. 34). Goodin’s (1985) ethical approach provides one avenue for understanding who bears responsibility, the ways in which we fulfill our obligations to the vulnerable, and some of the limits of that obligation and we borrow, quite freely, from that work here. Goodin’s moral responsibility theory seems appropriate as it is consistent with both Kant, consequentialist, and moral responsibility theories and with our everyday moral intuitions. Goodin’s moral theory has been used to explain moral responsibility in contract medical research (Adobor, 2012), elderly care (Tong, 2014), and outsourcing of foreign labor (Pierlott, 2004).

Robert Goodin’s (1985) theory of vulnerability, puts duties to protect the vulnerable at the center of moral obligation. This is Goodin’s “vulnerability principle”. For Goodin (1985, p. 194, vulnerability is to be understood in terms of “one person’s interests being vulnerable to another person’s actions or choices.” Goodin (1985) presumes that those deserving of our moral obligations are already in a vulnerable position and our actions to help may further expose them to harm. According to this theory, those in positions of power have a special duty to those over whom they have power, especially those that are dependent on them. To Goodin, the mere fact that the patient depends on the doctors’ care to get better places them in a vulnerable position to start with. Research has shown that doctors have asymmetrical power when it comes to patients (Pellegrino and Thomasma 1996, p. 208). Therefore, Pellegrino and Thomasma (1996) argue that medical professionals, indeed, the whole of society, have a special obligation toward patients. Goodin’s (1985, pp.775–785) moral theory is at heart a relational one. His core theory can be summarized as follows:

  1. a)

    The basic articulation of the theory on responsibility is as follows: “A is vulnerable to B, if and only if B’s actions and choices have a great impact on A’s interests.“ (p.779).

Related to a physician strike, this first principle suggests that doctors have a moral responsibility as a strike is a deliberate choice that affects care to patients. Indeed, Goodin would suggest that patients are in “dire need” because they are not able to help themselves and that triggers an obligation to help them.

This principle of moral responsibility is similar to the moral claim (MC) argument (Driver, 2008), where the moral claim, MC is stated as:

MC if an agent A is morally responsible for an event, e, then A performed an action or omission that caused e”. In this case, an individual physician may not have called the strike, but it is the withholding of care in support of the strike that is assumed to be causing the e.

  1. b)

    Goodin (1985) adds that vulnerabilities are relative. “A is more vulnerable to B than C if B’s actions and choices make a greater impact on A’s interests than do C’s actions and interests” (p.779).

For example, doctors may go on strike because the government refuses to give in to their demands, thereby making governments morally responsible (Abbasi, 2014). However, the strike increases a patient’s occurrent vulnerability, and therefore one can argue that the choices of striking doctors impose greater vulnerability on patients than the government’s actions, presently.

  1. c)

    Goodin identifies three forms of responsibility (a) causal responsibility (the agent produces the result), (b) moral responsibility (you are to blame for it), and (c) task responsibility (it’s your job).

In the case of a strike, physicians may be violating the three forms of responsibility when some harm comes to the patient. The job title of a physician confers task responsibility, failure to treat a patient violates moral responsibility and any harm that results from the strike fits the causality rule. While the idea of causality is often held as an important element of moral responsibility (e.g., Driver, 2008), some argue that causality need not be present for a moral claim to apply (Leslie, 1991). Goodin (1985, p.781) argues that moral responsibility still applies even if we determine that someone “deserves” the harm. For example, a doctor cannot fail to treat a patient because they got into an accident while drag racing, an illegal act.

Goodin extends his theory of moral responsibility into a collective one. Goodin (1985) identifies two types of moral responsibility: “disjunctive and conjunctive responsibility”. According to Goodin, “moral agents have disjunctive responsibilities when one particular person within a group is in the best situation to help the vulnerable. While the person best placed is required to render assistance, other people’s responsibilities are not erased but merely reformed. Conjunctive responsibilities occur when a vulnerable person can best be helped when a group of people cooperates to render aid” (pp. 134–136). Goodin’s principle on collective responsibility is consistent with Feinberg’s (1970) theory of collective responsibility.

Goodin’s collective responsibility principle is instructive on the need for multiparty collaboration as a tool for helping the vulnerable. He states:

  1. d)

    “If B is a member of a group that is responsible, under the principle of group responsibility, for protecting A’s interests, then B has a special responsibility: (a) to see to it, so far as he is able, that the group organizes a collective scheme of action such that it protects A’s interests as well as it can, consistently with the group’s other responsibilities, and, (b) to discharge fully and effectively the responsibilities allocated to him under any such scheme that might be organized, in so far as doing so is consistent with his other moral responsibilities, provided the scheme protects A’s interests better than none at all” (p. 139).

Related to physician IA, physician groups and governments have conjunctive responsibility to aid vulnerable patients. Indeed, the principle of conjunctive responsibility mandates collective action on the part of all stakeholders in developing a healthcare system that is good enough that IA is an exception rather than the rule. In developing countries especially, governments are mandated to provide the facilities in hospitals. Considering the unique, tripartite relationship between patients, physicians, and the government (Isangula, 2012), all parties must work collectively to ensure that healthcare policies and resources are in place to prevent IA in the first place. Since vulnerability is a universal expression of the human condition, physicians and government must take joint action to prevent harm by avoiding IA. It is important to note that conjunctive responsibility does not excuse individual physician responsibility to the patient, it is only the nature of responsibility that is assuaged, not dismissed, and Li Y-T and Ng (2021) point to one way such an objective can be fulfilled. The authors suggest that involving the public as stakeholders in negotiations between physicians and governments could help. In making critical decisions on how best to help patients, consequentialists frameworks present important decision-making guidelines for ensuring that actions taken confer the most benefit on society. For example, Kant in his MM notes that we must promote collective social ends as part of our imperfect duties (Kant, 1996, p.5:113), Rawlsian ethics helps set up a social contract in which those who are worse off in society get treated fairly. There are several additional theoretical traditions focused on duty to the vulnerable.

Firstly, all the Abrahamic traditions emphasize duty to the vulnerable. For example, Rosner (1975, 1993) citing Jewish Rabbinical Law argues for the sanctity of a doctor’s moral obligation to patients. He writes: “a cardinal principle of Judaism is that life is of infinite value and clinicians cannot be justified in walking away from their posts… for a physician to strike, for whatever reason, is unconscionable and totally contrary to every standard of medical ethics and morality” (p.37). Rosner’s position is reflective of Judeo-Christian ethics and morality. The exaltation to “Love thy neighbor as thyself” is an indirect comment on cosmopolitan duties that arise out of our shared humanity. Secondly, moral values of showing kindness to others, especially the vulnerable, and having a moral character are at the heart of Islamic concepts of morality (Leaman, 2006). Mehrunisha and Arzoo (2018) note that within an Islamic paradigm the sick have a right to be treated, and those in a position to offer care, be they physicians or the government who may have the resources are duty-bound to provide it. This position is consistent with Goodin’s (1985) individual and collective responsibility. Thirdly, traditional moral philosophies such as African communalism, including the South African doctrine of ubuntu places moral responsibility for care on physicians (Ewuoso, 2021). As Mangaliso et al. (2021) put it, “ubuntu captures the communality of human life and the collective obligation we all have for the preservation of human dignity.” Mawere (2010), notes that “where people share the same idea of personhood and communal life, physician strike is a violation of the public trust-a complete failure to exhibit the prime duty and responsibility to other members of their community” (p.9).

Thirdly, the emerging ethical principles based on European communalism and humanistic traditions emphasize autonomy, dignity, integrity, and vulnerability, all reflective of the human condition as key elements for developing basic principles in bioethics and bio law (Rendtorff, 2002). Although directed to bioethics research, some of these provisions have lessons for physician action. The lesson here is that although all people, including physicians, have the freedom (autonomy) to determine what they want to do, both patients and physicians must be assured of their dignity. It is important to note that these principles following Kantian ideals remind us that people should be treated as an end, not a means only.

Finally, the medical profession itself recognizes the unique position it occupies in society and the imperatives for reducing harm. The Hippocratic Oath, based on Hippocrates’s love of medicine and humanity, to which physicians subscribe, is a testament to that recognition, and the profession’s focus on the vulnerable. Misselbrook (2013) argues that “the ethos of medicine is generally Kantian”, citing the British GMC publication Good Medical Practice as a good example of duty-based guidelines. Heubel and Biller-Andorno, (2005) note that the German Principles on the physicians’ role in patient care, use terms of central importance in Kantian ethics. The American Medical Association (AMA) Ethics Opinion on Patient-Physician Relationship affirms that it is the “physicians’ ethical responsibility to place patients’ welfare above the physician’s self-interest or obligations to others.“ Indeed, the AMA code of ethics, decidedly Kantian, prohibits physicians from using IA as a bargaining tool and urges members to seek policy changes that promote patient welfare “in ways that are not disruptive to patient care” (AMA Code of Ethics, 1.2.10).

Some limits to moral responsibility

Several conditions limit moral responsibility arising from the theoretical frameworks reviewed in this paper. These limiting conditions ensure that an agent is not compelled to always “bear the weight of the world on their shoulders” (Pierlott, 2004). For example, physicians cannot be expected to bear a more than reasonable responsibility for the welfare of patients (Sanchdev, 1986). Importantly, these limitations temper some of the claims placed on agential responsibility and reform their “deserts of blame or praise”.

First, the agent must have the capacity and act freely to deserve blame or praise (Rudy-Hiller, 2018; Björnsson, 2017). An agent who had no control over action is excused according to this condition. In the case of individual physician participation in an IA, it is conceivable that peer pressure may play a role in whether participants are acting freely or not. To the extent undue pressure accounted for participation in an IA, such agents may be excused from moral responsibility. Second, agents must be aware of the action, its consequences, moral significance, and the fact that alternatives existed. An agent must know the claim exists in the first place, as well as whether a more important claim replaces a current one (Goodin, 1985; Björnsson, 2017). The closeness of the agent to the claimant, their mutual dependence, and relevant knowledge (Kant, 1996) increase agential awareness of action. Unless the agent knows that a claim exists to start with, they cannot be morally responsible. Goodin (1985) suggests that an agents’ moral responsibility is further limited if others are willing to come to the aid of the claimant, or there exists another more pressing claim. Of course, agents can use self-interest as an excuse and Goodin seems aware of that possibility. In the case of physicians, it is difficult to argue that these conditions apply. Doctors are aware patients need help, have the capacity, may have the freedom to act, and know non-action results in harm to patients, at least in the short term. Another important epistemic condition for responsibility is an agent must also be aware of the consequences of an action, including awareness of the moral significance (Rudy-Hiller, 2018). Also, knowing that alternative courses of action exist is part of the epistemic conditions for moral responsibility (Zimmerman, 2002). Levy (2011) argues that agents should only be blameworthy or praiseworthy when they are aware that there were other courses of action they could have taken, in other words, an agent could have acted otherwise (Björnsson, 2017).

Finally, Kant’s notion of imperfect duties presents one limit on moral responsibility. Kant argues that while we must fulfill our perfect duties, there is some discretion when it comes to imperfect duties. As Robinson (2019) puts it, Kant seems to be suggesting that fulfilling our moral duties “should not be at the point of personal exhaustion and degradation” (p.123) and that agents must be free to choose contributions that they believe would be of benefit to society. While not exactly utilitarian, this latitude associated with imperfect duties is a limit to the moral responsibility of agents. Using the Nigerian doctor’s strike as an example, the non-payment of salaries imposed severe burdens on the physicians and so to deny them the right to strike would be to the point of “personal exhaustion and degradation” and the doctors may consider that the cost of calling a strike as a tool to change the situation is lower than the benefits from the strike.

The preceding discussion demonstrates that physicians must be aware that a strike can cause harm to patients, that they have other choices besides going on a strike, among others for us to hold them blameworthy. Instead of second-guessing whether physicians meet these conditions, we can use the actions physicians have taken for past strikes to reduce patient suffering as a proxy of their awareness of harm resulting from strikes. The evidence shows that doctors in Israel, the UK, Pakistan, and Australia were careful to limit strikes to what they label “non-emergency cases,” keeping emergency services open during their strikes (Abbasi, 2017). In the case of Israel, Brecher (1985) reports that doctors retained the option of fee for service to ensure that there was some continuity of service during their strike. In the case of the UK, urgent and emergency care were still attended to. While these palliative arrangements may be feasible, even if inadequate in developed countries, such cannot be said of developing countries such as Ghana, Nigeria, and Kenya where private medical facilities are beyond the means of the average citizen. The upshot here is that non-emergency cases may rapidly become emergency cases if not quickly attended to. Counihan (1982) argues against this sort of partial care during strikes, dismissing it as untenable. In their totality, these remedial actions are clear indications that doctors are fully aware of the potential harm that may come to patients because of their strike. Such actions may be well-intentioned, but they may be insufficient, and come no close to the moral responsibility and moral obligation concepts discussed in this paper.

Conclusions

In this paper, we explained the moral responsibility and agency of physicians as it relates to strikes. We applied theoretical pluralism to explore issues at the nexus of vulnerability and responsibility because such an approach may promote the development of more complex solutions. We argue that theoretical tools may offer practical guidelines for resolving some of the intractable contemporary moral issues such as those that arise out of this discussion (see. e.g., Kant, 1995; Schneewind, 1998; White, 2011). For example, the need to balance the interest of doctors, patients and society require multiple frameworks for making evaluative judgments on responsibility, praise, or blame. Making impartial decisions (Rawls, 1999), based on the human condition (Kant, 1995), developing tools for practical wisdom, and an appreciation of the virtues (MacIntyre, 1984), all offer important resources for addressing the contemporary individual and societal issues at the heart of this paper.

We have shown that while physicians have the right to act in their self-interest, their unique position makes them moral agents. The presumption that patients are in a dependent position is consistent with prior research that accepts the unique position of a doctor, viz-a-viz their patients (Beauchamp, 1997; Pilnick and Dingwall, 2011). Beyond individual responsibility, we argued for collective responsibility to the vulnerable (Goodin, 1985; Kant, 1996). Contrary to methodological individualism positions (Smiley, 2017), more recent accounts of responsibility suggest that there are instances where collective entities are best placed to prevent harm (Soares, 2003; Kant, 1995 Schneewind, 1998), thereby placing responsibility for patient care on both physicians, the government and society. We argued that in addition to developing practical and moral wisdom to help decision-making, physicians may need to develop virtues such as humility. For example, Keenan (2020) suggests that adopting an ethics of vulnerability based on the Imago Dei tradition can “become a foundation for a virtue ethics that orients us toward the right responsiveness to contemporary challenges” (p.56).

We suggest, based on utility maximizing views, that some strikes may be justified if their outcomes are long-term improvements in healthcare facilities that serve the majority. Indeed, in certain cases, such actions may be deemed to be for the greatest good (summum bonum). This may especially be the case in developing countries, given the often-dire state of medical care in those regions (Waithaka, Kagwanja, and Nzinga et al. 2020). Physicians, especially those in developing countries and those that work for public institutions often operate in resource-starved environments and face the moral dilemma of either protecting vulnerable patients or using strikes as a tool for securing better facilities. At the same time, we still need to be mindful of using the end-means philosophy as Kant reminds us that we have a “perfect duty not to use others as a mere means” (1996, p. 4:430). Indeed, as Weil et al. (2013) note, physician strikes need not be inevitable, and exploring other strategies, may be an option.

The study contributes to the emerging literature on moral responsibility, vulnerability, and obligations, particularly as it applies to the use of IA in the medical profession. In focusing on theories of moral responsibility, vulnerability, and their implications for moral agency, this study provides a more in-depth explanation of the issues, an improvement on the often atheoretical nature of some prior studies. Huddle (2005) notes that clinical practice is inextricably tied to moral values. Physicians face moral and value-laden situations all the time and therefore, training medical students in moral decision making, including sensitizing them on moral responsibility and obligations should prove useful. Thomas (2015) acknowledges that while there is the recognition of the need to include moral education in physician training, very little of that is done presently and the need to train medical students in moral development has been emphasized (Branch, 2000). Lyon (2021) argues that teaching virtue ethics in medical school will positively impact the moral character of physicians. More importantly, moral training should attenuate opportunities for moral disengagement (Bandura, 2016), a situation in which people minimize the extent to which they understand how much harm their actions cost others. The findings from this paper point to the need for moral education in the training of physicians as that may have a positive impact on clinical practice and ethical decision-making as practical wisdom (phronesis) promotes excellence in deliberation (MacIntyre, 1984; 1991). The current training of physicians remains largely instrumental for good reason. Teaching virtues, including the tools for acquiring them (phronesis), should help decision-making when it comes to such intractable issues as choosing between physician and patient welfare. There is some evidence that people can acquire moral wisdom through teaching and education (MacIntyre, 1984).

As is often the case, this study is not without its limitations. This is a conceptual paper, and its framework may be subject to refinement, and extension. The fact is this paper does not by any means resolve the many intractable issues often associated with moral responsibility and obligations and there is a lot more that can be said about moral responsibility and agency, and we call for additional studies on the issues. For example, it would be insightful to know what physicians think of moral responsibility. How we balance a physician’s right to strike against a patient’s right to care, perhaps using dialectical thinking, is another area worthy of further exploration.