Liberalism’s Problem of Strongly Pro-Social Work

I. A LIBERAL PARADIGM OF OCCUPATIONAL CHOICE

The target of my argument is a view on occupational choice that we can define as paradigmatic for political liberalism. According to this view, occupational choice is something that everyone is free to make on the basis of their own values and preferences, as long as those preferences do not violate others’ rights. Douglas MacKay formulates such a view based on Rawlsian principles, stating that (1) “freedom of occupational choice is a basic liberty” which cannot be infringed by reference to the common good, utility or perfectionist values, and (2) “citizens are permitted, as a matter of justice, to accept or decline offers of employment on the basis of considerations stemming from their values and preferences.”¹ Permissible personal values and preferences must be the basis of occupational choice because this is necessary for citizens to realize “the full exercise of their capacity for a conception of the good.”²

Before attacking freedom of occupational choice and claiming a profound problem for liberalism, we should first consider alternative interpretations of the right to freedom of occupational choice and evaluate modifications that aim to stay within a liberal framework. Distinct from the paradigmatically liberal view, there are views that entail justice-constraints on occupational choice. G.A. Cohen or Lukas Stanczyk hold such views and deny that this makes them incompatible with liberalism. Cohen rejects state coercion but emphasizes a moral duty to work at the best of one’s abilities in the socially most productive occupation, implemented in a socialist egalitarian ethos.³ Stanczyk also rejects direct coercion, but stresses that freedom of occupation does not entail a right to get whatever set of options one wants. As opposed to that, respecting such a right is compatible with the state designing the available options in a way that makes them socially productive.⁴

To distinguish them from the paradigmatic liberal view, we can refer to those proposals as socialist liberal views. Especially Cohen’s ethos has provoked controversy regarding its compatibility with liberal commitments. The most basic point of departure from what I call the paradigmatic liberal view is probably the expansion to personal conduct of the principles of justice, which the liberal paradigm depicts as principles governing institutions, not individuals.⁵ Cohen presents two arguments to defend against illiberalism charges when introducing principles of justice into an ethos governing individual choices. First, Cohen softens the subordination of individuals to social productivity by granting an “agent-centred prerogative.”⁶ Second, Cohen claims that “morally inspired motivation” is not state coercion and therefore not in conflict with liberty.⁷ His assumptions are controversial.

I think that Paula Casal is right in arguing that Cohen “needs to choose between the claim that an ethos cannot conflict with liberty and the claim that an ethos can conflict with liberty and indeed become so influential and oppressive that it must be included in the basic structure and subjected to the principles of justice.”⁸ She points out that Cohen himself accepts that for a sexist ethos; so he cannot uphold the claim that only legal norms are coercive, while ethical norms and expectations and their social sanctions are innocent from a liberal point of view. The crucial question is whether the liberal state is allowed to actively cultivate a few distinctive values to substantially guide people, without violating the condition that it must not prioritize any specific conception of the good.

Rawls provides an argument for a distinctive set of values and virtues a liberal state can permissibly cultivate. As a liberal, he accepts the distinction between the right (that is universal and enforceable) and the good (that is particular and needs protection from state regulation and coercion). However, he reflects on the social value of individual ethical choices. In the ideal theory of the well-ordered society developed in A Theory of Justice, Rawls considers it good in itself to have different conceptions of the good within a society.⁹ Realizing one’s own conception of the good is essential for self-respect, which is the probably most important primary good. Rawls, however, does not assume that realizing one’s own conception of the good is a purely individualistic, self-oriented pursuit that stands in tension with social productivity and responsibility. On the contrary, Rawls introduces an Aristotelean principle, implying that the good for one person is usually also good for others and includes a reference to the common good.¹⁰ Moreover, Rawls seems to imply that the value of different conceptions of the good is ultimately derived from the fact that persons have a variety of talents and abilities, and that society as a whole benefits from a universal freedom to develop one’s inclinations and hone one’s own specific abilities.¹¹

In Political Liberalism, by contrast, Rawls pays attention to the fact that this ideal picture is not represented in real-world societies. Usually, the state needs to actively foster personal qualities to ensure that citizens are working together under the principles of justice. Rawls remains clear that the state may not prioritize any specific comprehensive doctrine in ethics, religion or philosophy, but he argues for a distinctively political conception of justice that citizens must internalize and embrace.¹² The state may actively cultivate such a conception. The conception need not be a mere modus vivendi. It can be both “deep” and “wide” and represent an overlapping consensus of all reasonable comprehensive doctrines.¹³ Thus, the political conception is part of every comprehensive doctrine, but it does not follow from their other, non-political parts.¹⁴

Rawls admits that the common term of state “neutrality” can be unfortunate in this context.¹⁵ He claims that “justice as fairness is not procedurally neutral,” because with a political conception a society seeks a common ground. It may be neutral, though, “in terms of the aims of basic institutions and public policy with respect to comprehensive doctrines and their associated conceptions of the good.”¹⁶ It may, in distinction from that, ensure the good of a political society that is dependent on citizens cultivating political virtues.¹⁷ He aligns his liberal brand of a political society with the ideal of classical republicanism and distinguishes it from the idea of civic humanism.¹⁸ The former cultivates the political as an inevitable part of life that depends on a functioning constitutional regime, but not as a comprehensive doctrine; the latter makes the political life as a citizen “the privileged locus of a good life.”¹⁹ A liberal state can thus cultivate a sense of responsibility for necessary goods without violating liberal commitments, as long as it does not privilege a culture in which care for the public good is the most essential part of life.

Now, the question is whether a socialist liberal view on occupational choice can be defended within the limits of such an argument or whether it involves a more fundamental break with liberalism. Cohen’s ethos, in a nutshell, requires citizens to work to the best of their abilities in the socially most productive profession without additional pay. This, it seems, requires a rather strong idealistic commitment to advancing the good of society in more or less all one’s daily efforts. Whether any version of such an ethos could be defensible for a liberal is a question to which I return in the end.

I will now proceed to an investigation of the nature of what I call strongly pro-social motivation and suggest that it requires an approach to life and work that is far from universal in society. The non-ideal conditions of many contemporary societies differ significantly from the ideal picture that Rawls draws in A Theory of Justice, where he assumes a social element built into individual self-development. Under present conditions, self-realization and social utility do not necessarily coincide. Sometimes, the two values might even lead to incompatible and contradicting paths of life. Many people in contemporary societies experience that they need to decide whether they prioritize social utility and service or their personal excellence and career development. As I will illustrate in more detail in section III, modern societies are increasingly polarized with respect to the difference between individualistic professional self-realization and social utility or service-oriented conceptions of occupational choice, in which problematic class and gender differences manifest. I argue that there is a deep rational and motivational difference between the more individualistic layer of society and the more socially oriented one. While some of the injustices relating to those differences are due to socioeconomic differences and can be remedied by more distributive justice, a distinctive type of wrong will remain.

II. THE RATIONAL AND MOTIVATIONAL STRUCTURE OF STRONGLY PRO-SOCIAL CHOICE

The liberal paradigm outlined above assumes that everyone is free to base their occupational decisions on a wide range of personal preferences and values – other-regarding preferences or social values being only one possible type of preferences a person can have. In this section, I argue that such a picture of human choice misses a crucial difference between choices that relate to individual preferences, goals and values, and choices that do not.

I want to introduce a distinction between pro-social and strongly pro-social choice in order to prevent the objection that individual goals need not be self-regarding, let alone egocentric. Pro-social choice may be based on other-regarding preferences, social values, or group goals that one identifies with as an expression of one’s individual worldview. Distinct from that, however, there can be strongly pro-social choice, which I define as involving risk and sacrifice that cannot be rationalized in such a way. Instead, there are situations in which an entirely different kind of reasoning is activated. While this kind of reasoning is especially relevant in times of war and catastrophe, the number of mundane situations in which it also occurs might be larger than we realize. Let me begin to introduce the difference in reasoning by considering the following choice situation:

Imagine you are offered the opportunity to pick one of the following three options:

1. You can stay home to care for a sick and dependent elderly family member.

2. You can withdraw in a solitary cabin to write a book.

3. You can go to the cinema.

Which of those lots would you prefer?

I assume that many people will find this choice situation somewhat weird or artificial. Typically, such choices do not present themselves in the way of picking a lot. While many people would be clear about not wanting to let down a dependent family member in need of care (option 1) just to go to the cinema (option 3), they would nonetheless be hesitant to say that they prefer the lot of staying home to the exclusion of all the others. By contrast, many who are asked which of the options they prefer would probably say: “It depends. I take the first if it is necessary but I would prefer another.”

This suggests that people do not always choose according to their own preference but, rather, sometimes see that their preferences are irrelevant in a particular choice situation. You might object that people choosing the first over the third option in real-life situations just act in accordance with a deeper or weightier preference – that they reveal a preference for realizing the personal value of caring for their families. That this explanation is often too easy can be seen in a case involving the weighing of the first option against the second. People who deeply care for writing a book, as part of a personally significant scientific or literary life-project, might experience such a choice situation as tragic – or they might not experience it as a choice situation at all, just feeling necessitated to stay with their family whatever the cost of personal value for them.

Nevertheless, accepting the first option because it is necessary, not because you prefer it, does not mean that you cannot genuinely be motivated by your family member’s wellbeing. Instead, we should conclude that, sometimes, such choices are not even framed as a choice of whether you prefer your family’s wellbeing to something else. Rather, the family context motivates a choice in which individual preferences, goals and values are simply conceived as irrelevant.

Cases like this suggest that the reality of human agency is not exhaustively described as the realization of individual preferences, goals and values. The theoretical options of redescribing the relevant preferences as regarding other people or a deeper social value often seem ad hoc. A better description of the phenomenology and normativity of such cases is that humans do not always rationalize their actions in terms of their own preferences, goals and values. Sometimes they reason differently.

Raimo Tuomela distinguishes between two basic kinds of thinking that are not reducible to each other: I-mode thinking and we-mode thinking.²⁰ Tuomela does not postulate supra-individual agents; indeed, he emphasizes that “the we-mode is in the minds and actions of individuals.” Still, he claims that “states of we-thinking and I-thinking are ontologically different kinds of states.”²¹ In proper we-mode action, an individual person is “functioning as a group member,” while in I-mode thinking and acting, the person is functioning as a private person.²² In line with my proposal of distinguishing strongly pro-social choice from mere pro-social choice, Tuomela also holds that we can be “privately social,” realizing our individual goals and values in a group context or in relation to a group.²³ It is possible that we thereby benefit the group or act for the group. However, the reasoning differs from the proper we-mode case. While I-mode group action is guided by individual goals and values (to benefit the group), proper we-mode action is guided by a group ethos that is constituted by collective “goals, values, beliefs, norms and standards.”²⁴ We-mode action must satisfy what Tuomela calls the Collectivity Condition, according to which “the goal [of the action] is satisfied for a member if and only if it is satisfied for all other members.”²⁵ The group is united by a collective commitment, which “glues” them together around an ethos and gives “joint authority to the group members to pursue ethos-related actions.”²⁶ The group ethos, rather than the person’s private goals and values, provides the motivational reasons for action. It also constitutes the normative framework for evaluating the action:

A publicly we-committed member who leaves the joint project or intentionally violates the ethos can be criticized by the others, whereas if he had been committed only “to himself” he would in general have been less criticisable socially for letting the others down. In the I-mode case, he can be criticized for causing harm to others and for not being a rationally stable person.²⁷

Tuomela also assumes that people are “disposed more easily to give up … in the I-mode case than in the we-mode case” because of social pressure. ²⁸ This is explained by the fact that, in proper we-mode agency, the person’s reasoning does not refer to individual goals and values but brackets them for the group ethos. This does not mean that people cannot act in the I-mode and in the we-mode at the same time. Tuomela holds that they might even do so on the same occasion.²⁹ However, it means that the I-perspective is not necessary for rationalizing and motivating action and that it can very well be bracketed or overridden.³⁰

Hans Bernhard Schmid has connected Tuomela’s work with situational psychology to argue that we switch between different modes of reasoning according to situational cues.³¹ Schmid argues that social situations can be designed in a way that makes people identify them as situations calling for we-mode reasoning and making I-mode reasoning inappropriate. Consequently, people set aside their individual judgments and preferences, at a personal cost to themselves. When people identify a situation as calling for we-mode instead of I-mode reasoning, they do not conceive of their own attitudes as normatively significant and instead choose in accordance with collective requirements.³²

I think that the distinction between I-mode and we-mode reasoning provides a good explanation for why the choice situation I described at the beginning of this section seems so artificial to us. The idea of picking one out of the three options just does not capture how choice situations involving significant social sacrifices, such as setting back the personal value of writing a book for a dependent family member, are typically structured. The idea that we pick from options that are presented to us as basically on a par fails to capture the fact that the dire needs of a family member can activate an entirely different mode of reasoning. We do not, or at least need not, deliberate about what we value the most; instead we simply respond to a social necessity. We simply act in accordance with a social requirement, without feeling any need to go back to our personal attitudes to rationalize and motivate what we are doing.³³

In other words, we respond as family members, or in the family-mode, when we encounter the urgent need of our dependent elderly relative, while we choose according to our own values and preferences, in the I-mode, if we are free to do so.³⁴

III. EMPIRICAL FINDINGS ABOUT OCCUPATIONAL CHOICE

Now, we might wonder whether the phenomenon of acting in the we-mode has any bearing on the reality of occupational choice. We might object that choosing a professional path is unlike a situation in which we have to set aside our own preferences, goals and values for people in need. Understanding occupational choices in that way, however, reflects an idealized liberal understanding of occupational choice that does not match empirical reality.

In many societies, there are noteworthy class differences in the likelihood of picking socially useful or helping professions, instead of paths characterized by extensive opportunities for individual self-realization. While there are certainly multiple causes of such differences, many psychologists deny that they are driven simply by the lack of financial opportunities. In the US, for example, social psychologist Nicole Stephens and her colleagues have established a fundamental cultural difference between social classes: the upper classes have a culture of independence, while the lower classes have a culture of interdependence.³⁵

In cultures of independence, children are typically understood as projects of their parents, in which parents invest a significant amount of money and resources to grant them the most successful future. Children growing up in a culture of interdependence, by contrast, are typically understood as helpers and supporters of their family. Not only do they receive fewer resources for themselves in their youth. Often, they are also expected to give more resources back to their parents during their entire lives.

While these differences have an undeniable connection to wealth and resources, the cultural differences also have a reinforcing effect and prevent alternative choices, even if financial support is provided from the outside. This is, for example, illustrated in Jennifer Morton’s philosophical work about the ethical costs of upward mobility – work that is informed by empirical evidence and personal interviews with students from disadvantaged backgrounds in the US higher education system.³⁶ Morton points out that disadvantaged students are still more likely to drop out of higher education, or decide to work in care and community service instead of pursuing higher education, even if funding and other support is provided. Morton’s interviewees describe not only financial and practical problems but also personal struggles that Morton describes as genuinely ethical in nature.

Morton’s typical interviewee is drawn away from thesis work and studying because she is needed by family members. Even if she is no longer needed by her own family, she still feels a strong pull to give back to society and help people, which she feels she cannot do equally well were she to pursue a career at university or in higher administration.

Now, should we conclude that if financial support and opportunity is given, this choice is unproblematic? Should we conclude that, in the absence of unjust coercion, someone prioritizing concrete helping professions over more individualistic career paths is just the expression of a pro-social preference that is stronger than that person’s individualistic preferences?³⁷

Indeed, it seems straightforward to assume that differences in the strength of pro-social preferences are, at least ideally, a matter of different personality types that lead to different life choices. The liberal rationale for differential occupational choices sketched above presupposes that whether you choose a more social or a more individualistic path in life just depends on what type of psychology or motivational set you have. According to Rawls’s idealized understanding discussed in section I, it is good for the whole society to have both individualists and pro-social people, just so long as nobody is unfairly excluded from or coerced into one or the other way of life.

Morton’s cases, however, suggest that this is not how many students from working class or immigrant backgrounds perceive what pulls them away from academic pursuits. It is not that they just discover that they care more for helping others, so higher education is not for them.³⁸ Some of Morton’s interviewees were talented and passionate about the subject they were studying and had a strong interest in academic pursuits. Nevertheless, they felt that there is so much urgent need elsewhere that they must set aside their own dreams about academic or artistic excellence and self-realization. Most of them experience this as a struggle and an injustice, however, once they see their more privileged fellow students not facing the same struggles and seeming to be fine with mainly pursuing their own interests and goals.

This phenomenon, I suggest, is evidence that there is such a thing as strongly pro-social occupational choice – an occupational choice that people make in response to a perceived social need, irrespective of their own preferences, goals and values. In such a case, the choice is an immediate response to social necessity that is not in need of being shoed up by reference to personal preferences or goals. In such cases, people respond as fellow citizens, as fellow human beings, as family or community members, supporting the common good or a group ethos, rather than acting as individuals who pursue their own goals and express their personal values and preferences.

Those culture, class and also gender differences in occupational choice are evidence that the conceptual distinction between we-mode and I-mode reasoning is relevant also in occupational choice. Even if we bracket the cases in which the lack of money or encouragement forces people into jobs that are considered more immediately socially useful, I think we can still find many cases where people make such choices on an entirely voluntary basis, setting back their own preferences, goals and values for a common good. Such a choice is inadequately described if it is understood as structurally on a par with the decision to pursue, for example, writing, music or art, which might be valuable but less immediately needed by others.

The description fails to capture that the strongly pro-social agent sets aside their individual preferences, goals and values for a collective good. In the following, I will explore why this failure is a normative problem. I will argue that a state must take a stance, instead of allowing people to suppose that individualistic occupational choices are entirely on a par with and as permissible as strongly pro-social choices.

IV. A DISTINCTIVE WRONG TO STRONGLY PRO-SOCIAL AGENTS

Freedom of occupational choice along the lines of the liberal paradigm pays no attention to the deep ontological difference between I-mode and we-mode reasoning. I argue that a just society must explicitly decide whether it promotes occupational choice as an I-mode decision or whether it expects occupational choice to be based on we-mode reasoning. Failing to draw this distinction and just allowing citizens to fall into different modes of reasoning depending on their situation, psychology or moral conviction commits a distinctive wrong to pro-social agents that I will now explore.

What could possibly be the problem with a society that stratifies into one group of people who make their strongest personal preference a basis for occupational choice and another group of people who respond to (perceived) social necessity? I think the main problem is that the response of the latter type only makes sense if the universal ethos of how you make occupational choices – of what that type of choice is essentially about – is characterized by the strongly pro-social rationale. This means that it is a universal expectation and a common understanding that picking an occupation is about taking a share in the tasks that need to be done and about participating in the provision of the common good. It is not, in the first place, about best realizing your personal values and preferences – even if you respect a few social constraints in that pursuit.

A liberal society which is not unified by an ethos that is actively guiding – not merely constraining – occupational choice offers citizens a range of options, some more pro-social and others more individualistic, and leaves them free to opt for one or the other. As suggested in section III, some people experience a pull away from their individually preferred occupational pursuits and feel the necessity of picking occupations with higher social utility. Other people, by contrast, feel encouraged to pursue their individual goals in their careers and do not experience the same distracting normative tug.

Let me first distinguish between four main types of reasons that may instigate strongly pro-social choice and examine how each can be related to injustice. First, and most obviously, there are economic reasons, and related socioeconomic access problems, for some people having less choice in selecting a career according to their own preference. This part of the problem can at least in principle be remedied by redistribution. If provided with economic stability, those people will immediately become more active in pursuing their personal preferences and values.

Second, there may be what I call positional reasons for being a strongly pro-social agent. Because of their location, surroundings and family setting, those people are more acquainted with social need and feel more guilty than others if they do not make their life primarily about taking responsibility for need. The likelihood of being confronted with need and the urgency to do something about it is also correlated with socioeconomic background in the actual world. In poor neighbourhoods, problems are more visible. Richer families are better able than poor families to rely on paid professionals for meeting their needs. However, redistribution will not rule out positional reasons for pro-social choice propensity entirely. Even under economic equality, some people will be more acquainted with hardship and need arising from frailty, sickness, disability and the like.

Third, there can also be personality reasons for some being more responsive to social need and tending to feel guilty if they do not devote their life to providing for a community. This factor is not always easy to distinguish from the fourth type of reasons, which I call reasons of moral conviction. Some people have personal ethical views that make social responsibility and contribution one of the most central – or even the central – aspect of life irrespective of their personal inclinations. Other people take their life to be primarily about realizing more individualistic values or achieving excellence in a field that is not directly in the service of human needs. While moral or religious convictions of that kind may correlate with personality traits as well as with economic factors, they are nonetheless a separate factor influencing strongly pro-social choice.

For the state to promote occupational choice as an I-mode project, it would have to work towards abolishing all four types of factors. We may say it is enough to rule out the economic reasons, and maybe provide some alleviation for the positional reasons, while reasons of personality and moral conviction are fully permissible grounds for a choice. Indeed, we may expect that we-mode reasoning as a phenomenon slowly disappears when a society is sufficiently affluent and equal. The remaining pro-social agents would then be I-mode reasoners who simply judge for themselves that social issues are what they personally care most about – exactly like the individualists who care most about their respective projects.

If the liberal state achieves this, and it works both structurally and on the level of individual psychology, there is in fact no problem. However, it is an idealistic assumption that societies – even if they are affluent and developed – can in fact flourish entirely without people capable of strongly pro-social motivation. As a matter of fact, for many citizens of OECD countries, the large amount of essential labour that most people would not prefer is rendered invisible by outsourcing it – or leaving it to poor immigrants who are, effectively, outside the social contract of the domestic liberal society and driven by sheer economic necessity or, often, desperation. Stopping relying on that exploitative source of labour, especially relevant in primary health care and food production, would very likely lead to unbearable costs and ultimately a breakdown of supply. This makes the success of a liberal association of I-mode reasoners, living and working together as equals, even more doubtful.³⁹

Moreover, it is an open question whether it is ethically desirable to abolish the phenomenon of social necessity to which one simply must respond irrespective of one’s personal preferences. Embeddedness in structures of necessity, and the corresponding mindset, that is common in small and close-knit productive communities but eroding in large modern societies may be of separate psychological or ethical value. Relying on some hyper-idealistic people to fulfil the professional roles that benefit from such a mindset, while allowing others to opt out, can be ethically detrimental even if it is functional and economically feasible.

I conclude that a productive community entirely based on I-mode reasoning is difficult if not impossible to achieve, even if it were an attractive ideal given affluence and technological progress.

Now, why should we think it is a problem when some people reason in the we-mode when making their occupational choice and others not, even in the absence of economic restrictions on choice? The main source of a sense of injustice seems to be the experience that one is necessitated into we-mode reasoning oneself, while others are permitted to reason in the I-mode instead. For oneself, the two modes are not conceived as fungible options in an individual deliberation. Instead, individual deliberation is restricted and bracketed and comes up only in the free spaces that the group ethos leaves. The nature of ethos-guided we-mode action entails that it can only work if it is collectively accepted, and the shared responsibility for the collective goals is universally accepted within the group. If the group in question is the population of a state as a productive community, the normative and motivational nature of the we-mode requires that everybody recognizes the priority of socially-necessary works, such as primary care and essential practical service, over more optional pursuits, such as for example creative, scholarly or athletic goals.

There is no need to deny that the latter goals are worthwhile and useful to pursue, even for wider society. However, there is an undeniable hierarchy between those two types of pursuits. In a small, close-knit subsistence community, this is easier to see.⁴⁰ There explaining to a community member who is seriously struggling with the common necessities that you do your share by engaging in ambitious scholarly work and need not contribute is not very convincing. Even if coincidence, serendipity or derivative value cause social benefits in the long run, the normative pull of more urgent, immediate needs is a pull on all community members, who are – as Tuomela formulates it – “in the same boat.”⁴¹ For the productive process to be a full-fledged we-mode action, the goal (the common good) and the distributive structure of the whole action (which actions relate to the common good in which way and in which priority order) must ideally be known and shared by everyone. In the end, I will turn to the obvious difficulty to conceptualize the productive process of a large-scale modern society as a we-mode action in this sense.

Now, I will ask whether a sense of injustice on behalf of some agents is warranted if the productive process fails to be a proper we-mode action. In a Rawlsian well-ordered society, it seems to be taken for granted that everybody’s reasonable life plan is fit for being an equal part of a just society – that, for example, the creative scholar and the essential health care worker could both pursue their tasks as we-mode goals, relating to the common ethos of a just society.⁴² In actual liberal societies, however, workers or students who are too prone to we-mode thinking – frequent in lower social classes – are more likely to drop out of ambitious scholarly or creative projects, while those succeeding in them are typically stronger in I-mode reasoning, and more immune to the pulls of we-mode agency, not only for economic or positional reasons, but also for reasons of personality or moral conviction. The result is a social layer characterized by we-mode reasoning, and a layer characterized by I-mode reasoning. The hierarchy between such layers is of course worse if it is also accompanied by economic inequalities, but it is also problematic if it is a result of personality or moral conviction.

I suggest that leaving the necessities of the common life to those who personally care for them amounts to recognitional disrespect of pro-social agents in their function as pro-social agents. The strongly pro-social health care worker, for example, may accept a burdensome job with not only a significant risk to her own mental and physical health but also a risk for moral injury resulting from compromising one’s own moral integrity in case of professional failure or error. A society that respects both we-mode and I-mode reasoning as a basis of occupational choice may support and respect the health care worker fully as an individual person – in the same way in which it respects and supports the scholar as an individual person.⁴³ However, the crucial fact is that the health care worker, if she is like one of Morton’s academic dropouts, did not really act as an individual person but as a community member. In fact, she regrets not being in the shoes of the scholar, for whose job she would have been equally talented. She would indeed have preferred the scholar’s path, if the social need she was exposed to and responsive to hadn’t pulled her away. If a person comes from a family and class background in which it has always been self-evident that you subordinate yourself to social need and necessity, she feels betrayed if she learns that this is not how every citizen thinks – that there are classes of citizens in which families educate and prepare their children to best realize their individual goals and values.

This sense of betrayal, I think, is not only fuelled by socioeconomic contingencies but also has a justified basis in the structure of we-mode action as opposed to I-mode action. If the state fails to recognize those different modes and establish a robust collective ethos to guide we-mode action, it accepts sacrificial and risky services from people who only provide them because they believe that it is required by a commonly shared ethos; and it lets the risks and burdens fall entirely on them as private people, who are depicted as having had the freedom to choose another life instead. The strongly pro-social agent acts, as it were, in the belief that her we-mode attitude is rationalized by a shared ethos, while on this depiction she is in fact on an individual project. In its failure to expect everyone to choose their occupation guided by a social ethos, and thus rule out entirely individualistic career paths, the state communicates that there is no shared ethos; it communicates that the pro-social requirements are only normative for those who choose as their individual project pro-social agency instead of some other project with a more individualistic content. To be more transparent about its implicit terms, a liberal society would actually have to communicate explicitly that occupational choice is an I-mode project and make sure that everybody understands this. The potential pro-social agent should be told: “Don’t pick a pro-social profession unless you are really sure that you have no other preferences, dreams or interests that you would rather turn to if there was no social need!” If that is not explicit, it amounts to betraying pro-social agents.⁴⁴

I already suggested that a society would not be functional, and probably not attractive, if it discourages responsibility and sacrifice in such a blunt way and actively promotes I-mode decisions. It may be functional though, if it does not actively take a stance at all and allows for individuals to work in different modes for different reasons. But this is failing to recognize the special nature of pro-social agency, which can only be respected if the state actively takes a stance. The state must actively cultivate the view that occupational choice is not only constrained by social factors but is the very locus of taking social responsibility and contributing to social need. If the state fails in that expressive task, the strongly pro-social agent, who might have been completely happy with her life if she saw that her efforts and sacrifices are basically what life in society is like, may feel that she would have chosen differently had she known what the type of choice in question represents for everyone except herself – namely a locus of individual self-realization, not a locus of making your contribution.

V. POSSIBLE SOLUTIONS

I have argued that a state failing to draw the distinction between agents choosing their occupations in the we-mode and agents choosing them in the I-mode wrongs agents with strongly pro-social motivation. I also argued that unless a society can and wants to flourish entirely as an assembly of I-mode reasoners and explicitly communicates this, a society should ensure a collective ethos that makes occupational choice about making best use of one’s capacities for advancing the common good. This, I indicated in section I, is Cohen’s egalitarian ethos in a nutshell. In their interpretations of Rawls, both Tuomela and Cohen claim that the very conception of justice as fairness should serve as a collective ethos that members of the well-ordered society embrace in their personal conduct.⁴⁵ This presupposes that occupational choice is not a purely private matter but is genuinely done in the we-mode. As an ideal theory, the theory of justice as fairness seems to work with the assumption that all occupational choices add up to the common good so that, at least in the well-ordered society, every individual choice deserves equal respect. But as I have already said, this does not reflect the actual professional world and the reality of occupational choice, as the social value and urgency of jobs differs widely.

What can the state do to promote a universal social ethos of occupational choice? I can identify two main strategies. First, we could advocate for social services that are mandatory for every citizen, as Debra Satz does. As in my own criticism of polarizing societies, Satz deplores the “unmooring of elites from the majority of other citizens” in contemporary societies.⁴⁶ She suggests a mandatory public service requirement in order to create a new type of “robust common experience.”⁴⁷ Such an experience does no longer exist in polarized societies and is crucial for building a “democratic ‘we’” unified by a “solidaristic ethos.”⁴⁸ This is important for sustaining trust in democracy and pro-social motivation. Satz regards the service as a “price of admission” into a democratic society that grants us the status as free and equal rights-holders.⁴⁹ As such, it is compatible with Rawlsian political liberalism, because it is justified as maintaining the good of a political society and does not impose a particular comprehensive doctrine about the good life.

While each concrete policy proposal must be evaluated in light of empirical evidence and political contingencies of time and place, I think that a scheme of mandatory service could in general be a possible step towards solving the problem of strongly pro-social work.⁵⁰ At least a few basic tasks that are important for the common good will be identified and declared as being everyone’s responsibility.⁵¹ If this is ensured, making further occupational choices in an I-mode manner will be less problematic, or even less likely. The obligation to do the work oneself has clear advantages over the proposal that citizens can fulfil their social responsibility just by paying taxes, which can be detrimental to the social ethos because it makes letting down the community an option that can be bought with money.

From a liberal perspective, a public service requirement is least problematic if it extends over a limited time, such as for example one year of life. Though this can have some positive ethos-building effects, it is doubtful that one year of mandatory public service is enough to rule out the problem of strongly pro-social work completely. The problem arises because after that one year the workforce then bifurcates into occupations that encourage, cultivate and reward I-mode reasoning and occupations that are filled only because enough people take their personal preferences and values to be less relevant than on-going social need (whether because of economic constraints or a working-class-typical psychology or conviction).

A better solution, I suggest, would involve an active shaping of job bundles that do not allow for an accumulation of self-realizing pursuits, on the one hand, and reproductive or service-oriented tasks, on the other. Proposals along those lines have been put forth by Casal and Stanczyk. Casal suggests that the state should rule out options that deprive people of their occupational autonomy or restrict their options by leaving job options with less autonomy to the less talented, low-skilled or those vulnerable because of their gender. She criticizes Cohen’s ethos for not recognizing the distinct value of occupational autonomy and instead formulates an “ethos designed to secure occupational autonomy for all.”⁵² Stanczyk argues that “welfare-improving restrictions on the basic liberties of citizens” must be possible, if they are needed for the state to secure important public goods such as “police protection, education, medical care, legal representation, to name only a few.”⁵³ As opposed to mandatorily assigning jobs, Stanczyk advocates making education dependent on a duty to serve in underserved areas and a (temporary) prohibition on moving to privileged areas offering higher pay.

Both proposals promote the design of occupational bundles that are more responsive to social utility and the public good than the privileged career paths in most current liberal societies. Such measures can help creating a sense that learning a profession or taking up a job is genuinely about taking up a responsibility within the joint endeavour of securing the public good. Finding one’s personal good and value on the way may be good too, but it should be secondary to the question of how to best use one’s talents for the public good. It is only in such an ethical culture that the sacrifice and risk-taking of strongly pro-social agents is fully rationalized, recognized and respected.

VI. OBJECTIONS AND TENSIONS WITH LIBERALISM

Respect for strongly pro-social agency, I argued, can only be secured in a productive community unified by a social ethos that requires everyone to work with the best of their abilities to best serve the common good. This reassesses occupational liberty as most of us know it. The crucial question is, however, if a liberal state can find justified interventions or if it should rather accept to some extent of the problem of strongly pro-social work.

Indeed, the environment in which the distinctive we-mode reasoning behind strongly pro-social agency flourishes and is most reasonable are small, close-knit subsistence communities in which everyone’s full subordination to basic need is required and the needs are so basic that they are rather obvious and uncontroversial. Modern affluent societies, by contrast, set free a lot of productive capacities and at the same time create new needs that some regard as less urgent than others. The social ethos necessary for avoiding the problem, however, benefits from a clear prioritization of tasks and common knowledge of their relative relevance for the common good. Political liberalism has less of a problem in justifying moderation of occupational choice in areas in which it is necessary to secure the uncontroversially basic functioning of health care, political order or defence. It is more difficult to defend a detailed rule that seems required to ensure that no social group will concentrate their productive efforts on what most others find irrelevant to the common good. In such cases, we-mode agency fails and the strongly pro-social agents who relied on it can fall between the cracks. That will also happen if the strongly pro-social agents themselves live with the moral conviction that their sacrifice and social concern was right from a moral perspective. For this reason, everyone individually following their own moral conviction does not guarantee the prevention of the problem. This would only work if all were unified in their moral conviction and ground a factual collective action scheme on it.

This brings me to an additional reason for why we-mode agency itself is highly ambivalent: the overall goal and action scheme has no guarantee of being morally right. It can be outright criminal, racist, nationalist or otherwise flawed and parochial. If we are highly optimistic about technological and political progress, we may even hope that we-mode agency in the realm of occupational choice (and most other large-scale existential decisions) is going to die out and makes space for I-mode reasoners regulated by liberal principles. Still, a society not at all dependent on strongly pro-social agents is so far not in sight – or, as indicated above, it will depend on exploiting outsiders to the domestic social contract instead. As long as strongly pro-social agents exist and play a support role in the productive process, they are wronged by a failure to ensure that the productive process is conceived as joint agency, not only as an assemblage of individual activities regulated to ensure that all needs are met.

If we reach a point at which strongly pro-social agency does not play any crucial role in society, we may in fact conclude that the strongly pro-social choice propensity of some is nothing but a maladaptive psychological relic that should not be cultivated and passed on to their children. If some still exercise that psychological disposition and feel betrayed by others failing to do likewise, their resentment would be unjustified. However, it does not seem that we have arrived at such a state yet. As things develop, wars and crises caused by demographic structures or resource scarcity may well increase the likelihood that sacrifices you would rather not make will be crucial for our collective functioning.

Of course, it is likely that we will continue functioning if just enough, but not all, people discharge sacrificial duties. Requiring that of all may then seem like a waste of human resources and unjustified deprivation of flourishing. This is the most difficult issue. We may indeed argue that large modern societies can never become like small subsistence communities with neat interpersonal structures of mutual justification and an immediate sense of being in the same boat. Indeed, we may argue that the productive process on our developmental stage can never be a full-fledged we-mode action. When Tuomela argues that the conception of justice guiding the political community can be a we-mode guiding ethos, what he seems to have in mind is the decision-making and responsibility for our basic institutions, not the whole productive process.⁵⁴ Cohen’s ethos, by contrast, demands occupational choice be permeated by the ethos as well. Whether Cohen can defend this claim within liberalism, as he wants to, may depend on the strength of the personal prerogative – and also on the extent to which occupational choice shapes a person’s life. In most capitalist societies today, this extent is quite huge. If a citizen can only reject a life-shaping profession if it is detrimental to her personal religious or moral conviction, and otherwise has to be strongly guided by the common good in her daily pursuits, it seems that this resembles the moral outlook Rawls discusses as civic humanism or another comprehensive doctrine of the good life.⁵⁵ You are expected not only to take some minimal care for the functioning of basic political institutions but also to accept social contribution as the guiding value of extensive, life-defining choices. Moreover – and this distinguishes we-mode reasoning from an I-mode moral conviction arrived at when you embrace any moral theory, such as contractualism, utilitarianism or a religious doctrine – you must accept a collectively designed action-scheme that does not leave the judgement of what is most important right now to you alone. You are not acting according to your own best judgment, but instead you suspend your own judgement to do your share in a collectively established scheme.

Besides Casal’s worries with Cohen’s ethos noted in section I, there is also a compelling argument by Emily McTernan. She argues that a Cohenian ethos can be discriminatory against the (voluntarily or in-voluntarily) non-productive and less productive by cultivating a “unified sense of value” and suggesting social productivity as the “dominating value.”⁵⁶ Implementing a social ethos of occupational choice strong enough to conceptualize production as we-mode action integrating the whole work force amounts to an extensive shaping of common culture, sentiment and value and potentially stands in tension with liberal core commitments.

The civic virtues and attitudes, about which there can be an overlapping consensus and which Rawls allows the liberal state to actively cultivate, are not sufficient to guarantee such an integrated process. The closest Rawls comes to acknowledging the distinction between I-mode and we-mode might be in Political Liberalism, when he introduces the distinction between the reasonable (justice-oriented) and the rational (oriented towards one`s individual conception of the good) as two distinct and necessary presuppositions for a liberal citizen.⁵⁷ Still, it is doubtful whether the liberal sense of reasonableness is strong enough to rationalize an extensive occupational life subordinated to (re-)producing the common good. Note that Rawls says that “reasonable persons … are not moved by the general good as such but desire for its own sake a social world in which they, as free and equal, can cooperate on terms all can accept. They insist that reciprocity should hold within that world so that each benefits along with others.”⁵⁸

While Tuomela mentions the possibility to conceptualize the liberal nuclear ethos of a cooperative political society as a group ethos, Philip Pettit rejects this political ontology for Rawlsian liberalism.⁵⁹ He argues that the political society cooperating under justice as fairness is neither a group agent (solidarism) nor an aggregation of individuals (singularism). Instead, Pettit argues, the Rawlsian political society is what he calls a civicity – a third ontological type, looser than a fully concerted group agent with a shared goal and organizational structure, but still more integrated than a libertarian association.⁶⁰ Democratic decision-making and pluralism, Pettit argues, rule out the possibility for a political society to function as a group agent. The liberal reasonableness, it seems, is not equivalent to we-mode rationality, in which individual judgment is suspended and a group ethos is directly action-guiding. The reasonable liberal citizen cares for realizing an individual life plan under fair conditions but does not necessarily exhibit the natural propensity to negotiate the pursuit of their individual plan with the demands of a collectively established action-scheme (unless actions are required for preventing the breakdown of the political society as such).

Pettit’s ontological proposal shows that there are already good arguments to not envision the ties that unify a liberal civic society as a group ethos, as Tuomela suggests. Thus, it is even more questionable to extend the group ethos from basic commitments to upholding liberal institutions to the whole productive process. The solidaric citizen committed to securing the common good with their labour is thus more strongly social than the reasonable citizen in Rawlsian liberalism. To conclude, whoever cherishes liberalism about occupational choice must face the problem of strongly pro-social work.

Still, we can justifiably take a few measures, such as the ones suggested in section V, to alleviate the problem even without radically rejecting the advantages of modern liberal societies. Instead of making production a full-fledged we-mode action, we may strive for production to be guided by a basic common ethos, whilst allowing for plural views about how and with which activities to support the common good. It may still happen that the care worker or farmer feels that she does more for the common good than some culture or knowledge worker. But the sense of betrayal would be very limited if everyone showed equal concern for the common good, and the respective outlooks on life and occupational choice of different groups of citizens were therefore less polarized. Then what the knowledge worker does can be conceived as part of a joint action scheme even by those who have a hard time recognizing that occupational choice equally acceptable as the choice of a nurse. While shared knowledge is beneficial for maintaining full-fledged we-mode action, lacking shared knowledge in all detail will not necessarily lead to a breakdown of we-mode agency if enough trust in others is given. Strong trust in others’ willingness to use their talent for the common good, and in their judgment of how this is best done, can move the civicity envisioned by Pettit closer to an ethos-guided group without trampling democracy and individual rights.

Finally, even falling short of the ideal of full-fledged we-mode action may not be a knock-down argument against a liberal productive society. Its specialization and division of labour yields many advantages. Nonetheless, this should not give us permission to simply ignore the fact that we are still relying on people with propensity for a certain type of choice and that that type of choice is fully rationalized, recognized and respected only in a community in which this choice propensity is universal.

ACKNOWLEDGEMENTS

I am indebted to Herlinde Pauer-Studer for comments, discussions and encouragement to pursue this work. I am grateful to Anca Gheaus, Jens Jørund Tyssedal, Friedemann Bieber, Willem van der Deijl-Kloeg and Huub Brouwer for comments on an earlier version of this paper and the exchange at our pre-read workshop The Goods and Values of Work in August 2023. I thank the participants of the LMU Munich PPE Workshop: The Future of Work in December 2023, as well as Laura Valentini and the participants of her Colloquium in Social and Political Philosophy. As a source of inspiration for thinking about occupational choice and pursuing creative approaches in political philosophy, I benefitted from the students of our extra-occupational master programme Philosophie Politik Wirtschaft at LMU. Finally, I thank three anonymous reviewers and the editor of this journal, Bob Goodin, for help with improving and finalizing the paper.

COMPETING INTERESTS

The author declares that she has no competing interests.