John Roemer and the imprecise language of personal responsibility

Roemer escapes this feigned opposition between merit and destiny by elaborating a theory of responsibility that is less crude than those that have animated the debate so far. In Equality of Opportunity (Harvard University Press, 1998) he writes: 'I intend to introduce a distinction between responsibility and accountability, a distinction that is not usually made in everyday language (...) Holding a person accountable for an action,' the philosopher continues, 'means that he or she must pay the consequences: he or she should, perhaps, compensate those who have suffered harm as a result of that action, or be punished by society for it. In other situations, a person responsible for behaviour that generates consequences for himself must accept them: society has no moral mandate to change them. Now, in many cases, though not in all, we might think that a person can only be held responsible for a behaviour if he or she is responsible for it. (We cannot, for example, hold a hypnotised person culpable for the harm he caused someone while under hypnosis)' (p. 18). Up to this point, the philosopher's discourse seems to move only on the terminological and syntactical level. But then Roemer goes deeper and focuses on the opposite connection. On the fact, that is, that 'holding a person responsible does not always mean that we should hold him or her imputable. Thus, for example, we might decide that a teenager is responsible for poor school attendance because she has decided to skip class frequently, but nevertheless not hold her responsible for this, and we might even try to take measures to make up for her consequent educational deficit if the circumstances of her life explain her behaviour (...) I say that individuals should be held responsible for their degrees of commitment,' Roemer concludes, 'but not for their levels of commitment. The consequence of an action for a person should not depend on his circumstances, but on how much he commits himself. Those with the same degree of commitment should be held equally accountable for consequences; since they are equally accountable, their rewards (i.e., the levels of benefit or well-being involved) should be equal. We can only hold a person accountable for misbehaviour if it would have been reasonable, given the circumstances, for him to have behaved better: but the set of reasonable behaviours depends on the 'type' of person (...) taken as the set of behaviours observed in that type' (pp. 18-19). This distinction is therefore decisive. Being responsible for an action does not always mean having to bear the consequences entirely. I may have done something, I may have chosen it, I may have willed it in a sense that is not purely mechanical, and yet society may have reasons for not leaving me alone in the face of all that follows from that choice. The conclusion Roemer draws from this is clear: 'I believe it is morally wrong to hold a person responsible for not having performed an action that it would have been unreasonable to expect of a person in his or her circumstances (...) The distinction between responsibility and imputability allows us to maintain certain moral standards with regard to actions, but without punishing those who perform a reprehensible act if, in the circumstances, only a superhuman being could have prevented it' (p. 18).

Acknowledging that a person is responsible, therefore, does not necessarily imply holding them entirely responsible, or making them bear the entire cost of their conduct. Here a subtle moral space opens up, far from both paternalism and moralism. Justice should not treat individuals as children incapable of acting, but neither should it pretend that every choice is born under conditions of full mastery.

Think about school, for example. A girl stops attending regularly. She accumulates absences. She studies little. She arrives unprepared. Formally, the choice of such conduct is hers. No one has materially decided in her place. But if that conduct is born within a broken family, in a home without space, in a context where no one has ever turned school into a credible promise, how fair is it to make her bear the consequences of a choice that is also her choice? We can say that she is 'responsible' for the absences. But are we sure we can hold her 'accountable' in the same way as a peer raised in an environment where school attendance is protected by habits, expectations, resources and loving supervision?

John Roemer on this point invites us to distinguish moral judgement from social sanction. Not every mistake must become destiny. Just as not every waiver must turn into a fault. There is a difference between saying 'you have acted wrongly' and saying 'from now on you deserve all the worst consequences of what you have done'. It is a difference that securitarian and punitive ideologies tend to erase. And they erase it especially with the most fragile, because the language of responsibility often functions as a selective device. It is harsh with those who have few protections, but lenient with those who have many ways of turning their mistakes into accidents along the way.

Roemer's theoretical move is to link responsibility to commitment (effort), but not to commitment as an absolute quantity. In fact, as we have seen, he distinguishes between 'levels' and 'degrees' of commitment. Now this distinction shows its full moral import. People should be accountable for their 'degree' of effort but not for their 'level', because the latter is strongly conditioned by 'circumstances'. We must look at the 'degree' of effort, the relative position a person occupies in relation to those who share similar conditions to his or her own.

It is a counterintuitive idea because we are used to measuring commitment by the results it seems to produce. But these results are generated by the combination of circumstances and commitment. A method must therefore be found to eliminate the influence of circumstances in the evaluation. Measure commitment, that is, but only when circumstances are equal. We know, in fact, that the same performance can require very different energies. We know that for some, normality is sustained by an invisible network of help, while for others, every ordinary gesture costs extra effort. We know that courage does not always coincide with success, and that laziness does not always coincide with failure. We know that what happens to children from the first year of primary school onwards is strongly determined by what happened to them previously, in the period from gestation to the first day of school. Roemer gives the example of two children, Alan and Betsy. Alan belongs to a disadvantaged group while Betsy belongs to a privileged group. Both exercise the same absolute 'level' of commitment. But that same level places Alan at the 90th percentile of the observed distribution of effort in his 'type', i.e. in the group of pupils living in the same 'circumstances', while Betsy is only at the median of her 'type', even with the same level of effort. Faced with this situation, Roemer's conclusion is stark: 'Taking into account their respective circumstances, Alan put in more effort than Betsy' (p. 15). Given the same level of commitment, Alan's degree of commitment is greater because he has done something that, in his context, is far less common, less supported, less expected.

He who appears mediocre in absolute terms, therefore, may have exerted a very high degree of effort in relation to his 'circumstances'. Those who appear brilliant, on the other hand, may have merely glided along a favourable slope, taking advantage of life's advantages. Of course, this is not always the case and in fact Roemer does not turn every underdog into a hero nor every privileged person into a usurper. His argument is more sober and tells us that without information on 'circumstances', our judgement on responsibility is morally incomplete. Quoting Ortega y Gasset we could say 'Yo soy: yo y mi circunstancia' ('I am: me and my circumstances') (p. 15). Not just me. Not just circumstances. Justice moves in the space determined by the conjunction of these two realities. If it sees only the I, it becomes guilt. If it sees only circumstance, it becomes predestination. If it holds them both together, then it can perhaps begin to judge without humiliation.

This point is particularly important in a country like Italia, where responsibility is often invoked intermittently. We ask it of the children who drop out of school, the poor who do not 'get active', the unemployed who would refuse indecent jobs, the young who do not have children, the welfare recipients suspected of opportunism. We ask much less of the institutions that produce unequal schools, inadequate childcare, fragile transport, low wages, endless apprenticeships, labour markets where availability is confused with merit and vulnerability with flexibility. And young people flee abroad.

Responsibility, in this public usage, often becomes a vertical word. It descends from the top down. It is addressed to those who must justify themselves, rarely to those who organise the conditions within which others choose. Instead, Roemer urges us to make it symmetrical. If individuals are responsible for their choices, societies are responsible for the circumstances that make some choices reasonable, others improbable, others almost unthinkable. It is not a question of replacing personal responsibility with social responsibility. It is a matter of preventing the former from being used to conceal the latter.

The most difficult point concerns precisely what we can reasonably expect from a person. We cannot expect the same behaviour from lives so differently equipped, if we do not want to turn moral universalism into social blindness.

Of course, this approach can be disturbing. Who decides what circumstances count? Who determines what was reasonable to expect? How can we prevent understanding from becoming indulgence and indulgence from becoming injustice towards those who, in the same circumstances, have tried harder? Roemer knows the problem. That is why he does not entrust justice to individual compassion, but to a public procedure. The relevant circumstances must be discussed politically, defined collectively, made observable as far as possible. There is no algorithm that spares us moral conflict. But there is a more honest way of formulating it: not to ask generically how much inequality we are willing to accept, but which consequences we are willing to attribute to individuals and which to the social structure.

The stakes are high. A society that attributes too little to individuals risks weakening the very idea of 'agency', of personal initiative and autonomy. But a society that attributes too much to individuals becomes cruel, because it calls freedom what is often only differential exposure to the vagaries of fate. Responsibility, to be morally serious, must be situated. It must know from where it speaks, to whom it speaks, what it assumes when it judges.

In this sense, Roemer also helps us read the contemporary rhetoric of 'resilience'. Few words in recent years have had such good fortune despite being so shrouded in ambiguity. Resilient is one who resists, one who adapts, one who does not break. But when resilience becomes a duty imposed on the vulnerable, it changes nature. It is no longer a virtue; it becomes an elegant way of asking individuals to compensate themselves for the failings of institutions. It praises the girl who studies despite everything, the young man who works and graduates, the worker who reinvents himself after every crisis. But praise can turn into an alibi. If some make it despite everything, then everyone could have made it. This is where the exceptional case is used to feed the 'habit of injustice'.

This is not to deny admiration for those who endure, but to avoid making it a punitive measure for those who fail. That someone overcomes enormous obstacles shows the greatness of that person; it does not show the justice of the obstacles. A decent society should not construct public policies by asking everyone to be exceptional. It must ask what conditions allow ordinary people to exercise ordinary responsibilities without being crushed at the start.

The responsibility that remains, then, is no small thing. It is perhaps more demanding than the meritocratic one, because it is not satisfied with the surface of results. It does not simply say that everyone must answer for themselves. It tells us that everyone is accountable for what, given his or her circumstances, he or she could reasonably do. This addition changes everything. It introduces a social memory into the judgement. It reminds us that choices are personal, but are never born in an empty space.

Reading Roemer today is a necessary anti-ideological exercise. An exercise even more necessary today, when the moralising right tends to see in defeat an individual flaw and the more determinist left tends, still too often, to see the individual only as the place where structure reproduces itself. Roemer accepts neither simplification. He asks us to inhabit that uncomfortable zone where people are indeed agents, but agents who are prepared, supported, oriented, and authorised to desire in a fundamentally unequal manner. In his perspective, therefore, justice begins when we learn to judge without forgetting. Without forgetting that the language of responsibility can ennoble freedom, but also justify abandonment. Without forgetting that holding someone accountable for their life is only legitimate if we are willing, at the same time, to hold society accountable for the conditions under which that life had to choose. The Roemerian perspective, then, forces us to stop using responsibility as a judgment and to treat it, finally, as a demand for justice.