Rewiring Ethics: Collective Action, Recognition, and Fractal Responsibility

A. Collective Action Problems

Many situations of moral significance have the following structure: there is some large negative outcome that would be produced or prevented only by a pattern of actions across the individuals in a large group. The obtaining of this pattern does not significantly depend upon the choices of any one individual in the group. Moreover, each individual faces some significant incentive to do their part in collectively producing this outcome (usually some palpable cost or benefit for themselves or their family). Call such cases collective action problems. The choices we make about which job to take, whether to strike or protest, which neighbourhood to live in, which schools to send our children to, how to vote, and about the environmental impacts of our lifestyles can all have significant negative social impacts. And yet, in many of these cases, individuals face incentives to conform to the problematic pattern of actions.

Moral theorists standardly want their moral theory to yield the result that individuals should behave in ways that mitigate these negative social impacts. It is common for theorists to oscillate between two scales: either giving responsibility to individuals for their marginal contribution to some massive pattern, or giving responsibility to massive agencies to instruct individuals blindly.¹ But ideally, we want individuals to be doing something significant themselves while also participating in some larger collaboration of significance.

This article starts with a conjecture. Lots of people find work more fulfilling when it is recognisably worthwhile: when they are more transparently and palpably connected with the value they produce.² Relatedly, lots of people are more motivated to work when they are more transparently and palpably related to the value they produce. Call this the Localising Preference.

I assume this is intuitive enough. For instance, many people prefer to work for the needs of a local community rather than following centralised instructions as part of a massive corporation; many people also prefer to consume products that were made with their particular needs in mind rather than mass-produced commodities. We can find support for this conjecture in empirical psychology,³ management science,⁴ economic theory,⁵ and contemporary ethical theory.⁶ This also connects up with an important theme in political theory and moral theory. Recognition is a key aspect of the ideal of unalienation in Hegel and Marx.⁷ This is also an important theme among contemporary critical theorists, in particular Nancy Fraser⁸ and Axel Honneth.⁹ A related notion of uptake or completion is important in care ethics.¹⁰ Recognition is also a key aspect of the ideal of contractualism developed by T.M. Scanlon, which also inspired Waheed Hussain’s criticisms of market opacity.¹¹

The central theme of this article is that taking morally seriously the importance of this kind of recognition provides the basis for a new standard for dividing up responsibilities in collective action contexts. This approach enjoins dividing labour in ways that preserve recognisably worthwhile responsibilities where possible. This injunction would apply within a given group of individuals with some large responsibility, but also to the allocation of responsibilities to those groups of individuals themselves. This supports a fractal model of moral responsibility, where each agency is assigned recognisably worthwhile responsibilities as participants in collaborations that are themselves assigned recognisably worthwhile responsibilities as parts of larger collaborations, and so on. In this way, individuals can take responsibility themselves for suitably-sized needs while also taking responsibility for playing their parts in increasingly larger collaborative endeavours. Since there will be natural limits on the extent to which responsibilities can be feasibly allocated in recognisably worthwhile ways, this approach also yields a taxonomy of strategies, favouring local management of needs where possible, but also favouring large-scale management of large-scale values in suitable contexts.¹²

B. Quick Contrast with Rule Consequentialism

The literature on marginalist approaches to collective action is vast.¹³ I want to focus on, and make a proposal within the context of, more collaborative approaches.

One option is to defend an indirect consequentialism that treats the Localising Preference strategically, as an empirical fact about what would affect the consequences of different patterns of rule-conforming behaviour. This striking but characteristic remark from A Theory of Justice is significant here:

In designing and reforming social arrangements, one must, of course, examine the schemes and tactics it allows and the forms of behaviour which it tends to encourage. Ideally the rules should be set up so that men are led by their predominant interests to act in ways which further socially desirable ends. The conduct of individuals guided by their rational plans should be coordinated as far as possible to achieve results which although not intended or perhaps even foreseen by them are nevertheless the best ones from the standpoint of social justice. Bentham thinks of this coordination as the artificial identification of interests, Adam Smith as the work of the invisible hand.¹⁴

Rawls thought it was a job for social science to figure out what sorts of rules would produce the best overall outcomes.¹⁵ Notice that the ‘predominant interests’ we are to be led by are given antecedently – in advance of collaboration.¹⁶ Moreover, the overall results need not be intended or even foreseen by the individuals.

There are two problems here. One is extensional. Consider cases in which a localising rule (to anticipate: one assigning more recognisably worthwhile responsibilities) would bring about less overall value. In such cases, the localising option would promote more disvalue than the option that promotes the most marginal value – though perhaps not all that much. Rule Consequentialism would not advocate a localising rule in such cases. But I think that even if the overall result would not be value maximising, it may well be most morally desirable to follow a rule enjoining one to act in a localising way.

The other problem is methodological. I want to resist the idea that we design rules based on exogenous descriptive expectations about how individuals are likely to behave. That gives individuals too little moral responsibility. It treats moral motivation as analogous to non-agential influences on outcomes; this is a classic mark of moral alienation.¹⁷ And it threatens to make the relationship between substantive individual responsibilities and social value too opaque.¹⁸

In my view, it is a mistake for Act Consequentialists (or other less extreme marginalist approaches) to think that individuals have some individual responsibility for everything of value that they can affect. This seems to give individuals too much responsibility – and for each one not to give others enough responsibility. Act Consequentialists are not moral team players. But it is also a mistake for Rule Consequentialists to think that individuals are merely responsible for following whichever massive rule happens to be most impactful in the aggregate. This seems to give too little consideration to the specific moral responsibilities of individuals. It seems problematic to me that individuals have no guarantee that they will be able to ‘foresee’ the ways in which their actions contribute to the ‘best overall outcomes.’ It is mere luck in this case when the best rule enjoins individuals to take a more localising responsibility for particular individuals. Intuitively, the Rule Consequentialist approach gets the right result (when it does) for the wrong reasons.

C. The Recognisable Responsibility Principle

I offer instead:

The recognisable responsibility principle: Individual agents should be assigned recognisably worthwhile responsibilities where possible.

I’m going to spend most time on the most distinctive aspect of this proposal, the relevant moral significance of recognisably worthwhile responsibilities.¹⁹ Let me first say something briefly about the other elements in this principle.

Firstly, by ‘responsibility’ I have in mind something like Zheng’s notion of responsibility as accountability:

We are … responsible for an action in this accountability sense when it is appropriate for others to hold us to certain expectations and demands regarding our duties and tasks— and to sanction us when we fail to carry them out.²⁰

As mentioned, this approach is neutral about the conditions under which a responsibility is worthwhile. There is a range of views one could take here, from something being worthwhile when it meets the needs of others,²¹ or when it realises value,²² and perhaps even when the action is not reasonably rejectable on the basis of personal reasons.²³ A fuller development of this approach would need to take a stand on this question. I’ll default to talking about meeting needs.

It is important to keep in mind that the recognisable responsibility principle is a collaborative principle.²⁴ The principle distinguishes merely acting on recognisably worthwhile responsibilities – as one might without knowledge or concern for one’s potential collaborators – from acting in a suitable collaboration, in which one takes oneself to be playing one’s collaborative part. One’s first-order reasoning and one’s behaviour in the specific interaction may be superficially similar in the two cases. But one’s overall motivational orientation would be importantly different; this also has counterfactual implications for one’s motives and actions, and for the prospect of recognising oneself as participating in a larger collaboration responsible for larger values. The recognisable responsibility principle does not advocate a focus on one’s own specific responsibilities while neglecting the circumstances of others. Rather, it is based on collaboration, and hence, in the case where all agents are similarly situated, a form of generalised normative expectations. According to the recognisable responsibility principle, one upholds one’s responsibilities as a way of playing one’s part in a collaboration with others, where those others also have recognisably worthwhile responsibilities.

To anticipate: this is also relevant to the proper basis of the relevant kind of recognition, which is not merely that the activity is worthwhile, but that the activity is collaboratively worthwhile. This collaborative relationship is one we can stand in to the patient and other would-be agents. The recognition of this fact, that we are collaborating in worthwhile ways, can form the basis for a desirable form of moral community.

A. Explaining Recognisability

Let me build up to that last claim by starting with the simple question: What are the conditions under which a responsibility is recognisably worthwhile?

Here’s a first thought: a responsibility is recognizably worthwhile if the agent (or some other pertinently situated agent) is in a position to know that upholding the responsibility would be worthwhile.²⁵

But this won’t quite do. For in many collective action thought experiments (such as Glover’s ‘baked beans’ or Parfit’s ‘harmless torturer’ cases²⁶), we do know what the overall effect of our action and everyone else’s actions will be, and yet individual actions are not plausibly recognizably worthwhile, in an intuitive sense that I’m trying to capture. Here it is relevant that Karl Marx – making reference to Smith’s famous example of the division of labour in a pin factory – actually made fun of the idea that recognition in production might come merely from the fact that one knows that one is a trusty cog in a good system, whatever the specific content of one’s responsibilities:

Man remains a maker of pin-heads, but he has the consolation of knowing that the pin-head is part of the pin and that he is able to make the whole pin. The fatigue and disgust caused by the eternally repeated making of pin-heads is transformed, by this knowledge, into the ‘satisfaction of man’.²⁷

There you are every day, banging away at some fraction of some pin. You have read your Wealth of Nations, so you know that everyone’s taking a tiny part of the overall productive process is massively more efficient than each person taking responsibility for making the whole pin. But still, there you are, banging away at these fractions all day, miserable – and perhaps now also feeling guilty because you think you should be pleased that you are helping people efficiently.

This is closely connected with the idea that moral theories that give no importance to the Localising Preference are alienating. Interestingly, the idea that workers will be alienated if they lack some more palpable knowledge of the worth of their efforts is well-known among business leaders and military strategists.²⁸ Andrew Carton advocates that we ‘conceptualize leaders as architects who optimally motivate employees when they create a cognitive blueprint composed of a small and streamlined constellation of connections that link everyday work and the organization’s ultimate aspirations and then allow employees to mentally assemble more elaborate connections around that core structure.’²⁹ Carton offers this as purely instrumental advice for managers trying to motivate employees – because alienated employees are less productive.

These insights from business management and military strategy provide further empirical evidence for a widespread Localising Preference; their prescriptions point in the direction of assigning recognisably worthwhile responsibilities to individuals. But there is a risk in these cases that these managers are merely trying to change the perceptions of their work. This would collapse into the proposal that Marx made fun of: changing perceptions without changing material reality.

For the nature of our activity is also relevant, not merely a changed perception or greater knowledge of activity that, along with the coordinated activity of some large number of others, has sufficiently desirable consequences.

A more palpable sense that one’s efforts are meeting needs is what is missing in the pin factory. Even if you know, theoretically as it were, that your action has some impact in the void, this is no part of your productive experience. It is all the same to you whether the eventual causal impact is positive or negative or neutral. It is likewise all the same to the patient whether and why you did whatever you did. This is why one cannot simply redescribe cretinising or massively diffuse contributions into something recognisably worthwhile. It is also why the stratagems of the military and business leaders ring hollow: they are (some of them) mere rebranding.

My suggestion is not that we rebrand, but rewire: that, where possible, we restructure collective activity in ways that assign individuals recognisably worthwhile responsibilities, on some not-far-fetched understanding of what this amounts to.

One more distinction will help to clarify what I have in mind by a more palpable knowledge of the nature of one’s worthwhile activity.

Start with the notion of ‘completion’ described in the care ethics literature.³⁰ Here is Kittay:

Care is not something we do to something or someone. It is something we do for another’s benefit. There has to be an uptake on the other’s part …To underscore the active element in the reception of care, I will speak of the completion of care as the ‘taking-up of care’.³¹

At issue here is the taking-up of care by the cared-for – the recipient of the caring act of service. The central idea is that care involves respect: an open attentiveness to and engagement with the subjectivity of the cared-for.³² Care is not merely about imposing benefits, but about working with the patient to meet their needs as they see them – at least when we can.³³ As a thesis about the nature of ideal care, this seems plausible. I also think that uptake is morally important; this can be a helpful corrective to a widespread consequentialistic assumption in ethics that assumes that the ‘promotion’ of benefits is pro tanto morally good,³⁴ in favour of a more relational approach.

It is plausible to describe the ‘taking-up of care’ as the recognition of care by the patient. In this sense, recognition involves both a representation of something as a certain way and the manifestation of approval for that thing’s being that way.³⁵ This involves the manifestation of esteem, for instance in the expression of gratitude for a service.

But this is still not quite the kind of recognition under consideration. The central issue for the recognisable responsibility principle concerns the prospect of recognition by the agent that their own service is worthwhile.³⁶ I suggest that we get this by adding another layer to Kittay’s notion of uptake.

We can find this more complex ideal in the slightly different notion of ‘completion’ from the writings of the early Marx.³⁷ In Marxianly completed service, an agent does what they can to meet a patient’s specific need; their doing so is recognized as such by the patient; and the patient’s recognition is recognized, in turn, by the agent. This is what we are after. I understand palpable knowledge that one’s service is worthwhile, in the specific case in which one is serving another human being, then, as one in which: the agent is knows and approves of the fact that their service is worthwhile partly in virtue of the fact that the patient knows and approves of their service being worthwhile. This is the notion of agent recognition at stake in the recognisable responsibility principle. That principle does not require that there be such recognition. Rather, it enjoins us to empower agents by putting them in a position to recognize their activity as collaboratively worthwhile in this sense. In short, the kind of recognition at stake in the recognisable responsibility principle is being in a position to have agent recognition of patient recognition of the service as worthwhile.

This is obviously a rather stringent condition. However, as we will see, in cases where the various features of such a robust ideal cannot be satisfied fully, the recognisable responsibility principle can be satisfied by degrees and/or by making these limitations as such transparent.³⁸

B. The Value of Recognition

It would be a mistake to think that mere recognition is valuable in itself, as it were, without being recognition of something worthwhile, such as the actual meeting of someone’s needs. That mistake involves the fetishism of esteem. The recognisable responsibility principle is designed to avoid both overly materialist and overly idealist approaches, or more positively, to combine the strengths of both.

Consider this remark from the third part of the Theory of Justice:

What binds a society’s efforts into one social union is the mutual recognition and acceptance of the principles of justice; it is this general affirmation which extends the ties of identification over the whole community.³⁹

We’ll shortly take up Iris Marion Young’s extrapolation of this thought to a wider class of social relations. For now, I want to accept the thought that the primary source of the value of recognition of worthwhile service is, in the first instance, in the importance of being able to identify oneself as a participant in a suitably moral community. I am taking up the suggestion from Marx, Rawls, and many others in the socialist, feminist, and communitarian traditions, that such identification is an important human good, and one that will properly serve as a part of an ideal of moral community.⁴⁰

These relationships are of tremendous importance in moral life. There is a tendency to abstract too far from such quotidian details in analytic moral philosophy. There is often a presumption that only necessary conditions and universals can be starting points in moral theory. But an alternative, equally venerable, more Humean, tradition, starts with what David Wiggins has called the ‘truly irrevocable cares and concerns of human existence.’⁴¹ We can think that morality is about people, about community, and start with what people are really like and what community is really like – not just empirically, but phenomenologically. On this approach, it becomes easier to see how the phenomenology of doing something recognizably worthwhile can be important to someone, and that kind of importance could be a condition on the structure of moral responsibility.⁴²

C. Moralities of Recognition

Let us say that a morality of recognition maintains that moral standards bearing on actions obtain at least partly in virtue of standards bearing on a certain kind of ideal relationship between the relevant parties. Such views are more straightforwardly contrasted with consequentialist approaches on which morality is about promoting antecedently specifiable goods rather than acting in order to constitute moral community with others.

This recognition-based approach has roots in Karl Marx and John Rawls, and has been given particularly clear expression in T.M. Scanlon’s defence of moral contractualism.⁴³ In order both to show the methodological overlap, and the substantive differences, it will be helpful to contrast the two a little more. To be morally wrong, for Scanlon, is to be unjustifiable to others. The standard of justifiability is given by the contractualist principle: roughly, an action is unjustifiable if its performance under the circumstances would be disallowed by any set of principles for the general regulation of behaviour that no one could reasonably reject.⁴⁴ This forms the basis of the relationship between agents and relevant parties, which is, loosely speaking, one of respect as justifiability. Scanlon’s argument for this approach is phenomenological.⁴⁵ As Scanlon says:

The contractualist ideal of acting in accord with principles that others (similarly motivated) could not reasonably reject is meant to characterise the relation with others the value and appeal of which underlies our reasons to do what moralities requires. This relation, much less personal than friendship, might be called a relation of mutual recognition.⁴⁶

I am taking on board the idea that the value and appeal of mutual recognition might underlie the authority of a set of guiding principles in collective action contexts – which is most contexts, really.

However, there is an important contrast between the transcendental standard of recognition in Scanlon, where agents are merely enjoined to act in accordance with certain principles discernible largely a priori, and the normative-psychological notion of agent recognition under consideration here.⁴⁷

The distinctive suggestion that I am drawing out in this article is that recognition of others is not a formal property, or a merely epistemic condition, but a palpable fact about the human experience of living together under certain conditions. I am suggesting that recognition of serving one another is an important desideratum in the moral assignment of responsibilities. This seems to be an important part of the phenomenology of everyday moral life.⁴⁸

D. Two Objections (One Postponed)

Before applying this structure to various kinds of collective action cases, there are two natural kinds of objections one might raise about the foundational moral role that I’m assigning to human recognition.

One is that there are jobs that need to be done that cannot feasibly be organized in such a way as provide workers with a palpable sense of the ultimate worthiness of their efforts. I’ve been keeping this objection at bay so far with the occasion reference to localizing responsibilities where possible. I will shortly generalize this approach to offer a principled account of non-localisable cases.

The other concern, naturally raised from a more consequentialistic perspective, is that insisting on the importance of recognition can sound problematically self-interested. It is as though one is focused on the value to oneself of serving others rather than the value to the others. But that is a misunderstanding of the functional role of recognition in the approach I am advocating. The first point to make in response is that one’s reason for action in service is not recognition but the meeting of the relevant need. The value of recognition is a by-product.⁴⁹ The importance of recognition plays a different explanatory role, as a general standard on desirable divisions of responsibilities.

This connects with another point about collaboration: one is not merely rearranging one’s own responsibilities to make them more personally agreeable. Rather, the conjecture is that recognizable service is important to all of us. This explains the significance of structuring responsibilities for everyone to empower recognizable service. This also fits with the shift in moral perspective from the circumstances of patients to the circumstances of agents. A significant part of the quality of our lives together is determined by our own activities, not just the resources we have or the ways we materially impact on others. And yet the quality of service has been oddly erased from much moral theory.⁵⁰ I suggest that a morality that starts with the quality of service – everyone’s service – has much to recommend it. This is true both for agents assigned recognizably worthwhile responsibilities where possible, but also for ‘patients’ who are in a position both to recognize that their needs are being met and to recognize the work of those meeting them.

A. Localisable Collective Action

The most obvious practical implications of the recognisable responsibility principle concern localizing postures in consumption and production activities. The principle advocates spending more time and energy caring for one’s local social environment in production (in one’s professional and other contributive projects) and in consumption (patronizing local producers) than one spends on more distant social environments, even at some cost to overall value promotion.

The principle also advocates a certain amount of reasonable partiality in one’s projects – to care more about one’s own children and students than those of others. How does it do this? Easily: children and students are ready bases of important kinds of recognition.⁵¹ It would be natural to universalize responsibilities in ways that preserve these recognition-enhancing responsibilities even at some cost to overall efficiency or value promotion. (Plus, as we saw earlier from moments of extensional convergence between Act Consequentialism, Rule Consequentialism, and the recognisable responsibility principle, promoting overall value will not always conflict with distributing recognizable responsibilities.⁵²)

Perhaps a more interesting kind of application reconsiders the limits on what constitutes the group with which one is collaborating – which involves questioning the presumption of the ‘case’ in moral reflection. The idea is that one aims to break down a large group with a single end into smaller subgroups all with recognizably worthwhile ends. One can break those subgroups down again, and again, until one’s participation in the group is itself easily recognizable. The clearest example of this is in real life is the structure of armies. To reduce desertion rates, armies are divided into increasingly small units – small enough to make desertion humanly recognizable.⁵³ We can reproduce this structure in industrial action, for instance. Take the recent industrial action in UK academia. Forgoing wages and withholding teaching is individually costly to academics; the implications of one individual teacher’s doing so are fairly negligible in national negotiations. But rather than focusing on the national movement, one can identify with one’s own university, or school, or department. The strategies one employs might well be different in these different cases, if the strategy that increases the chance of success at your university conflicts with national strategy.⁵⁴

Importantly, the idea is not that in such cases one is choosing whether to collaborate with the university as the larger group or with the national group. Rather, the suggestion is that one is choosing whether to add an additional agential layer in the total collaboration. The additional layer – the university-sized group – itself aims to honour the recognisable responsibility principle. In fact, the national struggle for fair terms in academia itself can be seen as a collaborative layer within a larger struggle for fair terms across the economy as a whole – and that in some country, and so on. At each level of organization, the recognisable responsibility principle advocates assigning to all participants in some group some recognizably worthwhile end, to the extent possible; that group is itself assigned this responsibility as part of a larger group assigning recognizably worthwhile responsibilities; and so on.

This is the fractal ideal for social organization: at every level of social organization, agencies are upholding recognizably worthwhile responsibilities as a way to participate in a suitable collaboration with relevant others. Every agency has responsibilities in two directions as it were: their first-order responsibilities (designed to be recognizably worthwhile) and their responsibilities to the collaboration (to take on such first-order responsibilities as fits with a distribution of recognizably worthwhile responsibilities to the other participants in a way that meets the group’s responsibilities).⁵⁵

It is worth pointing out that the metaphor of ‘localism’ can be misleading. The presumption of localism comes from human limits on service and recognition. That being said, a proper division of responsibilities will probably require some people to be in charge of coordinating subgroups, and the larger groups of which they are a part. Some individuals have abilities (including based on training) that are well suited to taking responsibilities for municipal and national coordination of strategies; others have abilities better suited to meeting more personal needs.

In cases involving larger agencies, the notions of recognition, ability, and need will all be scaled up appropriately. The needs at stake will often be massive – such as the concerns of public health in a mining community, or the importance of democratic rules in a massive state. The relevant abilities will also be massive; clearly, groups can do lots of things that no individuals can do by themselves. We will also need some collective notion of recognition, such that there could be recognition by and of a massive agency, such as Preston or Cleveland, for instance. This provides an alternative source of justification for subsidiarity and municipal economic management, as in certain models of Community Wealth Building.⁵⁶

Municipalities might themselves aim to honour a version of the recognisable responsibility principle, thereby to realise a kind of community-level analogue of mutual recognition. They might aim to invest in local labour and capital in ways that are designed to foster relations of solidarity, and doing so in a way that is nevertheless responsive to some overall distribution of abilities and needs across a far larger population. This is precisely the moral ambition of some participants in the Community Wealth Building movement, which seeks to structure capital and labour management within a community in ways that provide short- and long-term benefits to that very community. By simultaneously seeing themselves as participants in a larger collaborative enterprise, these municipalities are also in a position to avoid concerns about self-interested protectionism.⁵⁷

B. Non-Localisable Collective Action

I have so far mostly focussed on cases in which responsibilities are localisable. But it will not always be possible to assign recognisably worthwhile responsibilities to individuals in collective action situations. This will sometimes be due to the nature of goods being collectively managed, and sometimes due to the state of prevailing technology. In this subsection, I’ll talk about the technological limitations; in the next, about collective values.

Climate change is a good example of a technological limitation on localisation. Climate change is already negatively impacting the lives of billions of people, and it has resulted from the actions of billions of particular people. But it is not possible to distinguish recognisable relationships connecting particular agents and patients, either retrospectively or prospectively, due to the global and systemic nature of the causal chains in play. This is importantly different from the kind of recognition that is possible in lots of production and consumption contexts. It is also importantly different from the recognition that is possible in military and industrial action cases, where the collaboration itself can be made more recognisably worthwhile. In cases where the nature of the phenomenon or the limits of technology render recognisably worthwhile responsibilities unavailable, we should opt for effective collaboration. Our contributions to this effective collaboration can still then be recognisably worthwhile in the following different sense: so long as the rationale for our responsibilities (stated in this paragraph) is as transparent to us as possible, we will know that we are contributing in the best way that we can.⁵⁸

The moral difference between the localising and non-localising responses in these cases is explained by a fact about the nature of alienation, which is that alienation is agentially-imposed. (I’m assuming that the relevant notion of alienation is the negative counterpart to the positive notion of recognition.) You are not alienated from someone separated from you by a river, just inconvenienced. But you would be alienated from someone who you could not spend time with due to homophobic social norms or your own racist attitudes or because they were afraid to say the wrong thing in case you fired them. In localisable collective action situations, where we do not distribute responsibilities in ways that empower recognisably worthwhile responsibilities, we are plausibly alienated from one another.⁵⁹ However, in non-localisable collective action situations, where we cannot distribute responsibilities in ways that empower recognisably worthwhile responsibilities, transparently distributing responsibilities in some other satisfactory way is the best we can do.⁶⁰ We are not alienated by the non-availability of greater recognition in these cases.

Somewhat ironically, I think we would be alienated by a moral approach that claims that one’s contribution is individually worthwhile in such cases – even if we are certain that our contribution makes some tiny difference to the massive outcome.⁶¹ For that would constrain us to misinterpret the point of the proceedings and the real significance of our contribution. And in fact, I’m inclined to agree with Walter Sinnott- Armstrong that we should freely draw on extrinsic incentives to ensure compliance with the collectively optimal strategy in some non-localisable cases.⁶² This should be within the remit of our collaborative design of responsibilities in non-localisable cases.

This also yields a response to a natural question from consequentialistically-inclined philosophers. What about rich people in an unequal world? Should they still prioritise the recognisability of local service over the far more significant good they can do opaquely, e.g. by simply relinquishing their fortunes to a well-run charity?⁶³ And if not, do we collapse into consequentialism?

In reply, this sort of case involves a combination of localisable and non-localisable considerations. Consider international economic injustice, for instance (analogous considerations will apply domestically). Global injustice calls for reparations and radical changes to international trade and international politics. These are not changes that most individual rich people can bring about by themselves. Take reparations, say. These reparations are recognisably worthwhile at some macro-level, which is where they are properly situated. A given rich person should do their part in financially supporting such reparations, for instance, through aggressive taxation. The rich can also actively campaign for aggressive taxation policies, including using their money to finance these campaigns. They might also focus their energies on campaigning within their own locality.⁶⁴

None of this involves a collapse into consequentialism, because there is no presumption that one must donate all one’s time and resources to the point of diminished marginal returns, and because individuals must still see themselves as making a suitable contribution to a collaborative effort.⁶⁵

The moral approach under development in this article lends itself to an integration between moral and political considerations. This project blends into normative politics. This is a feature, not a bug.⁶⁶

C. Collective Values?

What about the value of equality or democracy or autonomy or ownership rights? It doesn’t seem possible to assign responsibility for these values to a particular individual because they are, in part, essentially collective values. These are not examples of local values that we lack the technology to assign to individuals. Rather, these are values that are essentially realised in collective arrangements.

Not all of these collective arrangements will be massive. It would be a mistake to think that democracy is something instantiated only at massive scales, and even more of a mistake to think that democracy is constituted only by massive plebiscites. Much of democratic life is local; this will mean the delegation of large swathes of political authority to regional and local authorities.⁶⁷

Still, some democratic processes in some places are constituted by massive plebiscites. The importance of participating in them is explained by appealing to the prospect of recognition suitable to this kind of value. Ideally, participating individuals have a transparent account of why the policies managed at that level are best managed at that level and with precisely that amount of individual accountability (which is to say, not terribly much). Participating in massive plebiscites has value precisely in playing an almost performatively recognitive function: we all engage in the theatre of voting in part to manifest our commitment to democracy itself. This explains why expressivist theories of voting can feel quite compelling.⁶⁸ Earlier I said that there is a distinctive kind of value instantiated in societies structured to empower individuals to recognizably serve one another. One way we enact structures that have this value is by participating in democratic processes.

Does this strategy collapse into the wishful thinking that Marx made fun of? No, precisely because there is a transparent explanation for why this particular collective action has to be undertaken at this scale and in this way. Such an explanation is lacking in the pin factory, on the assumption that profit-maximisation is the preeminent organizing value there. In the case of massive plebiscites, the best moral explanation acknowledges the importance of recognition, in the sense that we have been considering. Given the nature of the values at stake in massive plebiscites, we should not expect individual actions to be efficacious. In fact, more strongly, in light of this transparent explanation for the inherent massiveness of this collective activity, we should expect ideal individual actions in such contexts not to be individually efficacious.⁶⁹ The ‘rationality of voting’ is mostly puzzling for philosophers and economists; ordinary people do not naturally deploy a standard of consequentialist rationality in this context.⁷⁰