Why the Social Connection Model Fails: Participation is Neither Necessary nor Sufficient for Political Responsibility

Abstract Iris Marion Young presents a social connection model on which those, and only those, who participate in structural processes that produce injustice have a forward-looking responsibility to redress the resulting injustice by challenging the structures that produce it (she sometimes calls this a political responsibility [Young 2011]). In Young's view, this is an all-things-considered, albeit discretionary, responsibility. I argue that participation in a structural process that produces injustice is neither necessary nor sufficient for having political responsibilities, and that therefore the social connection model must be rejected. A subtler model is needed, one that depicts participation in a structural process that produces injustice as sufficient (but not necessary) for having pro tanto forward-looking responsibilities to redress the process, unless the participating agent satisfies certain excusing conditions. I suggest the intuitive force of the thought that mere participation gives us political responsibilities can be explained by more fundamental considerations. Hastily, we might conclude that all participants have political responsibilities simply because most of them satisfy at least one of the following conditions: they cause injustice to continue, they are morally responsible for injustice, they benefit from injustice, they have communal ties with the victims of injustice, or they have the capacity to redress injustice.

that are blameworthy for the relevant injustices; the agents arguably lack any intent to do harm, and the injustice does not consist in a single past event but instead involves an ongoing process. Still, as Young observes: "the claims of the anti-sweatshop movement seem to have struck a chord with many individuals and institutions" (Young 2011, 126). Her diagnosis is that the complaints of the anti-sweatshop movement seem odd only if we apply a standard model of responsibility. The intuitive appeal of the activists' claims, she argues, must stem from an alternative model of responsibility; and this alternative model is the social connection model (SCM).
In summary, SCM consists of two claims: The Limiting Principle of SCM: An agent A shares with others a forward-looking responsibility to challenge a structural process S that produces injustice if and only if A participates in S.
The Allocative Principle of SCM: What A should do in order to challenge S can be decided, roughly, by considering the power, privilege, interest, and collective ability A has in relation to S in virtue of A's social position within S.
Young often uses "political responsibility" as shorthand for describing a forwardlooking responsibility that one shares with others 1 and "structural injustice" to denote unjust outcomes produced by structural processes. I will follow her in this. For the sake of brevity, I will also use Limiting and Allocative to refer to the two principles of SCM outlined above. The activists' claim is intuitively appealing. However, SCM does not satisfactorily explain this claim. Limiting might seem to give the right verdict when applied to the global garment industry. However, it has counterintuitive implications when applied to some other practices. Take child prostitution, for example, and consider a group of people who have an opportunity to challenge the structural processes leading to this problem. If they take this opportunity, we might assume, significantly fewer children would be vulnerable to being exploited for sex. Imagine further that the members of the group who have an opportunity to intervene do not participate in the structural processes leading to child prostitution. They do not buy sex from children, they do not profit from this trade, and so on. It seems that this group might have the responsibility to challenge the processes in question. However, Limiting entails the opposite. This indicates that we need more accurate criteria for deciding who has forward-looking responsibilities to challenge structural processes leading to injustice than SCM provides. I will suggest such criteria toward the end of the article.
The arguments for why SCM does not accurately capture the activists' intuition need more careful explanation than provided here in the introduction, and I will provide this in due time. First, we need a firmer grip on this model.
Limiting identifies the agents with responsibility for some particular deprivation, and Allocative determines what agents are called upon to do given that they have the responsibility. 2 More precisely, the parameters indicated in Allocative give agents "some guidance in reasoning about how to take action to try to undermine injustice" (Young 2011, 144). The distinction between the two principles is vital for the analysis of SCM, since the considerations Young finds important in connection with the former differ from those she considers important for the latter (although, as we will see shortly, there are some important connections: the first concerns the agent's participation in S, whereas the second concerns the agent's power, privilege, interest. and collective ability).
Limiting and Allocative are my interpretation of SCM as it is set out at greater length in Responsibility for Justice (I argue for this interpretation in sections III and IV). Taken together, the two principles omit one element of Young's model: the notion that we should shoulder our political responsibility to challenge structural injustice in cooperation with others. This element is of limited importance given the focus of this article, and most of the examples I use capture it anyway, since they typically involve numerous agents with the opportunity to ameliorate injustice through collective action.
Young describes political responsibility as a discretionary and blameless responsibility akin to role responsibility. She says it resembles the responsibility parents have for their children, or CEOs have for their companies. The idea that political responsibility is discretionary means that it is to some extent up to the agent to decide how to take it up (and, we might add, when to do this). It is, for instance, my responsibility as a parent to see to it that my children learn to swim, as this is something I should do, all things considered, but it is largely up to me how and when to do so. 3 Young uses some notions-"social position," "structures," "structural process," and "injustice"-that need a little explanation. Social positions are major general categories of agents. In the global garment industry, such categories may include consumers of garments, garment companies (or "labels"), the shareholders of these companies, trading agents, owners of the factories where the clothes are produced, and workers who produce these clothes. In Responsibility for Justice, Young states that structures are "notoriously hard to define" (Young 2011, 52), but she once described them as consisting "in the connections among these [social] positions" (112) and in the enablements and constraints that attend a certain social position. In Responsibility for Justice, Young often refers to structural processes instead of structures in order to "emphasize the dynamism of action in institutional contexts" (53). The idea is that when people act in everyday life they are drawing on the enablements and constraints afforded by their social positions, and when they do so they are (often unknowingly) reproducing the structural properties presupposed by their actions. For instance, when we buy clothes, our actions at the same time presuppose and reproduce the structural processes of the global garment industry; we are participating in these processes. Importantly, for Young's purposes, the structural properties presupposed and reproduced by the structural processes might involve injustice: Structural injustice . . . exists when social processes put large groups of persons under systematic threat of domination or deprivation of the means to develop and exercise their capacities, at the same time that these processes enable others to dominate or to have a wide range of opportunities for developing and exercising capacities available to them. (Young 2011, 52) This outline of structural injustice seems applicable to the global garment industry.
In relation to this, it might be worth noting that according to SCM, political responsibility for structural injustice is a responsibility to challenge the structural processes that lead to unjust outcomes, that is, to make them less prone to produce structural injustice. It is not a responsibility to, for instance, compensate the victims of injustice. 4 Allocative incorporates an important link to Limiting: what is relevant for deciding, roughly, what the agent should do in order to ameliorate structural injustice is the power, privilege, interest, and collective ability that the agent has in virtue of having a social position within the structural process that leads to the injustice. As Young puts it, "Responsibility in relation to structural injustice derives . . . from being positioned in the structures in relation to others and acting within these positions" (Young 2011, 180). So Allocative might entail, for instance, that consumers have a political responsibility to put pressure on their local retailers to stock merchandise made under fair working conditions, and that they have this responsibility in virtue of occupying a privileged position within the global garment industry giving them the capacity to exert such pressure; the enablements and constraints that come with the social position of being a consumer affords one this capacity. Global garment companies have other political responsibilities, which may include the political responsibility to interact with the owners of the manufacturers more directly in order to improve working conditions, and they have these responsibilities in virtue of occupying a social position giving them the requisite powers.
According to SCM, it is only those who participate in a structural process leading to injustice who have a forward-looking responsibility to challenge the relevant process, so the notion of participation is central. Young is not very explicit about how and where to draw the line between those who participate and those who do not. In her discussions of the global garment industry, she takes consumers, the global garment companies, their shareholders, and others to be participants in the industry. But we might ask, is an employee in the public sector whose salary is partly financed by tax revenue from the global garment industry also a participant? Another question Young leaves unanswered is, what happens when agents cease to participate in a structural process leading to injustice? Are the agents then relieved of the political responsibilities they had before? For instance, would you shed the political responsibilities you had as a shareholder if you were to sell all your shares? Would you be free of the political responsibilities you had as a consumer if you were to stop buying and wearing clothes made in working conditions akin to slavery in countries like Bangladesh? By extension, does SCM enable us to avoid all political responsibilities to ameliorate structural injustice simply by our moving out to the forest and cutting ties with civilization? These questions are beyond the scope of this article. Instead, I will focus on whether clear-cut and present participation is plausibly seen as a necessary and sufficient condition of political responsibility for structural injustice.
Section III argues first that SCM, as Young presents it, implies that participation in a structural process leading to injustice is a necessary condition of having a responsibility to challenge the process. This claim is then questioned. The idea that nonparticipants could have a forward-looking responsibility to challenge a structural injustice in virtue of their having the capacity to do so is canvassed. Maeve McKeown counters this argument by reformulating SCM to state that everyone who participates in the global economy has a political responsibility to challenge all structural processes leading to injustice in this economy. However, as will be explained, this version of SCM also faces counterexamples.
Section IV questions the idea that participation in a structural process leading to injustice is sufficient for having a political responsibility, drawing attention to cases where participants lack such a responsibility. Young has several resources to counter this objection. Most important, she can refer to the discretionary and blameless nature of political responsibility for structural injustice. Still, cases remain that SCM cannot explain, even when we take the discretionary and blameless nature of political responsibility for injustice into account. I suggest that participants might lack political responsibility (all things considered) for structural injustice either because they are excused from this responsibility or because they have other obligations that outweigh it. Section V questions the idea that mere participation in structural processes leading to injustice ultimately grounds our political responsibilities for structural injustice. I argue that the activists' intuition may be grounded in other more fundamental considerations. We might mistakenly conclude that all participants in a structural process leading to injustice have political responsibility for the process because most of them satisfy at least one of the following criteria: they benefit from the structural injustice in question; they have causal or moral responsibility for the injustice; they have the capacity to change things (possibly in cooperation with others); or they have communal ties to those disprivileged by injustice. By directing attention to these other considerations, we can also explain why nonparticipants might have political responsibilities to challenge structural injustice.
Before considering whether participation is necessary or sufficient for having political responsibility (in sections III and IV) and whether there might be other factors than participation that decide who has political responsibilities (in section V), we need to consider another preliminary question.

II. What Is Feminist About Criticizing SCM?
The question has been asked, why publish an article that consists primarily of a criticism of SCM in a journal devoted to feminist philosophy? 5 The short answer is that the underlying aim of this article is not to criticize SCM, but to sketch a more accurate explanation of the phenomena that SCM seeks to explain.
Still, one might worry that a project with the aim of explaining the activists' intuition is not necessarily feminist. Perhaps neither this article nor SCM qualifies as a feminist project. Even though Young is undisputedly a feminist philosopher, it might be argued that her later works on responsibility are less characteristically feminist than her earlier works on experiences of the lived body (for example , Young 1980;1984). 6 There are, however, reasons to think that both Responsibility for Justice and this article qualify as feminist projects. First, both concentrate on cases concerning disprivileged women: the question of who has political responsibilities vis-à-vis injustices in the global garment industry is a question about gender justice in a globalized world. Those disprivileged in the global garment industry-the workers in the factories-are predominantly women. In addition, Young's other main example concerns the vulnerability to homelessness of many single mothers in the US (Young 2011). For the sake of brevity, I do not discuss this example here. However, I do discuss how you might have political responsibilities to challenge structural processes that lead to teenage prostitution in distant countries, a harm that typically affects young women.
Second, the aim of explaining the activists' intuition-that is, of explaining how agents have responsibilities to redress structural injustice-might very well be an inherently feminist project, and be so whether or not you concentrate on cases concerning disprivileged women. To investigate structural injustice is to investigate how people's social and political identities (for example, gender, ethnicity, class, sexual orientation, and so on) might combine to create unique modes of discrimination. For instance, in order to identify the group that is under systematic threat of being exploited by the global garment industry, we have to look at several factors, such as gender, class, and possibly ethnicity. In other words, to study structural injustice is to study intersectionality. And of course, the study of intersectionality is characteristic of third-wave feminism. Relatedly, Young argues that although the lived body is a useful concept for thinking about subjectivity and identity, feminist theory also needs an account of social structures and structural injustices: As I understand them, feminist and queer theory consist not only in giving account of the meaning of the lives of women and men in all their relational and sexual diversity. . . . Feminist and queer theories are also projects of social criticism. These are theoretical efforts to identify certain wrongful harms or injustices, locate and explain their sources in institutions and social relations, and propose directions for institutionally oriented action to change them. This latter set of tasks requires the theorist to have an account not only of individual experience, subjectivity, and identity, but also of social structures. (Young 2005, 19-20) 7 In elaborating a theory of who has political responsibilities for structural injustices, this article (as well as Responsibility for Justice) takes the existence of structural injustice for granted, and asks instead who has the responsibility to redress such injustice. Given that studying structural injustice is a feminist project, this project arguably qualifies as feminist.

III. Participation Is Not Necessary For Responsibility
Intuitively, a bystander could have a forward-looking responsibility to ameliorate structural injustice in the very same way that he or she might have a forward-looking responsibility to save a drowning child. Young rules this possibility out, however, because Limiting makes participation necessary for political responsibility for structural injustice. Young states, "[o]ur responsibility derives from belonging together with others in a system of interdependent processes of cooperation and competition" (Young 2011, 105). Here are three further considerations suggesting that she rejects the intuitive view just mentioned.
First, SCM is not applicable to nonparticipants. Remember that Allocative means that what agents should do to challenge structural injustice is decided in a rough and ready way by the power, privilege, interest, and collective ability agents have in virtue of their social position within the structural process creating injustice. So, agents who do not participate in a certain structural process, that is, do not occupy a social position within that process, will lack social position in virtue of which we can decide, roughly, what they should do in order to challenge the structural process that produces injustice. This is an indication that participation is a necessary condition for political responsibility for structural injustice on SCM.
Second, it is clear Young sees social connection as a necessary condition of political responsibility to challenge structural injustice. She contrasts the nation-state view (as we might call it) with the cosmopolitan-utilitarian view, and rejects both. The nation-state view is the idea that special responsibilities devolve onto individuals simply in virtue of their being citizens of the same nation-state. This can be contrasted with the cosmopolitan-utilitarian view that "moral agents have identical obligations to all human beings, as well as perhaps to some nonhuman beings. There is a moral imperative to minimize suffering, wherever it occurs" (Young 2011, 137). She considers Margaret Gilbert, Hannah Arendt, and John Rawls to be advocates of the first view, and Peter Singer and Peter Unger to be supporters of the second. Although Young's main point is that the nation-state view is mistaken, she also asserts that "[i]n contrast with the cosmopolitan-utilitarian position, some account needs to be offered of the nature of social relationships that ground claims that people have obligations of justice to one another. It is not enough to say that the others are human" (139). 8 Finally, Young affirms the idea that participation is necessary for political responsibility in her discussion of Arendt's view of collective responsibility-writing, for instance, that "Responsibility in relation to injustice . . . derives not from living under a common constitution, but rather from participating in the diverse institutional processes that produce structural injustice" (105).
Writers such as Martha Nussbaum, Maeve McKeown, and Robin Zheng also seem to agree that Young takes participation to be necessary for political responsibility. On Nussbaum's interpretation of SCM, it is a necessary condition of one's political responsibility that one is "causally embedded in processes that produce a problematic result" (Nussbaum 2011, xx). My interpretation of SCM differs from Nussbaum's. On hers, two conditions are necessary and jointly sufficient for political responsibility-the other being that one must be "in a position to assume ongoing forward-looking responsibility (in cooperation with others) for ameliorating those conditions" (xx). Nussbaum's interpretation ignores quite a few passages where Young indicates that participation alone is sufficient for political responsibility. I will present some of these passages in section IV.
McKeown also takes SCM to entail that participation is necessary for political responsibility for structural injustice. Thus she writes: "Political responsibility is a burden placed on every individual by virtue of acting in these new conditions because we cannot help but reproduce structural injustice" (McKeown 2018, 501). The new conditions in question consist in living in a highly interdependent, globalized economy where our actions have impacts on millions of geographically dispersed people.
In her elaboration of SCM, Zheng distinguishes between two varieties of responsibility: attributability and accountability. The former concerns the question of what makes a person's action count as exercises of her own agency, whereas the latter concerns the division of labor within a moral community. 9 Zheng clearly takes SCM to be an account of accountability: "As a conception of accountability, the SCM ascribes responsibility to individuals on the basis of the fact that they are members of the moral community, and that they participate in the structural processes that collectively determine their shared social conditions" (Zheng 2019, 121). The passage shows that Zheng takes participation in a structural process to be a precondition of being held accountable for the injustices produced by this process-that is, a necessary condition that must be satisfied by those who share a responsibility to challenge the structural processes in question.
Pace Young, participation in a structural process creating injustice does not seem to be a necessary condition for having a forward-looking responsibility to challenge this process. Consider McKeown's example of contract slavery in Thai brothels. Teenage girls from northern Thailand, Laos, Burma, and Cambodia are sold by their parents to these brothels (sometimes via brokers). As part of the deal, the girls sign contracts with the brothel owners that place them in debt bondage for their purchase price plus interest. The girls are then legally bound by this contract to work for the brothel owner until the owner decides that the debt is repaid. This growing practice is driven by various factors: local economic conditions following rapid industrialization have given working-class Thai male laborers greater disposable income, social attitudes permit prostitution, and there is an assumption that children owe a debt to their parents (McKeown 2015).
Assume now that we are aware of the situation, and that, without significant cost, we can obstruct the structural processes that leave these teenage girls vulnerable to enslavement in brothels. We might, for instance, be able to implement a combination of measures reducing the demand for prostitution, changing some crucial social norms and improving the economic situation for the relevant groups. Assume also that we have not participated in the structural processes in question. We have not bought sex from these girls. We do not own brothels. We have neither implemented nor reproduced the idea that children owe a debt to their parents, and so on. In such a case, do we have a forward-looking responsibility to challenge the structural processes that lead to injustice? Intuitively, our ability to mitigate the harms being done to the girls is sufficient to ascribe forward-looking responsibility. This means Limiting is incorrect: agents can be responsible for ameliorating harmful structural processes without participating in them.
One might object that this responsibility is not political since it is not shared with others. However, upon closer scrutiny, the responsibility can be seen to be shared. Each of us shares this responsibility with many others. For instance, we have to cooperate among ourselves and with others in order to implement suitable measures.
At this point, we can ask whether Limiting really captures the activists' intuition that Young wishes to explain. Activists who protest the injustices in the global garment industry no doubt themselves participate in this industry in virtue of being consumers, and therefore SCM can make sense of why those activists take themselves to have a responsibility to protest. Things are different for environmental activists who protest the trade in rhino horn. Since they do not participate in the relevant practice (I take it), are they mistaken in thinking they have a responsibility to address the problem? And how about activists who protest fur-farming? Usually, these activists do not participate in the structural processes of fur-farming. They do not wear fur, they do not buy fur, and so on. Are they also mistaken about their political responsibilities? It seems that SCM cannot explain why these activists take themselves to have a responsibility to protest injustice. SCM appears to be tailor-made to apply to structural processes in which everyone, or nearly everyone, participates.
There is an additional practical problem with the idea that participation is necessary for forward-looking responsibility. It is clear that many remediable structural injustices will remain unchanged, since those who can be held responsible as a result of their participation may refuse to remedy the process. For example, the Thai workers/customers, the brothel owners, pimps, and so on are unlikely to challenge the structural processes that result in the enslavement of young girls since they have financial or other interests in perpetuating them.
McKeown calls this the falling-through-the-gaps objection. She suggests that we can hold onto a slightly modified version of SCM. Her argument is that "in the contemporary world our political responsibility is limitless" (McKeown 2015, 282;2018, 500) because everybody is connected to everybody through the global economic system.

Call this
The Broad Limiting Principle of SCM: An agent A shares with others a forward-looking responsibility to challenge a structural process S that produces injustice if and only if A participates in the overarching structural process, O, and S is a subprocess of O.
Broad Limiting preserves Young's idea that in order for A to have a forward-looking responsibility to challenge a certain structural process S, there must be some kind of connection between A and those disadvantaged by S, but it drops Young's further claim that A must also occupy a social position within S. Without this further claim, Broad Limiting loses some of its appeal. It may seem plausible that I acquire a forward-looking responsibility to challenge the structural processes that drive scores of people to work in slavery-like conditions in garment manufacturing when I buy and use these clothes, but it seems much less straightforward to say that I have a forward-looking responsibility to try to prevent contract slavery in Thai brothels simply because I buy bread at my local supermarket. Still, this modified principle avoids the objections raised so far.
It should be noted that McKeown does not take participation in (or reproduction of) the global economic system to be a necessary condition of political responsibility for structural injustice. She is open to the possibility that "there are other forms of connection that generate political responsibility that Young has not thought of" (McKeown 2018, 485). Broad Limiting remains faithful to Young's idea that participation is necessary for political responsibility.
Broad Limiting does not identify all of the agents with a responsibility to challenge a certain structural injustice. First, there may be agents who do not participate in the global capitalist economy but nevertheless appear to share with others a responsibility to challenge some structural (sub)process within it. Consider an off-grid, self-sustaining society of scientists that is completely cut off from the global economy and has been so for many years. Assume that this society has developed some effective means-say, a technical invention or a social policy-that would significantly ameliorate harm within the global garment industry if it were implemented in a way that is judged to be beneficial by those disprivileged by this industry. Moreover, the means in question could be implemented without risking the creation of high costs for the factory owners, or for other agents such as global garment companies, and without significant cost to the scientists. In such a case, would not these scientists have a responsibility to ameliorate the injustices in question?
Second, there might be participants in the global capitalist economy who have a responsibility to challenge a structural injustice outside this economy. Imagine a community living on a remote island, cut off from the rest of the world, and assume there is some ongoing structural process producing injustice in this small society (as judged by those disadvantaged by the injustice). A research group could, conceivably, accidentally become aware of the injustices and realize that without significant cost to themselves they could intervene in the island society, making the structural processes more just. The research group will ensure the intervention is beyond reproach. For instance, postcolonial theorists will be invited to oversee the implementation process so as to avoid the interventionist mistakes of the past. The perspective of those disadvantaged by the structural process will also be considered, since they probably know better than anyone how the injustices can be effectively ended, and the aim is to intervene without violating tribal sovereignty. In such a case, the research group seems to share with others a forward-looking responsibility to disrupt the unjust, but remediable, structural processes in this isolated society even though the society is not part of the global economy. Moreover, they share this responsibility with others in the sense that no one member of the research group could successfully challenge the structural processes alone. Cooperation within the research group, and with the people of the island community and postcolonial theorists, is needed in order to successfully redress the structural injustice.
It may be suggested that these examples are too hypothetical. However, the second, at any rate, is not obviously far-fetched, given that current estimates are that there are still approximately one hundred isolated societies across the world (cf. Kane 2018). Anyway, there are other less hypothetical examples. We know that other parts of the world are cut off from the global economic system. What if, at limited cost to ourselves, we could ameliorate structural injustice in North Korea? What if some group, at no great cost, could have obstructed the unjust structures in Cambodia during the Khmer Rouge regime? Would the group have lacked the political responsibility to do so simply because the Cambodian regime had hermetically sealed the country off from the rest of the world?
These considerations indicate that participation is not necessary for political responsibility for structural injustice, and that we should revise Limiting.
Limiting*: An agent A shares with others a forward-looking responsibility to challenge a structural process S that produces injustice if and only if A participates in S.
I should point out that, as Young sees it, one has political responsibility to challenge structural injustice in virtue of participating in a structural process that leads to injustice. Therefore, in Young's view, one always shares political responsibility with all others who likewise participate in the structural process in question. In Limiting*, this position is amended. The position now is that one might also share political responsibility for structural injustice with nonparticipants. One might, for instance, share a political responsibility with nonparticipants who can do something about this structural process.
Some may deny that the capacity to ameliorate structural injustice can ground political responsibility to do so and therefore reject the counterexamples to SCM I have presented. Indeed this seems to be roughly what Young does when she argues against the cosmopolitan-utilitarian view. 10 There is, however, an argument for saying that adherents of SCM should not reject the idea that capacity can ground political responsibility: there is a tension between holding that the capacity alone could never ground political responsibilities for structural injustice, on the one hand, and holding that the capacity is the single most important factor in deciding what the agent should do in order to ameliorate such injustice, on the other.
To make this point clear, I need to make a short detour. All of the parameters in Allocative (power, privilege, and so on), as Young understands them, concern the agent's capacity to challenge structural injustice. That agents with power, privilege, or collective ability have a capacity to ameliorate injustice is probably quite clear. It may be less clear that agents with an interest in the injustices being ameliorated actually have the capacities required to ameliorate those very injustices. What capacities do the workers in the factories of the global garment industry have? And why would these capacities be relevant to their possession of political responsibility? Here, Young's idea is that those disfavored by a structural process have an epistemic advantage when it comes to determining how the processes should be changed, and they could and should communicate their knowledge to others with the capacity to reform the processes. In virtue of having these capacities, the workers have a responsibility to challenge the structural processes they participate in, Young argues. The question that prompted this short detour was: Why would capacity be the single most important factor in deciding what the agent should do in order to ameliorate structural injustice, while being completely irrelevant to the question of who is responsible? This question seems to lack a satisfying answer-something that ought to trouble anyone who found SCM attractive to begin with.

IV. Participation Is Not Sufficient For Responsibility
According to Young, participation in a structural process that produces injustice is not only necessary, but sufficient, for the acquisition of political responsibility for structural injustice. For instance, "All who dwell within the structures must take responsibility for remedying injustices they cause, though none is specifically liable for the harm in a legal sense" (Young 2011, 105). The idea that participation is sufficient for political responsibility is perhaps most clearly spelled out here: "No one who participates in processes that produce structural injustice is exempt from responsibility to join with others to change those structures" (153). Writers such as Zheng and McKeown would agree that Young sees participation as a sufficient condition of political responsibility to challenge structural injustice. Zheng, for instance, says in her summary of SCM: "As a result of our causal participation in unjust processes, we are each assigned . . . the forward-looking task of collectively organizing to transform social structures" (Zheng 2019, 123; emphasis mine). On McKeown's reading, Young is claiming that one has a forward-looking responsibility to challenge a structural process if one reproduces this process in one's own actions. As McKeown puts it, "On the social connection model, we find all agents who reproduce the practice through their actions politically responsible to change the structures" (McKeown 2018, 498).
Young's sufficiency claim has been challenged. Carol C. Gould objects that SCM comes close to encouraging victim-blaming. It states that everyone who participates in a structural process creating injustice has a responsibility to challenge this process, 11 and therefore that those disprivileged by the process have that responsibility (including, for instance, workers in the factories of the global garment industries) (Gould 2009, 202-03). That SCM comes close to permitting victim-blaming is further emphasized by Young's claim that having an interest in rectifying the structural injustice gives one a greater political responsibility to do so.
It can also be objected that the idea that everyone who participates in a structural process creating injustice has a political responsibility for this process is excessively demanding-that SCM places unacceptably heavy burdens on some agents. It seems excessive to demand, for instance, of single parents holding several low-paid jobs that just about make ends meet, to also work for changing the structural processes of the global garment industry. It might also be said that it is excessive to demand that affluent middle-class people should take up their political responsibilities for all of the structural processes that produce injustice in which they participate. In modern society, most of us participate in an extensive range of harmful structural processes: the global garment industry, structural processes leading to global warming, the fishing industry, structural processes leading to gender or race inequalities, to mention just a few. Although it seems reasonable to expect an affluent person to work for change in some of these processes, it might still be considered excessive to demand that he or she must take steps to challenge all of them. (This problem is even more pressing if we accept Broad Limiting.) Young's most important response to the charge that SCM is victim-blaming and too demanding turns on her idea that political responsibility is discretionary. She writes, "responsibilities carry considerable discretion; one must carry out one's responsibilities, but how one does so is a matter of judgment according to what the responsibilities are for, the capabilities of agents, and the content of action' (Young 2004, 379;emphases in original). We might add that the discretion here gives us some slack not just in how we take up our political responsibilities for structural injustice, but also when we do so.
Responding to the complaint about victim-blaming, Young supplements the idea that political responsibility is discretionary with something we have already have touched on, namely the idea that those disadvantaged by a structural process often can in fact do something to challenge that very process. Most important, they often have an epistemic advantage: It is they who know the most about the harms they suffer, and thus it is up to them, though not them alone, to broadcast their situation and call it injustice. Unless the victims themselves are involved in ameliorative efforts, well-meaning outsiders may inadvertently harm them in a different way, or set reforms going in unproductive directions. (Young 2011, 146) Strengthening Young's point, we might add that the workers could also improve the situation by other means-say, by joining or, if necessary, forming a union. The situation of garment-factory workers in Cambodia has significantly improved in the last few years (from terrible to just really bad) mainly as a result of a successful union campaign.
Although Young's argument shows that more agents have political responsibilities to challenge structural injustice than we might initially have expected, it does not show that all participants in a structural process that leads to injustice have those responsibilities. Quite a few workers cannot feasibly broadcast their situation, or join or form a union. The risks of harassment, unemployment, and ensuing poverty for themselves and for their families are simply too high. This point becomes even more obvious when we consider enslaved teenagers in Thai brothels. Although some prostitutes in Thailand have a certain amount of autonomy, others are locked up and risk being seriously hurt if they refuse to work (cf. Bales 2012). The opportunities the latter have to broadcast their situation (or unionize) are clearly very limited. Similarly, low-paid single parents might lack a genuine opportunity to take action aimed at ameliorating the injustices of the global garment industry, and it may not be practicable for affluent middle-class people to address all the structural processes that produce the injustices in which they participate.
At this point, proponents of SCM might suggest that the discretionary nature of political responsibility means not only that it is up to us what to do, but also that it is up to us when to do it. A worker who makes clothes in Cambodia and who for many years has had no real opportunity to broadcast her situation or join a union might get an opportunity to do so in due course. As an affluent Westerner, I cannot shoulder all my political responsibilities for structural injustice now, but I can shoulder some of them now and some of them later on in my life. Still, sometimes people will presumably be unable to take up (all of) their responsibilities even with time-unable now and unable later. Individuals may die before getting a real opportunity to address each and every structural process producing the injustices in which they participate. Are they blameworthy for this failure?
Young has a further reply readily at hand. She argues that SCM cannot involve victim-blaming, because it does not involve blaming. 12 This argument does not concern only victims (for example, workers in the global garment industry); it generalizes. If SCM does not involve blaming, no one who fails to shoulder their political responsibilities for structural injustice can be blamed on the basis of a failure to respect SCM. However, the idea that SCM does not involve blaming has not stood unchallenged. In the foreword of Responsibility for Justice, Nussbaum argues that an agent might be blameworthy if he or she puts off taking up a political responsibility for structural injustice for too long, and others have objected to SCM on similar grounds (for example, Reiman 2012; Barry and Ferracioli 2013).
Young could concede this point without substantially changing her account. Her primary concern is to show that political responsibility to challenge structural injustice does not originate in blameworthiness (instead it originates in participation), and this concern is not threatened by conceding that we can be blamed for not taking up our forward-looking responsibilities. In addition, the concession would be in line with Young's analogy between political responsibility and role responsibility. For example, if companies are in financial trouble, CEOs have a responsibility to manage this problem in virtue of their role, and they could be blamed if they fail to take up this responsibility. This holds even if the financial trouble is not their fault. They will have this responsibility even if, for instance, they are new to their posts, or the financial trouble is due to unexpected market fluctuations. However, if Young were to concede this point, she would lose one counter-argument to the charges that SCM is victim-blaming and overly demanding. She might then be compelled to admit that participation is not sufficient for political responsibility.
Actually, it is not entirely clear that the claim that SCM does not involve blaming at all leaves SCM immune to the objection that it is excessively demanding and victimblaming. Even if Young insists that political responsibility for structural injustice is a completely blameless matter, she does stress that agents who do not take up their political responsibilities can be criticized: they "can and should be criticized for not taking action, not taking enough action, taking ineffective action, or taking action that is counterproductive" (Young 2011, 144). At this point, it would be useful to have a better understanding of what criticizing involves, if not blaming, and its role in SCM. Unfortunately, Young's discussion of what criticizing amounts to, and why it is fundamentally different from blaming, is underdeveloped and sometimes ambiguous. My best guess is that she would characterize criticizing as similar to blaming, but without the presupposition of fault in the person being criticized-rather like constructive criticism. For instance, she writes, "We must call one another to account if we are to change the structural processes, but we can do so without attributing malevolent intent to, or hurting, the persons we criticize" (165). Think of the task of commenting on a paper by a student. You might say something like "Do you know about so-and-so's work? No? You should read it and then rewrite the second section of your paper." In that way, you hint that the second section is inadequate and suggest how it can be improved, but you do so without blaming the student.
If this is what Young has in mind, we can ask why workers who lack the genuine opportunity to challenge the structural processes that create injustice they participate in would be subject to criticism. Equally, why would affluent people who put all the energy and resources they can spare into challenging structural injustice be subject to criticism? Why would the enslaved teenagers in Thai brothels be subject to criticism for not shouldering their political responsibility? Even if we put the critique in the most constructive manner, it will still be misplaced and inappropriate. Similarly, it would be inappropriate to demand of a student to read a certain work if it would be impossible for her to read it. Hence, to avoid the objections that SCM is victim-blaming and too demanding, it does not help Young to say that those who fail to take up their political responsibilities should be criticized, not blamed. Criticizing victims does not seem much more defensible than blaming them.
Limiting can be modified in at least two ways in an attempt to meet the objection that participation in a structural process that produces injustice is not sufficient for political responsibility. One is to add conditions under which an agent is excused from political responsibility; the other is to temper its demandingness by understanding political responsibilities as pro tanto obligations instead of all-things-considered (but discretionary) ones.
McKeown favors the first strategy and argues that some agents might be excused from political responsibility for structural injustice if "the conditions are such that they are unable to act upon that responsibility" (McKeown 2015, 276). This strategy is well suited to explain why workers in the global garment industry who face harassment, unemployment, and poverty if they broadcast their situation or join a union lack a forward-looking responsibility to challenge the structural processes of the global garment industry. Because of their situation, their responsibility is disabled, silenced, or overridden. Similarly, many enslaved girls in Thai brothels lack a responsibility to challenge the structural processes that make many young girls vulnerable to enslavement simply because they lack the ability to act upon that responsibility.
McKeown's strategy seems less well suited to explain some other situations. The workers in the global garment industry may be able to join a union without serious cost to themselves, but they refrain from doing so because this would have severe repercussions for their families. In joining a union, they might make it harder (for themselves) to get a job, and thus earn less. It may then be the case that although they can still sustain themselves, they do not have enough income to sustain their families. In such a case, it seems appropriate to say that they have a responsibility to ameliorate injustice, but that this responsibility is outweighed by their responsibilities toward their families. They are not excused from political responsibility, but they are justified in caring for their families instead of joining a union.
This brings us to the second strategy. Instead of introducing excusing conditions, we could abandon the idea that we must carry out the political responsibilities to challenge structural injustice we have. Instead of thinking about political responsibilities as all-things-considered requirements (albeit discretionary ones) we could understand them as pro tanto obligations. Pro tanto obligations are weighty but not conclusive moral reasons; they play an important role in deciding what you should do, all things considered, but they can be outweighed.
It is a neat suggestion, but this second strategy seems less straightforward in cases where the first strategy looked right. It seems odd to say that the enslaved teenage girls in Thai brothels have a pro tanto obligation to challenge the structural processes that oppress them, but that this obligation is outweighed by the obligation to avoid being (even more) seriously hurt. It seems rather that they are excused from this responsibility altogether. Therefore, we appear to need both strategies. The first is effective in explaining cases where agents cannot reasonably act on their political responsibility; the second is better at explaining cases where agents could act on their political responsibility but have an even stronger obligation to further some other goal.
If we modify Limiting to accommodate these considerations, we get something like: Limiting**: An agent A shares with others a forward-looking responsibility, that is, a pro tanto obligation, to challenge a structural process S that produces injustice if and only if A participates in S, unless A lacks a reasonable opportunity to act upon that responsibility.
The notion that the agent might be excused from a political responsibility to challenge structural injustice, or be justified in not taking it up, explains (away) certain aspects of the alleged discretionary nature of political responsibility. We can use Limiting** to explain why, at one point in time, we are not required to do something about a structural process that produces the injustice in which we participate, even though, at another point in time, we are required to act. Suppose a worker producing clothes under slavery-like conditions in a factory at t has no real opportunity to do anything about the structural processes of the global garment industry. Limiting** would entail that at t, she lacks a political responsibility to ameliorate these injustices. However, if later, at t*, her situation were to change, so that she now has the opportunity she previously lacked, Limiting** may well entail that she has a responsibility to do something about the situation. A final consideration is that other groups of participants in structural processes that produce injustice do not seem to have a forward-looking responsibility to challenge the processes in which they participate. Examples include people suffering severe mental illness and young children. Are they responsible for challenging structural processes in the global garment industry in virtue of their participation in these processes? McKeown responds to this question by saying that the severely mentally ill, the very young, and others might be exempted from political responsibility on the grounds that they do not satisfy the conditions of full agency (McKeown 2015). I think she is correct in this, but it would reach too far in this context to discuss these conditions more fully. 13

V. Alternative Grounds Of Political Responsibility For Structural Injustice
Since participation is neither necessary nor sufficient for having (all-things-considered) political responsibilities for structural injustice, an agent's mere participation in a structural process that produces injustice does not reveal anything about this agent's responsibilities to challenge this process. For this reason, we shall need to look for other explanations of the grounds of those responsibilities. Following writers like David Miller and Daniel Butt, I suggest that an agent, A, might have political responsibilities (or remedial responsibilities, as they call it) if A has a causal or moral backward-looking responsibility for injustice, if A benefits from injustice, if A has the capacity to ameliorate injustice (at reasonable cost), or if A has community ties with those who suffer from the injustice (Miller 2001;Butt 2007). One can be led to conclude that participation grounds political responsibility for structural injustice by the fact that participants in a structural process that produces injustice typically satisfy at least one of these conditions. But this is an error-at least, that is what I want to suggest. Participants might of course have political responsibilities, but the suggestion is that if they do, they do so in virtue of satisfying one or more of these other conditions. For the sake of brevity, I will elaborate the suggestion rather sketchily. I will also omit any discussion of capacity, since I treated this at length earlier in this article.
To begin with, those who participate in structural processes that lead to injustice usually reproduce these processes and thereby causally contribute to the continuation of the injustice. Thus, when you buy a new pair of shoes made abroad in working conditions contravening human rights, you contribute to the revenue of the company that sold these shoes. This company might in turn take this revenue as an incentive to order more shoes, and thereby you will have contributed to even more workers being exploited in the future. Since those who participate in structural processes that produce injustice cause this injustice to continue, we might think that they have special duties to remedy these injustices.
Should we resist the idea that the buyer of the shoes causes injustice? We might if, like Miller, we understand causation along the lines suggested by H. L. A. Hart and Tony Honoré: when we say that C caused E, we single out C as one particularly salient antecedent condition for E among many, perhaps because C is unusual or intentional (Hart and Honoré 1985). Since a single purchase of a pair of shoes is not on its own a salient antecedent condition for the injustice of the global garment industry, such a purchase does not (in the sense explained by Hart and Honoré and mobilized by Miller) cause this injustice. Young explicitly rejects the idea that participants in a structural process that produces injustice cause this injustice, but she seems to have a different understanding of causation. She argues that agents participating in structural processes that produce injustice do not cause harm since "it is not possible to identify how the actions of one particular individual, or even one particular collective agent, such as a firm, has directly produced harm to other specific individuals" (Young 2011, 96). The idea is instead that individual agents reproduce certain structural processes that in turn cause injustice.
On other understandings of causation, a single purchase of a new pair of shoes could cause injustice. Thus, following David Lewis, we might argue that A's purchase makes a difference to the manner and timing of the structural processes that lead to an injustice (Lewis 2000)-that is to say, it influences the structural processes-and that since causation is the transitive closure of influence, A's purchase is also one of the causes of the injustice: it influences the revenues of the garment company, and these revenues in turn influence future orders of clothes, which in turn influences how and when clothes are made in the factories. Alternatively, following writers like Wesley S. Salmon and Ned Hall, we could say that A is involved in the causal process leading to future injustice when A buys clothes (Salmon 1978;Hall 2004). So, whether we understand causation as the transitive closure of influence or in terms of causal processes, it will turn out that in buying clothes, A (together with others) causes injustice. This is not the place to assess the merits and demerits of these different accounts of causation, nor to assess which account of causation is relevant for assigning forward-looking responsibilities to agents. 14 I wish merely to suggest that some accounts of causation can be used to explain the intuition, had by some, that we causally contribute to future injustice (or risk doing so) when we buy clothes made under slavery-like working conditions, and to point out that these accounts of causation might ground the idea that, as buyers, we have a pro tanto obligation to ameliorate the relevant injustice because we are causally responsible for it.
We can now turn to moral responsibility as a ground for political responsibility. As Miller understands it, moral responsibility essentially involves an appraisal of an agent's conduct-for example, whether the agent intended, or foresaw, a harmful outcome that he or she brought about, whether the agent negligently failed to take steps to prevent some harm from occurring, and so on (Miller 2001). It seems that at least some agents participating in structural processes that produce injustice fit this description. Powerful agents such as multinational garment companies, for instance, intentionally and knowingly choose to perpetuate structural injustice, and therefore they might be morally responsible for this (cf. Gould 2009;McKeown 2015;Zheng 2019). In consequence, they might also acquire a forward-looking responsibility to set things right. Young's view that no participants are morally responsible in a backward-looking sense for structural injustice seems mistaken.
Next, quite a few of us who participate in structural processes that produce injustice benefit from them, and this might also give us political responsibilities. As Butt puts it: "by benefiting from an act of injustice, one can acquire obligations toward the victims of that injustice" (Butt 2007, 139). When I buy clothes made under slavery-like conditions, I pay less than I would have paid had these clothes been made in more humane working conditions. This may mean that I acquire obligations to those who made the clothes. In a similar way, the owners of the factories where the clothes are made, the global garment companies and their shareholders, usually benefit from injustice, and might therefore acquire political responsibilities to challenge the structural processes leading to injustice.
Last but not least, we who participate in the structural processes of the global garment industry might have responsibilities to the workers in the factories where our clothes are made because we belong to the same community. You might wonder in what sense a consumer of clothes in Europe or the US belongs to the same community as a worker in garment factory in, say, Bangladesh. When Miller presents the idea that those who have communal ties might also have special remedial duties toward one another, he understands community "in a fairly loose sense to capture special ties of various kinds such as those that exist within families, collegial groups of various kinds, nations, and so forth" (Miller 2001, 462). He continues: "The claim is that when people are linked together by such ties, . . . they also ( justifiably) see themselves as having special responsibilities to one another, responsibilities that are greater than those they have toward humanity at large" (462). It would not be stretching it too far to say that all those who participate in the structural processes of the global garment industry belong to the same community in this loose sense. In fact, Miller's claim here is strikingly similar to a claim that Young makes when she argues against the nation-state view: she says that social ties-even across borders-might ground special responsibilities that are greater than those we have to humanity at large. The most striking difference between Miller's and Young's positions is perhaps that Miller claims that those who have community ties actually (and justifiably) see themselves as having special responsibilities to one another, but Young's point is rather that, though we might not regard ourselves as having any social connections with those who are disprivileged by the global garment industry, we should recognize that we do have such connections, and therefore also that we have special responsibilities to them. 15 To sum up, agents who participate in structural processes leading to injustice typically satisfy one or more of the following conditions: either they causally contribute to injustice, or they are morally responsible for injustice, or they have the capacity to challenge the processes that lead to injustice, or they benefit from injustice, or they have community ties (in a loose sense) with those suffering from injustice. This might make it seem that participation in structural processes that produce injustice gives us a forward-looking responsibility to rectify those processes. However, upon closer scrutiny we see that it is the more fundamental considerations that confer this responsibility on us.
VI. Participation Is Neither Necessary Nor Sufficient For Political Responsibility I have argued that SCM must be rejected because participation in a structural process that produces injustice is neither sufficient nor necessary for having a forward-looking responsibility (all things considered) to challenge this process. Although SCM seems to account for the activists' intuition that Young seeks to explain in some cases, it cannot do so in all cases: it cannot explain why activists see themselves has having remedial responsibilities for structural processes creating an injustice that they do not themselves participate in (think of activists taking action against contract slavery of young girls in brothels, rhino-poaching, and so on). In addition, SCM is too demanding and cannot be wholly disentangled from victim-blaming, because it entails that everyone who participates in structural processes leading to injustice must take steps to obstruct these processes, even if they have no real opportunity to do so. Since participation is neither necessary nor sufficient for political responsibility to challenge structural injustice, we need another-more reliable-explanation for the activists' intuition. I suggest that we should think of political responsibility for structural injustice as something involving pro tanto obligations that stem from a plurality of features of the agent, including being causally or morally responsible for the relevant injustice, benefiting from the injustice, having the capacity to ameliorate the injustice, or having communal ties to those disprivileged by the injustice. This way, we can explain the activists' intuitions, not only about who has responsibilities for challenging the structural processes leading to injustice in the global garment industry, but also more generally about who has responsibilities to challenge a particular structural process leading to injustice.