A Journal of Philosophy, Applied to the Real World

Offsetting Class Privilege

University of Sheffield


The UK is an unequal society. Societies like these raise significant ethical questions for those who live in them. One is how they should respond to such inequality, and in particular, to its effects on those who are worst-off. In this article, I’ll approach this question by focusing on the obligations of a particular group of those who are best-off. I’ll defend the idea of morally objectionable class-based advantage, which I’ll call ‘class privilege’, argue that class privilege can be non-culpable, and put forward an account of the obligations those with class privilege have. My main claim will be that those with class privilege have obligations to ‘offset’ their privilege, in something like the same way high emitters have obligations to offset their greenhouse gas emissions.



The UK is an unequal society. Take income, for example. At one end of this spectrum of inequality, there are people who cannot secure employment at all, who are either destitute, or who rely on benefits of £72.40 per week.1. There are people on apprenticeships earning a weekly wage of £158.40, and there are people who work routine jobs for a minimum weekly wage of £345.60.2. At the other end, there are around 2.93 million people earning more than £1,117 per week, a figure which represents the bottom of the top 5% of earners in the UK (Jenkins 2015, p. 4). Or take occupation. Some people have jobs that come with low levels of social prestige and recognition, while others have jobs that come with high levels of social status and prestige. A poll in the United States revealed Banker, Actor, and Real Estate Agent to have the least prestige, and Firefighter, Scientist, and Teacher to have the most prestige, of the occupations surveyed (HarrisInteractive, 2007).

Or take education. 26% of the UK’s jobs require a degree, but most of the UK’s population do not go to university―the percentage who do is between 27.2% (based on data from the Office for National Statistics, 2013) and 40.2% (based on data from the Annual Population Survey) (Ball, 2013). At the high-school level, only 7% of the UK population goes to private schools, but graduates of private schools make up 75% of the UK’s judges, 70% of the UK’s finance directors, 53% of the UK’s journalists, 45% of the UK’s top civil servants, and 32% of the UK’s Members of Parliament (Monbiot 2010). Finally, the children of higher professionals are three times more likely than the children of people in routine work to get five good GCSE grades (ibid). Or take social capital. Some people have extended networks of friends, colleagues, and contacts in influential social positions. These people can be called upon for favours, or to assist in difficult times such as transitions in employment, or to alleviate financial pressure. Others have smaller networks, consisting of people in non-influential social positions.

All of this is hardly surprising from a descriptive perspective, given the country’s long feudalist history. But societies like these raise huge numbers of ethical questions for those who live in them. One such question is how we should respond to such inequality, and in particular, to its effects on certain members of the society. Many will be troubled by the situation of those at the bottom end of this spectrum. The broadest version of the issue I’m interested in here is whether there is anything that people in such societies owe to each other, as a result of these inequalities. Whether they do―and what it is they owe if they do―depends on a number of things.

Chief among them is whether some people are culpable in the fact of this inequality and its effects. Culpability is usually assigned on the basis of a person’s intentionally (or at least forseeably) doing harm. So there would be culpability for class privilege if, for example, some of the people at the top have intentionally made it the case that some of the people at the bottom are at the bottom. If there is culpability, either for the inequality itself or for the fact that certain people end up at the bottom, then much of the story about what some owe to others can be told in terms of the obligations of the culpable to make reparation for, or pay compensation to, those who they have harmed. (Or those who have been harmed as a foreseeable result of what they have done).

If there is no culpability, an answer to the question of what people in such a society owe each other may yet be given in other terms. Some might owe others assistance, on the simple grounds that some have the resources to provide assistance, and others need assistance. Or we might all owe particular things to anyone with whom we share a particular kind of association, such as the political association residents of the UK share with one another.

In this paper, I’m interested in pursuing a very different way of telling the story, namely in terms of benefiting. I want to ask specifically about the obligations of those at the top. This is to take seriously the intuition defended in Daniel Butt’s paper ‘On Benefiting From Injustice’, that beneficiaries of injustice3. have obligations that are stronger or more extensive than those that everyone has in virtue of either shared association, or capacity to provide assistance (Butt 2007; see also Barry 2003).

That will require doing three things: (i) defending the idea of morally objectionable class-based advantage, which I’ll call class privilege4. (Section II below), (ii) arguing that class privilege can be non-culpable (this keeps the story about who has what obligations in the domain of beneficiaries rather than shifting it to the domain of redress for culpable harm) (Section III), and (iii) putting forward an account of the obligations those with class privilege have (Section IV). After that, I’ll address an important objection to do with people being complicit in their own disadvantage (Section V). I’ll argue that class privilege is best understood as a failure of social mobility; that there are many class-privileged people who are not culpable in the fact of class privilege; and that nevertheless the class-privileged ought to ‘offset’ their privilege by taking on cost to undermine the current failures of social mobility. This is in just the same way that high emitters of greenhouses gases ought to offset their emissions.

One caveat before that. It is possible to give a range of different answers to the question of what the class privileged owe, because it is possible to approach the question in a more and less utopian way. The best moral answer to class privilege might be the dismantling of the very fact of class-based differences between people. A good answer might be to ensure that class position is decided fairly, for example, by lottery, or by choice, or by effort alone. I do not claim to be defending the best moral answer in this paper. I claim to be defending a good answer, one which is sensitive to changes that might be politically feasible as well as ethical.

Class privilege

I. Class, Class Advantage, & Class Privilege

When philosophers talk about concepts―like ‘class’―they typically try to capture as much of the ordinary understanding of those concepts as possible, although sometimes what they want to do with those concepts will lead them to propose revisions. A good place for us to start, then, is with the way ‘class’ is ordinarily understood. The question we’re starting with is ‘what is class’?

Traditionally, a British person was understood as belonging to one of three social classes: Upper Class, Middle Class, or Working Class. The upper classes were the aristocracy, the middle class were landowners, and the working class were those engaged in manual work. A recent BBC survey with over 160,000 respondents collected information about economic, cultural, and social capital, and concluded that there are now seven social classes in the UK: Elite, Established Middle Class, Technical Middle Class, New Affluent Workers, Emergent Service Workers, Traditional Working Class, and Precariat (Savage & Devine 2011).5. The UK Office for National Statistics uses a division based solely on occupations, and they present an eight-class, five-class, and three-class grouping, commenting that only the three-class grouping should be taken to be hierarchical. UK Geographics presents a six-class occupational grouping, made on the basis of Occupational Code, Employment status, Qualification, Tenure, and Full-Time Equivalent, see (UKGeographics 2014)).

These are four different ways of understanding class, all of which include occupation, some of which include much more. They distinguish between 3 to 8 class groupings respectively. They create both relations and hierarchies between groupings, because any given group stands in a particular relation to another, and the groups can be ranked in order of which has more and which has less of a particular good. For example, the Elite on the BBC understanding are at the top of the hierarchy when it comes to the possession of economic, cultural, and social capital, and they are better-off in relation to each of the six remaining class groups. Next, what is ‘class advantage’? Advantage is a simple matter of being better-off. Only the class group at the bottom of the hierarchy―for example the Precariat on the BBC understanding―lacks class advantage. The rest are better-off than at least one other group. Groups in the middle of the hierarchy will be advantaged relative to some and disadvantaged relative to others. What we’ve got so far is a story about class that permits an understanding of class advantage. What we’re missing is a moral dimension. Does it matter if some classes of people are advantaged? Let me first explain why that’s missing, and then go on to extend these initial suggestions in a way that makes class-based advantage morally objectionable.

If we care about equality per se, then these facts about social hierarchies and relations of advantage and disadvantage will be enough to start talking about what these people owe to each other. But there is a strong tendency in contemporary liberal political philosophy to think that some inequality can be permissible. Some defend this as being necessary to incentivize greater productivity, creativity, or entrepreneurship in society, which in turn can ‘trickle down’ to make the worst-off better off; some see it as an appropriate response to social contributions that require different levels of skill, training, effort, stress, or responsibility. So long as inequality is permissible, then there’s nothing wrong with the mere fact that there are social classes. So we need more than just the story about what groups there are, and which people are in which group. We need something that suggests unfairness or injustice in the fact that certain people are in certain groups.6.

Of course, we can’t go and look into the backstory for every person in the UK, to check how each has ended up in the group they’re in, and whether this history involved any unfairness or injustice. But we can check for unfairness or injustice in a more general way. For example, we can look at data on the social distribution of particular things―all and any of the things discussed earlier, like occupation, income, social capital, cultural capital, or indeed property holdings, honorifics, education―and check whether these things cluster in an improbable way according to particular traits or features. We should expect them to cluster according to each other: there will tend to be a correlation between education and income, for example, or between income and property holdings. What we shouldn’t expect is them to cluster according to some feature of a person that should be irrelevant from a moral point of view.

This kind of approach is often taken when it comes to features like race and gender. For example, we might make a graph showing the distribution of income between people in the UK, and then we might check this distribution for clusterings by gender, which is to say, whether there are more men then women in the top income categories, and more women than men in the bottom income categories. If we observe this clustering, there might, of course, be a perfectly reasonable explanation. For example, it might turn out that more women than men are working part-time, and the higher-income jobs require a full-time commitment; or that more women than men have chosen occupations that come with lower levels of remuneration; or that the highest-income jobs are those that were historically the most exclusionary of women, and this has resulted in there being more women in junior positions (the men who are now in the most senior positions were junior at a time where there were few if any women in the companies).

As implausible as these explanations might be in the case of gender, the more general point is that there can be such explanations. While distributions of particular things might look at first glance to be clustered in a problematic way, this will not always turn out to be morally objectionable. The problem that remains is to say what the traits or features are that we look for when we want to check whether a social distribution of something like social capital reveals improbable clusterings. If the distribution reveals clusterings by gender, we might say there’s gender privilege; if it reveals clusterings by race, we might say there’s race privilege. What would a clustering by class look like? In other words, what is class privilege?

There are two very different ways to answer this question. The first involves taking a cue from research into other forms of social discrimination. Familiar forms include discrimination on the basis of race, and discrimination on the basis of gender. There are particular social markers and social signals of race and gender, and these can trigger stereotypes and generalizations about race and gender groups. A person who has negative beliefs about women may encounter a particular individual, read off her appearance that she is a woman, and then apply those negative beliefs to her. For example, a man might believe that single women over 30 are desperate to marry and have children, meet a woman who signals sexual interest in him, and decide that what she wants from him is a tenure-track to marriage and family. They can also result in negative treatment, for example testimonial injustice where what a person says is less likely to be believed (Fricker 1999). Are there negative beliefs―stereotypes and generalizations―about some or all of the cluster of features we’ve identified as determining or relating to social class?

It’s clear that there are. Owen Jones catalogues a number of these in his book Chavs, with its revealing subtitle ‘The Demonization of the Working Class’ (Jones [2011] 2012). The most pervasive of these is perhaps that instead of extending sympathy to those in Working Class groups whose industries were destroyed under Thatcher, leading to high levels of unemployment and desperation, many in the Middle Class groups believe that unemployment or dependence on benefits is a preference. But there are many different features that might act as markers or signals of class. Consider employment, where initial selection of candidates works through CVs. Both names and addresses may signal one’s class group, as they have been found to do for racial groups.

In their (2004) experiments on racial discrimination in the United States, Marianne Bertrand and Sendhil Mullainathan found that job applicants listing addresses in whiter, more educated, or higher-income neighbourhoods had a higher probability of being called to interview (Bertrand & Mullainathan 2004, p. 1003). They also found that those applicants with typically White names (‘Allison’, ‘Brad’) were 50% more likely to be called to interview than those applicants with typically African American names (‘Aisha’, ‘Darnell’) (ibid, p. 998 & p. 1012). In a field experiment of UK employers’ social class discrimination, Michelle Jackson (2009) found that applicants with a name, school type, and interests associated with the social elite were more likely to receive a reply from employers, and that the single feature that made the most difference was name (Jackson 2009, p. 680 & p. 681).

Or consider face-to-face interactions. The following markers may all act as class markers: conventions of appearance (e.g. clothing, grooming), regional dialect, vocabulary, etiquette, and ability to converse on particular topics. If a person has negative beliefs about ‘the poor’, or ‘the working class’, and meets a person who has one or more markers of being in these groups, then she may apply her negative beliefs to this individual. This suggests that in just the same way as there can be gender- or race-based discrimination, on the basis of harmful stereotypes and generalizations about gender or about race, there can be class-based discrimination, on the basis of harmful stereotypes or generalizations about class groups.

Jean-Claude Croizet and Theresa Claire conducted research designed to test whether the idea of ‘stereotype threat’―namely that a person can be caused to underperform merely by being made aware of stereotypes that predict members of her social group to underperform, demonstrated in the case of both race and gender (Steele 1997)―can be extended to the case of class (Croizet & Claire 1998). They showed that it can. (Notice that this can go in both directions; to again use the BBC understanding, those in the Precariat might have negative beliefs about the Elite, and apply these to particular individuals on the basis of markers of appearance or social interaction that signal membership in the Elite. Stereotypes are bad for everyone, but the effects of class stereotypes are much worse for those at the bottom than those at the top).

The second way to answer the question looks instead at determinants of class position, rather than markers of class position. The way Croizet & Claire measured class is interesting for us. They equated class with socioeconomic status, and grouped students into either high or low socioeconomic status groups. But they did the groupings by accessing the students’ administrative records, and looking at the occupation of the parent who is the main provider for the student’s family. Students assigned to the low socioeconomic status condition had parents who were manual labourers, unemployed, and in administrative jobs, while students assigned to the high socioeconomic condition had parents who were managers, professionals, researchers, and college professors (ibid, p. 590).

This gives us a feature we might use: the occupation of a person’s parents. If greater numbers of the people with high social capital or high occupational prestige have a parent with a high socioeconomic status job, and this correlation cannot be explained away, then we might well have morally problematic class-based advantage, namely, class privilege. This correlation is also demonstrated in a study tracking the relationship between fathers’ incomes at the time their sons are born, and sons’ incomes at age 30. Fathers’ incomes are highly predictive of sons’ incomes in the UK (see discussion in Pickett & Wilkinson 2010, p. 160 & p. 289). One study found correlations of between 0.4 – 0.6 for fathers’ and sons’ incomes, and between 0.45 – 0.7 for fathers’ and daughters’ incomes, where 1.0 is complete determination of one by the other (see discussion in Aldridge 2004, pp. 20-27; Paxton & Dixon 2004).

Thus we have two different ways of understanding class privilege, both which capture intuitive features of our ordinary understanding of class, and both which involve unfairness or injustice; the first by way of explicit or implicit discrimination on the basis of class stereotypes and generalizations, the second by way of political and institutional obstacles to social mobility. The unfairness or injustice is what takes us from class and class advantage to class privilege. A person is privileged when she has markers which lead others to treat her favourably, or equivalently, lacks markers that lead others to treat her unfavourably. She is privileged when she has a parent in a better-off class group, which predicts that she herself will end up in a better-off class group.

II. Social Group Privilege, and Group Member Privilege

Are you privileged? You can refer to one of the sources mentioned earlier in this section, and depending which you choose, use your occupation, income, or other information to determine your social classification.7. How do you know whether that group has privilege, and whether you have privilege as a member of that group? The second part of this question is more difficult than the first. You know the group has privilege when it’s one of the better-off groups, and when rates of relative social mobility are low, because that means a major part of the explanation for why you’re in that group is that one or both of your parents had a certain occupation, or income, or social status, etc.

(There’s a further complication here: which groups count as ‘better-off’ and which count as ‘worse-off’? There are many ways to divide the two, for example, taking the middle group (on the BBC understanding the New Affluent Workers) as the dividing line, so that the three above them are the better-off and the three below are the worse-off; or putting the line at a particular point we take to represent ‘a life worth living’. We could say that every group higher in the hierarchy than the worst-off (on the BBC understanding, every group except the Precariat) counts as better-off. Conversely, we could say that only the best-off class group counts as better-off, because it is better-off than all the others. Any such decision would be more or less arbitrary. A principled way to divide the two would be to use a ‘hypothetical baseline’, which is to say, a non-actual distribution of goods against which we can compare the current distribution of goods―whichever goods we’re interested in. A good baseline would be one where a lack of social mobility does not preserve class advantage and disadvantage. Take occupation, for example. We might compare the current distribution of occupations and its clusterings by class group to a hypothetical distribution of occupations distributed fairly, for example by chance, across class groups. Each individual could then compare her actual position to the likelihood of her hypothetical position. The closer these two are to each other, the less she would be ‘better-off’ in the actual distribution, and the further apart these two are, the more she would be ‘better-off’ in the actual distribution).

You also know the group has privilege when the goods whose distribution we’re interested in are ‘zero-sum’, which is to say, there are a fixed amount of them, so that one person’s having more means another’s having less.8. Social prestige is not zero-sum: we can imagine a world in which all different kinds of jobs are accorded respect and recognition. Income is not zero-sum either. But occupation might be, because there is a more or less fixed range of things that need to be done, and number of people needed to do them. In specific instances, it certainly is: a company wants to hire three new people, and nine people apply. Any one person’s being hired means there is one less position available to the others. If the competition for the positions is fair, there’s no problem. If three are unfairly or unjustly disadvantaged by features they have and the remaining six lack, then the remaining six have privilege, regardless of which of their number is hired.

It’s this last thought that’s the most difficult to make sense of, and takes us back to the second part of the question raised above. Are all members of a group privileged? The best way to unpack it is to consider four different individuals. Two have fathers belonging to the Traditional Working Class and two have fathers belonging to the Established Middle Class (again making use of the BBC understanding). Let’s say these individuals are all daughters, and so have as much as a 0.7 chance of ending up in the same class group as their fathers. But that is not the same thing as their class group being determined by their fathers: they could be one of the 0.3 who shift between class groups.

Now imagine that two of the daughters, one from each group, in fact end up in the same class groups as their fathers. The daughter of the Established Middle Class parent experiences a wide range of opportunities which would not have been extended to her if she were not in that group, and she makes use of those opportunities. The daughter of the Traditional Working Class parent experiences a much narrower range of opportunities, a range which would have been more expansive if she were not in that group, and she makes use of those opportunities. And finally imagine that the remaining two of the daughters, one from each group, in fact manage to shift groups.

This happens entirely by luck: although the daughter of the Established Middle Class father has all the markers that would usually lead to favourable treatment, she happens to pursue an interest in which those markers do not translate into actual advantage. She experiences less opportunity for that reason, and as a result is ‘downwardly mobile’, which is to say, ends up in a class group lower in the hierarchy than her father. Symmetrically, although the daughter of the Traditional Middle Class father has all the markers that would usually lead to unfavourable treatment, she happens to pursue an interest in which those markers do not translate into actual disadvantage. She experiences more opportunity for that reason, and as a result is ‘upwardly mobile’, which is to say, ends up in a class group higher in the hierarchy than her father. Note that no one in this story squanders any opportunity, or does anything else that would make her responsible for where she ends up.

The daughter of the Traditional Working Class father who remained in the same group, and the daughter of the Established Middle Class father who shifted down between groups, might end up in a comparable situation. Likewise, the daughter of the Established Middle Class father who remained in the same group and the daughter of the Traditional Working Class father who shifted up between groups might end up in similar situations. If we were to consider each pair of daughters who have ended up with similar holdings of the relevant goods―occupation, income, social prestige, cultural capital, social capital―we might find it undesirable to describe one as having class privilege and the other as having class-based disadvantage.

Having a father in the Established Middle Class might have given the downwardly mobile daugher better chances, but those chances didn’t materialize into actual holdings. Given that she’s ended up in a comparable situation to a person we describe as disadvantaged, shouldn’t we rather think she’s disadvantaged too? And similarly, having a father in the Traditional Working Class might have given the upwardly mobile daughter worse chances, but those chances didn’t materialize into actual holdings. Given that she’s ended up in a comparable situation to a person we describe as advantaged, shouldn’t we rather think she’s advantaged too?

Against this thought, we can see that those in better-off class groups enjoy greater security over their positions. Even though the daughters of the Established Middle Class can end up in class groups lower in the hierarchy, they are much less likely to; even though the daughters of the Traditional Working Class can end up in class groups higher in the hierarchy, they are much less likely to. Considering the differences between these four daughters allows us to see two things.

First, both daughters of Established Middle Class fathers have class privilege. The first daughter has more privilege than the second, because her likelihood of ending up well-off translated into actually being well-off, while the second daughter’s likelihood did not. This will be important later in the paper, because I’ll suggest that the class privileged ought to ‘offset’ their privilege. The first daughter will have more privilege to offset than the second, so the first will have to take on more cost in order to satisfy her obligations than the second (see discussion in §IV).

Second, neither of the daughters of the Traditional Working Class have class privilege. The first has more class-based disadvantage than the second, because her likelihood of ending up with roughly the same income as her father translated into her actually ending up with roughly the same income as her father, while the second daughter’s likelihood did not materialize, and she in fact shifted into a better-off class group. This has the nice implication that even though she might end up in a class group which we’d think of as better-off, she doesn’t have obligations to offset, because she’s not class privileged. Privilege is determined not by mere membership, but by the backstory about membership and security in access to holdings.

The lack of social mobility that keeps the disadvantaged in place across multiple generations simultaneously keeps the advantaged in place across multiple generations. In other words: obstacles to equality of opportunity work out well for some. The idea behind there being unique obligations for beneficiaries of injustice is that those who benefit from unjust, unfair, or otherwise morally objectionable actions, events, states of affairs, histories, etc. have―or have stronger―obligations than others who are not beneficiaries (and who are not culpable, which I’ll say more about in the next section). In the rest of the paper, I’ll focus on the content and strength of these obligations.

III. Is Class Privilege Culpable or Non-Culpable?

When we culpably cause others to be badly-off, we will generally have very strong obligations to redress their situation. Some think our obligations can be so strong that they require us to take on cost in excess of the good it would do the badly-off person to have redress made. We don’t need to enter into that discussion here; but we can accept for the sake of argument that if the class privileged are culpable in their having class privilege, or in others’ lacking class privilege, then their obligations will be substantially stronger than if they are merely beneficiaries of obstacles to equality of opportunity. In this section, I assess whether the class privileged can be understood as culpable, first for having privilege, and second for the ways in which they are disposed toward that privilege.

I. Culpability for Having Class Privilege

I talked above about employers discriminating against job applicants. If this discrimination is explicit, then they’re obviously culpable. If it’s implicit, then they may not be (see discussion in Holroyd 2012). Certainly there may be specific individuals and groups who are culpable, either in what they have, or in what others lack. These individuals should certainly be held accountable. But is there anything we can say about class groups as a whole, or about most members of particular class groups?

It’s easy enough to see what people might want from a concept of privilege that would require it to involve culpability. They might be interested in privilege understood as the receipt of stolen goods, or at least the possession of goods that are the legacy of colonial theft, violence, and injustice. While this is a plausible way to think of many goods in a country like the UK, it’s not clear that it will allocate privilege along class lines (presumably all UK residents are privileged in this sense, rather than only those in the better-off class groups). They might be interested in privilege understood as profiting from political injustice against co-nationals, where e.g. failing to support mining communities in a transition to new employment industries is a political injustice, and those who profit are those whose interests are supported instead. The difficulty here is that government spending goes into a broad range of areas, so it’s again unclear that those who profit from this injustice are those in the better-off class groups―even if it’s clear that those people profit more than others.

They might be interested in privilege understood as complicity in a system designed deliberately to protect the advantage of some at the expense of others, or privilege understood as the sustaining, perpetuating, enabling, or upholding of that system. Supporting private schools by sending one’s children to them might be a good example of this kind of complicity. They might be thinking of privilege as a ‘club good’, exclusion from which is a harm to non-members. Owen Jones has described the British Conservative Party in this way, as a ‘coalition of privileged interests’ (Jones 2012). Michael Monahan writes in The South African Journal of Philosophy that privilege requires active participation on the part of the privileged (Monahan 2014).

George Yancy, writing recently in the New York Times, argues that the privileged can be culpable simply in virtue of group membership: men in virtue of being members of the group of all men, white people in virtue of being members of the group of all white people. For him, to be white in a race-divided society is to be racist; to be a man in a gender-divided society is to be sexist. He gives a range of disparate justifications for this claim.

For gender, they include: that despite men’s best intentions they perpetuate sexism; that men are complicit in industries that objectify women; that men see women through the male gaze despite intentions not to objectify women; that men share collective erotic feelings and fantasies which themselves are complicit in the degradation of women; that even if men fight against their sexism there will be moments of failure, and they will oppress women, so they cannot be fully innocent (Yancy 2015).

For race, his justifications include: that white people perpetuate racism; white people ‘harbour’ racism; white people benefit from racism; white people are part of, and reap comfort from, a system that gives them advantages while giving black people disadvantages; that white people are tied to forms of domination; that white people are wilfully ignorant of their ties to forms of domination; that white people have ‘signed a contract’ that guarantees them, but not others, social safety (Yancy 2015). Yancy says explicitly that not doing these things intentionally is not enough to free people from responsibility for them.

That’s about as much as I can offer in favour of the culpability of having privilege. Against such culpability, I can offer two arguments. The first is that we can cause harm, yet not in a way that we are morally responsible for. Think of actions that fall below some threshold of moral accountability, such as individual instances of rudeness; or actions that are not known to be (or even merely widely recognized to be) harms, as individual greenhouse-gas emitting actions before 19909. were not; or actions taken with care and without malice that nonetheless by luck turn out to do damage to another person. In these kinds of cases we can be a cause (or part of the cause) of harm without meeting the stronger conditions required to be morally responsible for what we cause, like intention,10. knowledge, foreseeability, ability to do otherwise, and so on (these vary between accounts).

The former kinds of case are particularly interesting because these ‘below the threshold’ actions can add up to social harms that are particularly damaging for those upon whom they fall, and yet ethicists struggle to account for any moral responsibility to remedy the harms (see e.g. Glover & Scott-Taggart 1975). Many see climate change in this way, because individuals’ greenhouse gas emissions don’t appear to be intentional causes of harm, but their cumulative effects involve great harm for a great number of people. Causing harm with one’s actions taken alone is not the same as causing harm through one’s actions taken together with many other people’s actions (any action taken alone may not amount to a harm while the actions taken together may do), and causing a harm―whether alone or together with others―is not generally thought to be sufficient for culpability, if the other conditions are not met.

Perhaps what is in the background is the thought that the privileged could get together and take action to make it the case that they weren’t privileged anymore, and the fact that they don’t do so is an omission for which they are culpable. I have argued elsewhere against the culpability of these kinds of groups, on the grounds that they lack the control necessary to describe what they cause or don’t cause as intentional actions or omissions (Lawford-Smith, 2015). Of course, Monahan and Yancy could be tacitly suggesting a revision to the requirements for moral culpability, following something like a ‘strict liability’ model as exists in tort law. But they’d need to make an argument for this, and as far as I have been able to find, they haven’t.

The second argument against the culpability of having privilege is that we can be entirely uninvolved in the causation of harm, and yet be a beneficiary of it. There are plenty of cases of this kind of ‘innocent’ benefiting given in the literature on benefiting from injustice (see e.g. Butt 2007; Anwander 2005; and the papers collected in Page & Pasternak 2014). Without the relevant kind of causal involvement, there’s no question of culpability. We might still be interested in these kinds of advantages, because they may yet establish obligations. Most of the discussion about benefiting from injustice has been about articulating the obligations of innocent beneficiaries, whether in cases of historical wrongdoing, or in the contemporary case of climate change (with the exception of Pasternak 2014 who takes up the issue of beneficiaries who might in various ways fail to be fully innocent, which I’ll come back to in the next sub-section).

We already have well worked-out moral theories that tell us about the normative implications of those who cause harm, contribute to the causing of harm, are complicit in harm, and so on. For those in privileged class groups who count as harming in one of these ways, we can simply apply what we already know about those kinds of cases. For example, Christian Barry and Gerhard Øverland have done a lot of work on the responsibilities that follow from a person’s contributing to harm (Barry & Øverland 2015); Chiara Lepora and Robert Goodin have provided a very thorough discussion of the ways a person can be complicit in harm, and what might follow from that in terms of holding the complicit responsible (Lepora & Goodin 2013). We don’t, however, have a well-worked out moral theory that tells us about those who merely benefit from harm, especially in the more distinctive ways typically involved when we think about privilege.

The disagreement with Monahan and Yancy, and any others who think that having privilege is culpable, is over the proportion of privileged people who can plausibly be classified as culpably involved in the preservation of their own advantage, compared with the proportion who cannot be. My suspicion―as naïve and charitable as it may be!―is that when it comes to class, there are a great many people who cannot be classified as culpable, even if there are some who can. I’ll proceed by focusing on the obligations of the larger non-culpable group.

ii. Culpability for Dispositions Toward One’s Class Privilege

Avia Pasternak is one of the few people who talks about ways of benefiting that are not fully innocent (Pasternak 2014). She makes a set of distinctions about the ways people can benefit that are useful in thinking about class privilege. She distinguishes (1) being unaware, and not reasonably able to be aware, that you’re benefiting from wrongdoing; (2) receiving benefits passively rather than actively seeking them; (3) not desiring the benefit; and (4) not being able to avoid receiving the benefit without unreasonable cost (see discussion in her paper for references to authors who discuss the moral upshot of each of these). On this view, what it would mean to be truly innocent in one’s class privilege would be to lack knowledge, desire, activity, and freedom in being privileged. Being implicated in one or more of these ways can, Pasternak argues, change the content and strength of the obligations one has in virtue of one’s benefiting.

Although it’s an empirical question, it does seem plausible that there are a majority of people in the better-off UK class groups who would meet (2) and (3)―not actively seeking the advantages they have, and not desiring them. Presumably many people in the UK desire a more equal and more socially mobile society. (1) and (4) are more difficult. How many people meet (1)―being aware or not reasonably able to be aware that they’re benefiting from wrongdoing―depends on how much of the UK’s class-based inequality can be attributed to wrongdoing by specific actors, compared to how much is a matter of wrongs emerging from long-established policies, systems, institutions, norms, and so on, and the extent to which the former, if true, is common knowledge. The more that it is the former, and the more that is common knowledge, the more it is open that those who are aware of this will not be able to escape the charge of knowledge of benefiting from wrongdoing.

How many people meet (4)―not being able to avoid receiving the benefit without unreasonable cost―depends on facts about what the class-privileged person’s secure opportunities are. Obviously no child can choose to walk away from her class-privileged parents in order to neutralize her starting position. But class-privileged parents often choose to game the educational system, and could obviously choose to send their children to state-funded schools instead of private schools. In that sense, many class-privileged children benefit from others’ wrongdoing on their behalf, and some have argued that this can also be a ground of very strong obligations (see discussion in Goodin & Barry 2014). The upshot is that some class-privileged people will have stronger obligations than others depending on how many of these conditions she meets.

Moral Obligations from Class Privilege

Now that we have a decent handle on non-culpable class privilege, we can start to think about the obligations that those with class privilege might have. In a recent thread in Legal Theory, some have approached the issue of class privilege by proposing more stringent legal responsibilities for the ‘gatekeepers of social advantage’, including landlords, employers, and university admissions boards (see discussion in Khaitan 2015). This is a sensible legal approach, but as an articulation of the moral responsibilities coming from class privilege it won’t be precise enough. Many who happen to be gatekeepers will not themselves have privilege, and because even for those gatekeepers who do, measures designed to ensure fair equality of opportunity in access to advantage do not do anything to affect the privilege of the gatekeepers themselves. (At best, such measures will make access to advantage fairer in the future, and so affect who the future gatekeepers of advantage are). We’re asking what obligations arise for whoever has class privilege, not for whoever controls the future distribution of privilege―although the former might end up being partially directed towards the latter.

Discussions about the obligations of beneficiaries have tended to focus on individuals’ actual benefits (see e.g. Anwander 2005; Butt 2007; Goodin 2013; Goodin & Barry 2014; Haydar & Øverland 2014; Heyward 2014; Pasternak 2014). In developing the concept of class privilege, I have argued that people can have class privilege because they are more likely to receive benefits, even if they actually do not benefit. Those born to parents in better-off class groups are more likely to end up with more economic, social, and cultural goods than those born to parents in worse-off class groups.

Discussions about the obligations of beneficiaries have also tended to focus on discrete identifiable acts of injustice, from which specific kinds of benefits can be traced to specific individuals (although cf. Barry & Wiens 2014; Heyward 2014). In developing the concept of class privilege, I extended the scope of what people can benefit from, to cover structural injustice and social inequality.11. And I extended the scope of what benefits can consist in, to include e.g. social and cultural capital. So unlike when benefits are held in the form of money or material goods to some discrete degree, the class-privileged person often cannot simply ‘give up’ her privilege and be done with the matter.

In fact, characterizing the obligations of the class privileged in the way others have characterized beneficiaries’ obligations―for example to disgorge benefits (Goodin & Barry 2014, pp. 371-372), or to relinquish benefits to the subjective extent that you value them (Butt 2007, pp. 140-143)―would seem to misfire. Giving up benefits might mean cutting family ties, or throwing away educational opportunities, or walking away from challenging and rewarding jobs, or simply trading places with another class privileged person, or otherwise making oneself comparatively worse-off for no obvious net gain in advantage to someone else (merely trading places in an ad hoc way with someone worse-off doesn’t obviously serve the cause of justice, because it might not be the best or the fairest way to compensate a person who has been unfairly disadvantaged, and it might not do anything to change the distribution of goods into the future).

It makes more sense to think in terms of the privileged having obligations to offset their privilege. I borrow the metaphor of ‘offsetting’ from discussions about climate change, where it is accepted that it would be very difficult for individuals in many contemporary domestic institutional settings to fully eliminate their greenhouse gas (GHG) emissions. ‘Offsetting’ captures the idea that emitting to some degree, although surely not to just any degree, is non-culpable (see discussion in §III), and yet can and should nonetheless be neutralized. Applied to class privilege, it suggests that it would be very difficult, and in some cases undesirable, for individuals to avoid having privilege, and yet that such individuals can and should neutralize the privilege that they have.

The explanation for why emitters should offset is that climate change threatens serious harm to persons, animals, and the environment; the explanation for why the class privileged should offset is that a lack of social mobility is a serious harm to those with parents in worse-off class groups. Worlds characterized by class privilege are bad, even if the people who have class privilege are not (necessarily) themselves bad. The idea of offsetting also makes the object of the obligations clear. When we offset our GHG emissions, we neutralize the harms they might otherwise do by removing or preventing GHG emissions elsewhere (e.g. by planting new trees, or preventing deforestation), and thereby make a small contribution to the mitigation of a major harm.

Similarly, then, when a person offsets her class-privilege, she must be attempting to neutralize the harm done by a system that distorts the distribution of goods in a society according to class. She can do this to maximum effect by channeling her offsetting to undermine the source of her class privilege and others’ corresponding class-based disadvantage. In summary:

What the class-privileged owe: Members of class-privileged groups must offset their privilege by taking on costs in order to undermine the sources of class privilege and class-based disadvantage.

The class-privileged can offset their privilege by taking on costs up to a point that is commensurate with their group-based advantage, either as time, effort, money, or other material resources (see also discussion in Barry & Lawford-Smith, ms.) What kinds of things count as ways to take on the relevant costs, and thereby offset privilege? The following are potential contributions that go to the source of class privilege (although this list is not exhaustive):

  • Challenge classist comments made in social situations
  • Show social respect and recognition to members of worse-off class groups
  • Engage in leisure activities where you are likely to interact with people from a range of different class groups
  • Take steps to collectivize into groups organized against class injustice (Young 2003)
  • Publicly boycott companies and corporations known to be involved in classist hiring or employment practices
  • Stand in solidarity with members of class groups experiencing discrimination or oppression (e.g. the working class) (Kolers 2014)
  • Undertake research into class-based social differences and whether they have alternative explanations, and share findings
  • Write to MPs, sign petitions, raise awareness about morally problematic class-based social differences
  • Encourage workplaces (your own and others’) to use anonymized CVs when hiring to mitigate class bias
  • Encourage workplaces (your own and others’) to add ‘class’ to existing diversity policies for hiring
  • Donate money, goods, or labour hours to charities and organizations working against class injustice
  • Vote for political parties whose platforms include action against class injustice
  • If you are a parent, send your children to state-funded schools (see also discussion in Swift & Brighouse 2016)

Notice while these would all look fairly uncontroversial as normative implications of race or gender privilege, they are slightly more surprising when it comes to class. In the first case, there’s not yet much of a public consensus on what kinds of comments count as classist (as Owen Jones nicely demonstrates in Chavs (2012)). So we might need to add an item to the list, to first figure out exactly what kinds of comments count as disparaging, discriminatory, prejudiced etc. against people based on class. Social norms are part of what maintain class privilege, and norms can be partly eroded with minor social sanctions that threaten esteem, such as verbal challenge. Recognition can make a difference to those who have been unfairly disadvantaged, and greater integration between class groups can provide opportunities for giving recognition, and more generally for challenging and breaking down stereotypes.

Whether we need anonymized CVs (as it has been argued that we do to combat implicit race and gender bias), will depend on whether we can read class off names, addresses, educational institutions, etc.―more empirical work needs to be done on this. If class can be read off appearance, dialect, or other features made visible in social interaction, then there will be further issues of implicit bias to be faced up to (there may also be explicit bias, but as explained, this takes us back to culpability). The main struggle will be to change long-standing institutions and policies, those related to education perhaps chief among them.

Conscientious readers might wonder how important obligations relating to class privilege are, compared to other kinds of obligations we might have. This is a huge issue so I can’t say much about it here. But the most important point is to make is that there’s continuing disagreement about the extent to which it’s permissible to show partiality in moral matters to those within one’s own national borders. To the extent that it is, class injustice is one of the most prominent sources of injustice to persons in the UK, so the obligations I’ve outlined above will be very important. To the extent that it isn’t, the conclusion will be very different. After all, the UK is a rich country, and there are many people in the world who are much worse-off than the worst-off here. For those who deny that it’s permissible to show partiality to those within one’s own national borders, rather than remedying class privilege in one’s own rich country, it might be more important to take action against climate change, or against global poverty (for more on this question of making moral tradeoffs see Lawford-Smith forthcoming).

Main challenge

In this final section of the paper I want to address a challenge to this account of class privilege and the moral obligations that it comes with. What has been driving this whole story is the fact that some people are likely to end up in a worse-off position in a distribution of a given good, simply because of their social class group. I’ve argued that this can give particular kinds of obligations to those likely to end up in better-off positions simply because of their social class groups. But what if there’s an explanation of people ending up in a worse-off position, that isn’t simply the lack of social mobility in the UK? What if members of worse-off class groups are complicit in their own disadvantage, for example because they have internalized preferences against taking up certain kinds of opportunities?

(This challenge is not unique to class, it shows up in particular for gender as well. We might notice that there are fewer women than men in full-time employment,12. and be concerned that this suggests morally objectionable gender inequality in the workplace. A critic might counter that large numbers of women prefer not to work, so they can care full-time for their young children).

I said earlier that the correlation between fathers’ incomes and daughters’ incomes is between 0.45 and 0.7 (where 1.0 would mean fathers’ incomes fully determine daughters’ incomes). Another way to think about this is that for every 100 daughters of fathers in worse-off class groups, between 45 and 70 of these daughters will end up with roughly the same income as their fathers, and between 55 and 30 of these daughters will end up with incomes significantly different to their fathers’. But notice that we’re reading this data on the assumption that the daughters’ preferences lead them to take up the opportunities they are presented with, so that the explanation of as many as 70% of the daughters of fathers in worse-off class groups ending up with the same income as their fathers is the UK’s lack of social mobility. If the daughters’ preferences lead them to reject some or many of the opportunities they’re presented with, then their disadvantage will have an alternative (or additional) explanation.

There is at least anecdotal evidence in the UK to suggest that at least some members of worse-off class groups have been complicit in their own disadvantage, for example by internalizing preferences against taking up opportunities that would provide more social mobility. Examples include being disposed against higher education, or disposed in favour of certain kinds of manual or routine occupations which generally come with less social standing and less remuneration. To the extent that this is true, a member of a worse-off class group could be making a free choice to adopt the norms or values of her class group, in which case she is not only being disadvantaged by a lack of social mobility, but is also determining her own disadvantage.

I say ‘could’ rather than ‘would’ because it’s not clear whether we should see this as a free choice. It matters whether the choice is made reflectively, with knowledge of what is at stake. Preferences can fail to count as genuine when they are the result of coercion or social conditioning. Group identification dynamics can be like this: others ascribe negative characteristics to a group, and members of the group adopt and affirm these characteristics in order to reclaim social esteem; or the group itself adopts certain values, perhaps in opposition to other groups, and conditions them into new members (in particular children). But in those cases the preferences do not explain away the disadvantage; they’re part of the disadvantage.

What is tricky about this challenge is that it puts us between a rock and a hard place. We could agree that those in worse-off class groups sometimes have preferences that lead them to reject opportunities, but say that these preferences are coerced or conditioned, so that we are not forced to agree that they’re complicit in their own disadvantage. Only genuine preferences, those endorsed reflectively and with knowledge of the consequences, could make them complicit. But there’s something uncomfortable about looking at someone’s preferences and telling her that they’re not her real preferences―that she would prefer different things if she hadn’t been conditioned by her class group to want those things. Doesn’t it add insult to injury to tell those who deny the value of education and prefer to make an earlier start in the labour market that they’re simply mistaken about what’s good for them?

On the other hand, if these preferences are genuine, then it’s hard to see how the disadvantage counts as morally objectionable at all. The challenge from earlier was to move from class and class advantage to class privilege, which we did by locating unfairness or injustice in the backstory of who got to be in which class groups. If those who end up in worse-off class groups are there because of the choices they made, rather than the opportunities that were not made available to them but were made available to others, then unfairness or injustice drops out of the picture.

Must we choose between adding insult to injury, and denying that there is class privilege? In fact we can squeeze out of this difficult position entirely. It matters whether the disadvantage of the person in the worse-off class group is ‘overdetermined’, which is to say, caused by two different things either one of which would have been sufficient. If the sole cause of the disadvantage is her own choice―if her preferences are genuine, she prefers a job that comes with less social prestige and less remuneration, and if she had preferred differently then she could have ended up in a different job―then we’re forced to deny that there is class privilege. The disadvantage is not morally problematic; so there’s no injustice or unfairness in the backstory about the distribution that compromises the advantages; so there’s no class privilege and corresponding obligations.

But if her choice is only one of the causes, then we neither have to deny that her preferences are genuine, nor give up on the idea of class privilege. Whether her preferences are genuine or not, the fact remains that were she to have chosen differently, she would still have been disadvantaged. Her disadvantage is overdetermined because the external obstacles to social mobility remain in place. (One thing this does imply, though, is that undermining obstacles to social mobility might not be sufficient to equal opportunity in class-group determination. If people have preferences that lead them to reject particular kinds of opportunities, we might see similar patterns to those we see now, even in a society with full equality of opportunity.)

In summary, the lack of social mobility in the UK causes inequalities between people that are not solely a matter of individuals’ choices. Those who do well out of these inequalities are privileged. Even when they are not culpable for having privilege, or for the dispositions they have toward their privilege, such people have obligations to take steps to address this inequality. One effective and politically achievable way for them to do so is through offsetting their privilege in one or more of the ways suggested above. Offsetting gives the privileged a concrete way to address a serious moral problem in their own society. While I have focused this discussion on class inequality in the UK, none of the ethical issues are restricted to the UK context. So this discussion should be useful to anyone worried about the ethical implications of class-based societies.

Acknowledgements: This work is funded by the European Commission. I’m grateful to audiences in Philosophy Departments at the University of Stirling and the University of Melbourne, and the MSPT Graduate Workshop at the Australian National University, for discussion on this paper; to Komarine Romdenh-Romluc, Ten-Herng Lai, RJ Leland, Sarah Hannan, and Nancy Yang, for comments on the written version; and to Dominic Wilkinson, two anonymous reviewers, and four anonymous editors of the Journal of Practical Ethics for comments and suggestions.


Aldridge, S. ‘Life Chances and Social Mobility: An Overview of the Evidence’, Cabinet Office, Prime Minister’s Strategy Unit (2004). Online at http://www.swslim.org.uk/documents/themes/lt10_lifechances_socialmobility.pdf accessed 23rd May 2016.

Anwander, Norbert. “Contributing and Benefiting: Two Grounds for Duties to the Victims of Injustice”, Ethics and International Affairs 19 (2005), 39-45.

Bailey, Alison. ‘Privilege: Expanding on Marilyn Frye’s ‘Oppression’’, Journal of Social Philosophy 29/3 (1998), pp. 104-119.

Ball, Charlie. ‘Most people in the UK do not go to university – and maybe never will’, The Guardian, Tuesday 4th June, 2013. Online at http://www.theguardian.com/higher-education-network/blog/2013/jun/04/higher-education-participation-data-analysis accessed 23rd May 2016.

Barrett, H. Clark., Bolyanatzc, Alexander., Crittendend, Alyssa N., Fesslera, Daniel M. T., Fitzpatricke, Simon. Gurvenf, Michael., Henrich, Joseph., Kanovskyj, Martin., Kushnickk, Geoff., Pisorf, Anne., Scelzaa, Brooke A., Stich, Stephen., von Ruedenn, Chris., Zhaoh, Wanying., Laurence, and Stephen. ‘Small-scale societies exhibit fundamental variation in the role of intentions in moral judgment’, Proceedings of the National Academy of Sciences (2016) [online first].

Barry, Christian. ‘Global Justice: Aims, Arrangements, and Responsibilities’ in Toni Erskine (Ed.) Can Institutions Have Responsibilities? (Basingstoke: Palgrave, 2003).

Barry, Christian. & Kirby, Robert. ‘Scepticism about Beneficiary Pays: A Critique’, Journal of Applied Philosophy 32/4 (2015) [early view].

Barry, Christian. & Lawford-Smith, Holly. ‘On Satisfying Duties to Assist’, manuscript at 25th May 2016.

Barry, Christian. & Øverland, Gerhard. Responding to Global Poverty (Cambridge: Cambridge University Press, 2015).

Barry, Christian. & Wiens, David. ‘Benefiting from Wrongdoing and Sustaining Wrongful Harm’, Journal of Moral Philosophy (2014), pp. 1-23.

Bell, Derek. ‘Global Climate Justice, Historic Emissions, and Excusable Ignorance’, The Monist 94/3 (2011), pp. 391-411.

Bertrand, Marianne. & Mullainathan, Sendhil. ‘Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labour Market Discrimination’, The American Economic Review 94/4 (2004), pp. 991-1013.

Brighouse, Harry. & Swift, Adam. Family Values: The Ethics of Parent-Child Relationships. (New Haven: Princeton University Press, 2016).

Butt, Daniel. “On Benefiting From Injustice”, Canadian Journal of Philosophy 37 (2007), 129-152.

Croizet, Jean-Claude., & Claire, Theresa. ‘Extending the Concept of Stereotype Threat to Social Class: The Intellectual Underperformance of Students from Low Socioeconomic Backgrounds’, Personality and Social Psychology Bulletin 24/6 (1998), pp. 588-594.

Fricker, Miranda. ‘Epistemic Oppression and Epistemic Privilege’, Canadian Journal of Philosophy 29/1 (1999), pp. 191-210.

Frye, Marilyn. ‘Oppression’, in The Politics of Reality. (Freedom, California: Crossing Press, 1983), pp. 1-16.

Glover, Jonathan and M. Scott-Taggart. 1975. ‘It Makes No Difference Whether or Not I Do It’. Aristotelian Society Supplementary Volume xlix, pp: 171–209.

Goodin, Robert. & Barry, Christian. ‘Benefiting from the Wrongdoing of Others’, Journal of Applied Philosophy 31/4 (2014), pp. 363-376.

GOV.UK. ‘National Minimum Wage and National Living Wage rates’ (2016). Online at www.gov.uk/national-minimum-wage-rates accessed 21st May 2016.

HarrisInteractive. ‘Firefighters, Scientists and Teachers Top List as “Most Prestigious Occupations” According to New Harris Poll’, August 1st 2007. Online at http://media.theharrispoll.com/documents/Harris-Interactive-Poll-Research-Pres-Occupations-2007-08.pdf accessed 21st May 2016.

Heyward, Clare. ‘Benefiting from Climate Geoengineering and Corresponding Remedial Duties’, Journal of Applied Philosophy 31/4 (2014), pp. 405-419.

Holroyd, Jules. ‘Responsibility for Implicit Bias’, Journal of Social Philosophy 43/3 (2012), pp. 274-306.

Jackson, Michelle. ‘Disadvantaged by discrimination? The role of employers in social stratification’, The British Journal of Sociology, 60/4 (2009), pp. 669-692.

Jenkins, Stephen P. ‘The income distribution in the UK: A picture of advantage and disadvantage’, Centre for Analysis of Social Exclusion 186 (2015), pp. i-31.

Jones, Owen. Chavs: The Demonization of the Working Class (London: Verso, [2011] 2012).

Khaitan, Tarunabh. A Theory of Discrimination Law (Oxford: Oxford University Press, 2015).

Lawford-Smith, Holly. ‘Climate Matters Pro Tanto, Does It Matter All-Things-Considered?’ Midwest Studies in Philosophy, forthcoming.

———. ‘What ‘we’?’ Journal of Social Ontology 1/2 (2015), pp. 225-250.

Lepora, Chiara. & Goodin, Robert. Complicity and Compromise (Oxford: Oxford University Press, 2013).

McIntosh, Peggy. ‘White Privilege: Unpacking the Invisible Knapsack’, Peace and Freedom, July/August (1989).

Monahan, Michael. ‘The Concept of Privilege: A Critical Appraisal’, The South African Journal of Philosophy 33/1 (2014), pp. 73-83.

Monbiot, George. ‘Plan after Plan Fails to Make Oxbridge Access Fair. There is Another Way’, The Guardian, 24th May 2010. Online at http://www.theguardian.com/commentisfree/2010/may/24/oxbridge-access-fair-top-universities accessed 1st November 2015.

Office for National Statistics, 2011 Census. Key Statistics for Local Authorities in England and Wales, 11th December 2012. Online at http://www.ons.gov.uk/ons/rel/census/2011-census/key-statistics-for-local-authorities-in-england-and-wales/rft-table-ks201ew.xls accessed 1st November 2015.

Office for National Statistics. Full Report: Women in the Labour Market, 25th September 2013. Online at http://webarchive.nationalarchives.gov.uk/20160105160709/http://www.ons.gov.uk/ons/dcp171776_328352.pdf accessed 25th May 2016.

Page, Edward. & Pasternak, Avia. (Eds.). ‘Special Issue: Benefiting from Injustice’. Journal of Applied Philosophy 31/4 (2014).

Pasternak, Avia. ‘Voluntary Benefits from Wrongdoing’, Journal of Applied Philosophy 31/4 (2014), pp. 377-391.

Paxton, W., & Dixon, M. State of the Nation: An Audit of Injustice in the UK (London: IPPR, 2004).

Pickett, Kate. & Wilkinson, Richard. The Spirit Level: Why Equality is Better for Everyone. (London: Penguin, 2010).

Savage, Mike. & Devine, Fiona. ‘The Great British Class Survey – Results’, BBC Science, 3rd April 2013. Online at http://www.bbc.co.uk/science/0/21970879 accessed 1st November 2015.

Steele, C.M. ‘A threat in the air: How stereotypes shape the intellectual identities and performance of women and African Americans’, American Psychologist 52 (1997), pp. 613-629.

Totaljobs.com. ‘How to claim jobseekers allowance’ (2016). Online at http://www.totaljobs.com/careers-advice/money-and-legal/how-to-claim-jobseekers-allowance accessed 21st May 2016.

UKGeographics. ‘Blog: Social Grade A, B, C1, C2, D, E’, UK Geographics 23rd February 2014. Online at http://www.ukgeographics.co.uk/blog/social-grade-a-b-c1-c2-d-e accessed 1st November 2015.

Yancy, George. ‘Dear White America’, New York Times December 24th, 2015. Online at http://opinionator.blogs.nytimes.com/2015/12/24/dear-white-america/ accessed 22nd January 2016.

Young, Iris Marion. ‘Political Responsibility and Structural Injustice’, The Lindley Lecture 2003, University of Kansas, May 5th 2003

1 This figure is the Jobseekers Allowance (JSA) for people aged 25 years and over, and is the same for both contribution-based and income-based JSAs (which a person is entitled to depends on whether she has made sufficient past contributions to National Insurance) (totaljobs.com).

2 These numbers are based on the (hourly) National Living Wage and National Minimum Wage rates that apply from 1st April 2016, for the category of people aged 25 years and over, and calculated to a weekly wage on the assumption of the UK’s legal maximum of a 48-hour working week (GOV.UK 2016).

3 Here I’m extending the idea of benefiting from (discrete, identifiable acts of) injustice (i.e. perpetrated by one individual against another and sending benefits to a third), to cover benefiting from structural injustice, social inequality, and other states of affairs that are morally problematic and yet may fall short of being unjust.

4 Different features than class―for example race, and gender―will be more or less relevant to social inequality in the context of different countries. Class is one very important feature in the UK, which is why I’m focusing on it here.

5 The total number of respondents was 161,458. These were mostly from England (86%) with small proportions from Scotland (8%), Wales (3%) and Northern Ireland (1%). 56% of respondents were men and 43% were women, the average age was 35, and 90% of participants described themselves as white. (These figures are not fully representative; the 2001 census put the proportion of white people in the British population at 81.9% (Office of National Statistics 2011), and the proportion of women as being slightly higher than men: 32.2 million women compared to 31 million men (ibid)).

6 For other accounts of privilege which similarly look for morally objectionable aspects of advantage, but instead focus on race or gender, see (McIntosh 1989; Bailey 1998; Frye 1983).

7 E.g. http://www.bbc.com/news/magazine-22000973 accessed 23rd May 2016.

8 See also the interesting discussion of benefiting ‘at the expense’ of another, in (Anwander 2005).

9 This is a generous date; some think it is much too late. See discussion in (Bell 2011).

10 At least, this is a condition for moral responsibility in what behavioural economists call ‘WEIRD’ societies: Western, educated, industrial, rich, and democratic. Significantly less importance is placed on intention in non-WEIRD societies. See discussion in (Barrett et al. 2016).

11 Existing accounts of obligations to address structural injustice are given on the basis of social interdependence, and do not assign unique obligations to those who do well out of the injustice. See discussion in (Young 2003).

12 In the UK, roughly 90% of men aged 28-44 are in full-time work, compared to roughly 70% of work (these numbers change slightly for different age groups). (Office for National Statistics 2013, p. 5).