Does being a ‘bad feminist’ make me a hypocrite? Politics, commitments and moral consistency

Abstract

A ‘bad feminist’ is someone who endorses feminist ideals and values but finds themselves falling short of them. Since bad feminists exhibit an inconsistency between what they say and what they do, this can generate worries about hypocrisy. This article investigates whether and when members of political movements with certain ideals ought to worry they are being hypocritical. It first provides a diagnosis of why worries about hypocrisy seem common in the political arena. I argue that accusations of hypocrisy are apt when one is in insufficiently committed to the values entailed by one’s pronouncements. It is particularly hard to assess what constitutes sufficient commitment in politics because many issues are multi-factorial, overdetermined, and involve numerous competing considerations, making it difficult to assess how genuine commitments to values should manifest in behaviour. Since there are many ways that one can act inconsistently with their pronouncements, mere inconsistency is not a signal of hypocrisy, particularly if one is disposed to take on costs to further that movement’s ends. I also consider and respond to a number of considerations against calling out hypocrisy, including worries about our accuracy, worries about the effects such accusations have on discourse, and worries about how accusations incentivise particular kinds of behaviour.

1 Introduction

Bad Feminist is a series of essays and reflections by Roxane Gay (2018). A New York Times bestseller, and appearing on lists of best books of the year at NPR, The Boston Globe and Newsweek, many of the essays reflect on instances in which Gay engages in behaviour at odds with her feminist values. As she puts it, she enjoys music that she knows is terrible for women. She loves pink. She sometimes finds it easier to play dumb in front of tradesmen than to accurately represent her knowledge. She isn’t as well-read in key feminist texts as she would like to be.

Part of the popularity of this book, and other media like The Guilty Feminist podcast by Deborah Frances-White, is attributable to the recognisability of these experiences. Many of us have strong convictions and deeply held values we are willing to defend and expect others to uphold. Nevertheless, we are imperfect: we succumb to weakness of will, we have competing priorities, and sometimes doing what is right feels too difficult. Such experiences are not particular to feminists; many readers also endorse saving the environment and mitigating climate change, improving our governmental and economic systems, or upholding civic and political values. And yet we drive cars, claim deductions on our taxes that may be questionable, use services with track records of exploiting workers, and share media that increases polarisation. We also observe such behaviour in others. It is also not unusual to see avowed socialists using the latest iPhone, or committed free-market politicians accepting ‘hand outs’ from the government.

What should self-identified bad feminists (environmentalists, socialists, etc.) think of themselves? What should they think about each other? These questions interact with a variety of topics, including the ethics of blaming and shaming as public actions and the epistemic and moral dimensions of acting collectively in response to complex problems. A number of philosophers have investigated how we ought to assess ourselves and others given how individuals’ social positions are influenced by and contribute to unjust states of affairs, looking at the implications for virtue (Tessman, 2005), vice (Dillon, 2012), holding agents and groups responsible (Doan, 2016, 2020) and the political ‘purity’ of our actions (Shotwell, 2016).¹ Less attention, however, has been given to hypocrisy, of which inconsistency between one’s pronouncements and one’s behaviour frequently generates worries. Accusations of hypocrisy often leave unstated why, exactly, it is bad, but there are very strong cultural norms against it, and most of us strive to avoid it.²

This article considers whether and when bad feminists—understood as individuals who, by their own lights, falls short of feminist values despite endorsing those values—are hypocrites. It first provides a diagnosis for why worries about hypocrisy are especially common in the political arena.³ I argue that agents are hypocritical when they are not sufficiently committed to the values implicit in or underlying their pronouncements. What counts as sufficient commitment, however, depends on a range of factors: what the value in question requires, what other goods are at stake, and what capacity and opportunity the agent has to comply with this requirement. Political issues are particularly ripe for worries of hypocrisy because they are social, complex, multi-factorial, and can conflict with many other goods, so it is particularly difficult to work out which actions are evidence of insufficient commitment.

I then argue that we can make progress by asking whether agents who make pronouncements are disposed to bear relevantly similar costs to those that their pronouncements would entail, taking into account that differences in circumstances may make it more legitimate for some parties to bear particular costs over other agents. Finally, I consider whether we ought to, in practice, accuse others of hypocrisy when we see it. Even if hypocrisy is bad, a number of other considerations may warrant adopting a general policy of refraining from making accusations. These include worries about our accuracy, worries about what effect these accusations have on discourse, and worries that such policies can incentivise undesirable behaviour. I argue that none of these possibilities, upon closer inspection, warrant regularly refraining from calling out hypocrisy.

2 The commitment account of hypocrisy

What does it mean to be a hypocrite? It is common to think of a hypocrite as someone who says one thing and does another. As a first pass, it seems to involve some kind of inconsistency. But inconsistency alone doesn’t seem to quite get at the heart of hypocrisy. Someone who believes both P and not P has inconsistent beliefs, but they aren’t hypocritical. Similarly, someone who says that they will go to the shops today and then doesn’t isn’t hypocritical. Neither are people who are simply liars, though they are criticisable in another distinct way.⁴ I believe we can make progress in our understanding of hypocritical pronouncements by considering a prominent account of hypocritical blame, from Todd (2019; with similar sentiments from Friedman, 2013; Piovarchy, 2023; Isserow, 2021; Rossi, 2018). On this account, what is objectionable about hypocrisy is not what the hypocrite does, nor does it concern inconsistencies between pronouncements and behaviour themselves. Rather, inconsistency between pronouncements and behaviour are evidence of something further. Todd argues that agents lack the standing to blame when they are not sufficiently committed to the values that would condemn the wrong. That lack of commitment is ultimately what matters is strongly evidenced by two things. First, agents seem to retain their standing to blame when their inconsistent behaviour is not evidence that they lack commitment to the relevant values. If the wrong was a long time ago, such that the person has reformed their ways, apologised, and is not disposed to commit similar wrongs, it seems they can now appropriately blame others who commit similar wrongs. Second, agents can be hypocritical even if they haven’t committed the wrong under discussion, or done anything at all. They might simply be complicit in the wrong: someone who intentionally profits off of terrorism might be unwilling to detonate the bombs themselves, but it would be hypocritical of them to blame said terrorists. Alternatively, agents might be disposed to violate the value themselves, but simply lack the means of doing so. Such a person is a subjunctive hypocrite (Piovarchy, 2023; Isserow, 2021).

This account not only helps us understand our objections to hypocritical blame, it has no problem being extended to cover pronouncements too: agents ought not make pronouncements about what should be done if they are not sufficiently committed to the kinds of values that those pronouncements entail or imply. This helps explain why the term ‘hypocrite’ comes from the Greek ‘hypokrisis’, meaning “acting of a theatrical part”, such as in a role in a stage play. Though actors in a play are not hypocrites, the relevant idea is that both involve false pretence. Both blame and pronouncements are ways that people falsely represent their commitments to certain values.

I believe this account is particularly well-placed to help us make sense of our mixed feelings towards bad feminists (and their feelings towards themselves) and other agents involved in politics and activism. I propose these mixed feelings stem from uncertainty about what counts as ‘sufficient commitment’ in any given case, as multiple factors are relevant and it can be difficult to weigh them. Let’s now consider some of these factors. I will begin by considering how they affect our judgements about agents who endorse reducing greenhouse gas emissions. Starting with this domain will make the general ideas easier to get a handle on, as the negative effects of agents’ actions are somewhat easier to quantify and relatively linear (i.e., more emissions results in more harms).

One factor concerns the importance of the value in question and what is required to uphold it. Different values call for different kinds of behaviours, and different levels of stringency. This means that, depending on the value, our assessments of commitment can allow for differences in frequency of compliance and what counts as compliance. Regarding the former, someone who leaves the lights on one time, contributing to more carbon dioxide being emitted, can count as sufficiently committed to the kinds of values that condemn the harms that come from climate change. But someone who commits murder (even just once!) cannot be considered committed to the kinds of values that condemn murder, at least at that point in time.

Regarding the latter, behaviours that cause greenhouse gas emissions typically won’t be condemned (in a sense relevant to assessing commitment) if the action intentionally ultimately lowers emissions overall. Suppose I get hired in a job that emits extremely large amounts of carbon dioxide. But suppose my entire motivation for doing so is to make my way to the top, so that I can reduce the amount of emissions from the inside and better campaign for lower emissions across my industry, and I have good reason to think I will succeed. Here my actions are not evidence that I am insufficiently committed to the relevant values (nor arguably that I’ve done anything wrong), because my actions prevent an even larger amount of emissions. If anything, those emissions are a product of my genuine commitment to these values. But values that condemn murder are not like this; most people tend to think that respecting these values requires we consider our own behaviour in more deontological terms. Suppose I hunt serial killers and delight in their murder. In doing so I prevent more overall murders. It would nevertheless be hypocritical of me to argue that we ought not murder one another, on the basis that murder is wrong and fails to respect human life. Values which condemn murdering typically don’t treat ‘unless you can prevent further murders’ as an exception.⁵

A second factor concerns what else was at stake. The value in question often needs to be weighed against other bona fide values as well as goods, interests, and rights. Suppose the reason I have taken the high-emissions job is not that I will one day be able to reduce emissions by a greater amount, but that I am committed to some other worthwhile value. Perhaps I have dependents, and this is the only job available that will enable me to fulfil my duties. Assuming one ought to prioritise their dependents over their emissions, here my emissions-producing behaviour is not evidence of insufficient commitment to values that condemn emitting greenhouse gases; rather, it is only evidence of my (rightful) commitment to my children. Admittedly, it must also be the case that were a low-emissions job that enabled me to meet my duties available, I would take that job. I won’t avoid charges of hypocrisy if it was simply luck that there were no alternatives, and I would have chosen a high-emissions job for trivial benefits. It is not enough that I have the right beliefs about the value and comply with the value, I must also be disposed to give it a certain level of priority over less weighty considerations.

While one should prioritise their dependents in the case just described, this will not be true of all cases. Suppose I could significantly reduce emissions for the whole country by making my dependent trivially worse off, and I refuse to do this. This counts towards thinking that I am not sufficiently committed to reducing emissions, and I may lack the standing to blame someone who goes for weekend joy-rides. There is a trade-off to be made, and if I truly care about reducing emissions then at some point I should give it priority over other considerations.

Having identified clear cases where one ought to prioritise one value, and others where one ought to prioritise another, there will inevitably be harder cases in the middle where we are unsure whether a choice is evidence of insufficient commitment. To be clear, what counts as commitment does not depend entirely on what is ultimately the right action, though the two are correlated. Suppose that while a lower emissions job allows me to be a minimally decent parent, a higher emissions job would enable me to be a fantastic parent. There might be some fact about which option the parent should choose. But even if they choose the wrong option, it doesn’t necessarily show they are thereby insufficiently committed to the second value. That the two are roughly comparable, or that parties can reasonably disagree about what should be prioritised, suggests that any failure to do what is ultimately right is not necessarily evidence of insufficient commitment. We can allow for some tolerance in hard cases.

Finally, we need to also note that a violation will not be evidence of insufficient commitment when complying with the value would be too difficult or costly. Sometimes agents simply don’t have the capacity or opportunity to do what the values require. Sometimes agents are entitled to a certain amount of slack. Even if we think that one really ought to prioritise reducing emissions over benefitting their child, we might think doing so would be too emotionally difficult, given their love for their child. Or perhaps one could only get the low-emissions job by enrolling in a very difficult, expensive and competitive degree, with limited chance of finding a position. In such cases we might think that given the costs to the individual, their failure to choose this option is not evidence of insufficient commitment, because it would be unreasonable to expect them to bear such costs.

In sum: to assess whether behaviour is evidence of insufficient commitment to a particular value, we need to take account of (1) the importance of the value and what it requires (i.e., the manner in which the value is to be upheld or respected, and thus the frequency and nature of any violations), (2) what else was at stake and (3) the agent’s capacity and opportunity to uphold the value.

3 Commitments and bad feminists

Let’s now apply our framework to feminists and feminist values. When it is easy to assess (1), (2) and (3), it is easy to assess whether someone’s ‘bad’ tendencies make them a hypocrite. Suppose Gerri identifies as a feminist, believing that women are equal to men and that things should improve for them. However, she has other criticisable values that she prioritises much more. As a politician, she readily votes against policies which would improve women’s status in order to gain more self-serving power. She frequently does things which conflict with feminist aims when this provides trivial benefits to herself. Arguably Gerri counts as a bad feminist, insofar as she has certain values and falls short of them. But her actions are particularly egregious. It’s not that she simply ‘falls short’, it’s that she barely even tries to such a large degree that makes it questionable whether she really endorses those values at all. Gerri seems like a hypocrite, her pronouncements are an apt target of criticism, and she lacks the standing to blame others who fail to comply with feminist ends.

In contrast, when we are uncertain about (1), (2) and (3), it is much harder to know whether to consider someone hypocritical. I propose these kinds of uncertainties are frequently present in politics and the kinds of behaviours worried about by bad feminists. They are typically not like the case of Gerri, where we can clearly identify costs, benefits, competing goods, and capacity to comply. Rather, they tend to involve norms, practices, double binds, institutions, collectives, and systems. As a result, the net effect of any action does not typically result from a single agent violating a single value. As Young (1990) famously points out, many sources of oppression are not the result of ill-intentioned agents acting immorally. Instead, they often result from well-meaning individuals carrying out small actions which, over time, can add up to unjust circumstances. These features make it difficult to assess how strongly people ought to comply with relevant norms, and how to consider what else is at stake.

To illuminate these difficulties, consider Gay’s example of enjoying music with lyrics that objectify and degrade women, conflicting with feminist values. Although such values generate prescriptions like ‘don’t objectify women’ which appear simple enough to understand and codify, the extent to which these values prohibit things that contribute to the objectification of women is much harder to cash out. Suppose I buy an album which has content objectifying women. Is this contribution significant enough to mean I am not sufficiently committed to the value? What about buying a different album without such lyrics from the same artist? Or buying albums by artists hosted by the same record label? Or paying for tickets to a festival at which said artist will be present, even if I do not intend to see their performance?

Assume this value does give us weighty reasons to not engage in actions which contribute to the objectification of women. Let’s consider now what else is at stake. One might think very little is at stake as all I have to do is not buy some albums. But if our sense of what things contribute to this music’s prevalence expands, then so does the number of potential conflicts. My child really likes this music, and won’t be swayed by my reasoning. Does my interest in maintaining harmony and making them happy outweigh this? What if I’m out with friends at a club and the song comes on to their excitement; does having fun dancing help contribute? If it does, I may frequently have to be quite the buzzkill.

Assume that our values give us reasons sufficiently strong enough to outweigh competing considerations. Let’s now consider my ability to comply with this value. At first glance, this also seems easy as I just have to not buy a single album, and not signal my endorsement when at clubs. But upholding this value could potentially be quite demanding. If all record labels have some artists who use such lyrics, will I have to give up listening to music altogether? What if I’m unsure whether a song has objectifying lyrics; am I obliged to actively examine lyrics any time I encounter a new song? Suddenly complying with this norm is going to take quite a bit of time, and I’ve got other important things to do. What if I’m not sure what counts as objectification? How much time and effort do I have to devote to studying feminist thought to understand? And what if my sources disagree? Working out who is reliable and where there is consensus is going to impose even more costs.

As even this admittedly simplified example shows, having identified a value or norm doesn’t always generate clear prescriptions for what one ought to do or how to weigh competing considerations.⁶ And thus for our purposes, it doesn’t give us a straightforward answer to what behaviours act as evidence of (in)sufficient commitment.

It is easy to think that the difficulty described above is primarily epistemic in nature, in that while there are answers to these questions, we are sometimes simply not well-placed to discern them. But a further source of difficulty comes the fact that many injustices have multiple contributing factors, which can make identifying particular factors as ‘the’ cause, or the most relevant cause, contentious (Smiley, 1992). Consider again the case of Gerri the politician. This seemed like an easy case to assess, as there was a clearly identifiable individual with the capacity to change things and few competing values. The problem is that even though Gerri ought to improve things, there are things that much less powerful individuals can do to produce the same results. Even if one is not a politician, since politicians are collectively elected, there are opportunities for collectives to produce the desired outcomes. As Collins (2013), Lawford-Smith (2015) and others have argued, for instance, when it is possible for collectives to bring about change, individuals can incur duties to collectivise: to come together and form groups who are capable of things that said individuals otherwise would not be. But how far can this go? Even if the existing system is very resistant and thus difficult to change, citizens can, in principle, work to change that system, given enough time.

The multi-factorial and casually overdetermined nature of many injustices means there are potentially limitless ways in which individuals can promote feminist ends. Though we can ward off demandingness worries by keeping in mind that agents can count as sufficiently committed to a value when an action upholding that value would be too difficult, costly or unreasonable, it seems there will often be other actions available that do not face the same costs, trade-offs, or lack of opportunity.

I believe we are now in a good position to understand why political movements are ripe for perceptions of hypocrisy. The issues that said movements focus on are often social, complex, difficult, and require lots of co-ordination across multiple levels of society to solve. Individuals making small contributions can at once seem trivial in isolation and powerful in aggregate, which makes it difficult to work out how to assess the effects of any given action, as well as what competing goods may be present. Interaction effects and causal overdetermination are frequently present, which make it difficult to identify sites at which efforts can be focused and clear norms can evolve. Additionally, many of these issues involve co-ordination problems: which option is best or most likely to produce change is often dependent upon what everyone else is doing.

All of these features make it difficult to assess what the relevant values require, what actions individuals should take to uphold them, how to weigh said values against other values or goods, and what capacities or opportunities agents have. Given this, it should be no surprise that individuals often feel somewhat hypocritical: it is much easier to identify an injustice and proclaim we should do otherwise than to know which form of doing otherwise is best. It should also be no surprise that individuals will regularly fall short of their ideals: their ideals concern issues that are big and complicated, which individuals can often only have limited effects on.

But this number of variables also means that there are a number of ways in which politically concerned agents can make a pronouncement, act inconsistently with said pronouncement, and yet not be hypocritical. Perhaps complying with the pronouncement would be too difficult or costly, or would require sacrificing something else which is genuinely more important. Perhaps other parties have a much stronger duty to change their ways in virtue of their position or circumstances.⁷ Perhaps one lacks the opportunity to comply, or is committed to other values which we think that they ought to prioritise. All of these possibilities mean that politically engaged individuals need not jump to feeling hypocritical simply on the basis of a perceived inconsistency. However, these difficulties cut both ways: sometimes we need to identify and call out hypocrisy. How optimistic can we be about our ability to overcome these difficulties? I believe we need not despair, as the aforementioned challenges can be overcome by focusing on the kinds of costs that would be entailed by the pronouncement.

Let’s return to Gay’s example. Suppose music which objectifies women is incompatible with feminist values, and should not be encouraged.⁸ This plausibly generates some duties, e.g. the artist ought not make music with those lyrics in the first place, or the record label ought not release said album, or consumers ought not incentivise this kind of music by paying for it. Perhaps these are all true at once.⁹ There is some uncertainty regarding who has what duties, how strong they are, and to what extent they might be outweighed by other considerations. But regardless, meeting those duties is going to require that someone incur costs to meet them. And we can ask whether people calling for change are committed enough to bear similar costs.

Suppose we think that individuals ought not incentivise such music, and thus ought not buy albums with said lyrics.¹⁰ If we then buy said music without an excuse, we will straightforwardly count as hypocritical. We are saying that this value is sufficiently important to require that agents refrain from incentivising said music, and then failing to do this. But suppose we instead think that change ought to come from the record labels and artists themselves, and this is the focus of our pronouncement. It’s plausible that our actions then won’t count as hypocritical. It would have to be the case, however, that the artists and labels would bear fewer costs than us by making change (plausible, given their high wealth), or that the costs they would have to bear are far more legitimate or reasonable to expect than those that would be faced by us as individuals (also plausible, given their power, capacity, role-specific duties, etc.).¹¹

We do need to be sensitive to costs said artist would bear that average listeners may not be facing. It is not uncommon for such lyrics to generate controversy, which in turn generates attention and sales. Expecting the artist to remove these lyrics is thus expecting them to incur much higher financial costs. To be clear, even if this is millions of dollars, this might be relatively little cost in a moral sense, since an established artist may have high wealth and thus be very able to bear those costs. But even so, there may be cases where an artist is facing much higher costs than we initially thought; we can imagine an up-and-coming artist who has many dependents, few other opportunities, and little experience navigating the music industry for whom changing their lyrics would be particularly costly. It is also not impossible that an artist could only succeed in the industry by including particular lyrics (‘sex sells’). In expecting the artist to not use said lyrics, we would thus be expecting them to bear the cost of the increased chance that they never ‘make it’.

This example shows how, in instances where we are unsure what prescriptions a certain value generates or how to weight incommensurate goods, we are nevertheless capable of assessing whether a pronouncement is hypocritical. In observing someone make a pronouncement, typically calling for change, we assess what would be required to bring about this change, or what kind of change they are calling for. We then ask what kinds of costs would be incurred, and assess whether the agent is disposed to bear similar kinds of cost (making adjustments for differences in the agent’s circumstances that affect the legitimacy of such costs). If they are, then they will not count as hypocritical. If they would not bear such costs, however, then it is unreasonable for them to expect others to do so.

4 Taking up the slack

Thus far we’ve identified several variables relevant for assessing commitment to a value, and noted cause for optimism, as there are many ways in which one can act inconsistently with their pronouncements and yet not be hypocritical. However, I believe there is one last factor that is particularly relevant to pronouncements in the political domain that we need to grapple with, which has not yet received attention in the literature on hypocrisy.

Previous analyses of hypocrisy tend to focus on relatively simple norms, such as prohibitions against murder or adultery, which are relatively easy for agents to avoid violating as there are often few competing considerations. We’ve seen that when it comes to political matters, things get more complicated as it isn’t clear how strongly our values prohibit particular actions, and there are often many competing considerations. Still, the overall account still works: if we can assess what the value requires, competing considerations and capacity to comply, we can assess whether someone is hypocritical.

The problem is that there will inevitably be cases where what costs we initially think are at the limit of being reasonable for people to bear may not, in aggregate, reach the threshold for what costs are required to bring about change. Thus far we have primarily been thinking about what costs could reasonably be expected in terms of difficulty, or interests, or the importance of the competing values: if φ-ing seems to violate the value, but not φ-ing seems like too much to expect, then an agent who φ’s does not count as insufficiently committed to the relevant value. But particularly in the political arena, where we are concerned with very difficult issues and calling for particular kinds of broad-scale change, we have no reason to think that once we have added up the costs that people can be expected to bear, we will thereby collectively have the means required to bring about change. This is especially the case if we haven’t done some empirical investigation into what kinds of changes would be required. We may weigh up the costs to individuals, the importance of any competing values or entitlements, take into account capacity and opportunity to comply, and yet come up short. Basically, given the way the world is, something’s gotta give. We need to keep one eye towards what our pronouncements entail.

To be clear, there are many issues where this is not the case. Consider again Gerri the politician, who could single-handedly improve things. Though many individuals working collectively could produce the same outcome, it could be that these costs would be so great that pronouncements to change things are directed at Gerri in particular. In such cases, it may be that ‘the buck stops’ with her. We can make pronouncements about what she should do, and blame her for not doing it. But there will remain a further question of whether we can make pronouncements about what other individuals should do in response to said failure. If one thinks the buck shouldn’t stop there, as the issue is sufficiently important that Gerri’s refusal to do her duty means others need to ‘take up the slack’, then one is committed to there being cases where what is required to create change outstrips what is reasonable to expect of individuals, and this is something our account must grapple with.¹²

Suppose we have a group of agents who seem, on a variety of measures, sufficiently committed to feminist values, and who are at the limit of what we currently think is reasonable to expect of them. However, this is not enough. Perhaps they are in a co-ordination problem: although they all share the same values and ends, they can’t work out what to prioritise and who to assign to particular roles, and extra work is needed to solve this. Or perhaps the ends they value can only be achieved through negotiation and compromise with another party that holds different values. Or perhaps a new issue has only just been discovered, and these individuals are now the only ones who can educate others about it. Whatever the problem, the point is that nothing will improve unless someone bears higher costs.

This puts us in a bit of a dilemma, which a commitment-based account of hypocrisy cannot, by itself, resolve. On the one hand, it seems that on most measures, these individuals are sufficiently committed to feminist values. They are conscientious, they care about the cause, they act in ways we would expect, and they are willing to bear some costs. It is these very values that make them want to say, ‘This status quo is not acceptable! Someone should do more!’. This counts in favour of thinking that pronouncements for others to do more are not hypocritical. But this means agents can make pronouncements about what other agents in similar positions to them ought to do, while refusing to do so themselves. Such behaviour seems like a paradigmatic instance of hypocrisy. They are calling for others to bear certain costs that they themselves are unwilling to bear. Given these considerations, what should we make of them?

I propose we can make sense of our mixed feelings with a distinction between hypocritical behaviour and being a hypocritical person (Rossi, 2020; McKinnon, 2006). In focusing on values thus far, we have been concerned with both actions and character, as both matter. But it seems that this is an instance where, though our feminists’ character and commitments in general are not in doubt, their actions are nevertheless incompatible with the relevant values. Are they hypocrites? Perhaps we don’t need a simply yes or no: we can say that their actions are somewhat hypocritical, without this making them hypocritical people overall.¹³

This is something to keep in mind when thinking about commitment and ‘bad feminists’. While ‘feminist’ seems like a property that is relatively stable, someone is a ‘bad feminist’ in virtue of failing to live up to one’s ideals or values, which can come about by far more temporary actions or omissions. This is plausibly why ‘I was a feminist for two days last week’ sounds a bit odd,¹⁴ but ‘I was a bad feminist for two days last week’ does not. In sliding between the two, one might erroneously take a single failure to mean one is a bad feminist (true, given one has fallen short of their values), which in turn means that one is thereby not sufficiently committed to feminist values (false, given many isolated failures are not diagnostic of insufficient commitment).

However, there is something to the thought that if one acts in value-discordant ways frequently enough, or with enough severity, while lacking a good excuse or justification (which, as we’ve seen, are not especially rare) then this could make someone a hypocritical person, and undermine their status as a feminist altogether. Ferracioli and Terlazzo (manuscript) argue that if a feminist is someone who simply believes in the equality of the sexes, then the concept ‘liberal feminist’ would be redundant. ‘Feminist’ would not be doing any work, since liberals already believe that everyone is equal. Ferracioli and Terlazzo propose that a better extension for ‘feminist’ is someone who is disposed to actively resist sexist norms and expectations, even when doing so is costly. This dovetails nicely with how I have been thinking of commitment: being disposed to take on costs is good evidence of commitment to values. Additionally, it helps make sense of the thought that someone who continually acts in ways that egregiously conflict with feminist values isn’t ‘really’ a feminist; rather, they are simply mistaken about their values. Their routine failure suggests that, despite their purported beliefs, they do not value equality sufficiently. It is worth noting that self-identified bad feminists who worry about this topic, or feel disappointed by their failures, are exhibiting precisely the kind of reaction we would expect someone who is genuinely committed to feminist values to experience. This worry is an expression of one’s concern for getting things right.

Some people may baulk at the idea that they ought not make pronouncements unless they are disposed to bear relevantly similar costs. While reminding readers that there will often be differences in circumstance meaning other parties ought to take up the slack first, or that the buck stops elsewhere, I think that this constraint serves an important purpose. In particular, it helps us co-ordinate our efforts, and ensure our pronouncements are the kinds of things that can be given uptake and be actionable. If pronouncements came with no requirement that one be disposed to bear similar costs, then it would be easy for agents—having met the minimal threshold for commitment in other ways—to make wildly impractical pronouncements and blame others for not meeting them. They would have little incentive to find out how the world works, to look for hidden costs that are unreasonable to expect others to bear, to find what particular causes are producing injustice, and what kinds of interventions will produce what effects.¹⁵ This, in turn, would mean that collectives would have little means of working out what kinds of pronouncements are going to be effective, where to focus their efforts, or which parties are particularly well-placed to remedy injustice. If any pronouncement is as good as any other, and no consideration needs to be given to what the pronouncement entails, then some independent mechanism for adjudicating between and prioritising certain pronouncements would need to be developed, and this would only be a further source of difficulty.¹⁶

Additionally, allowing pronouncements to be made independent of the agents’ disposition to take on costs is ripe for being taken advantage of by grandstanders (Tosi & Warmke, 2020). If people, having avoided charges of hypocrisy (due to being somewhat committed to the relevant values) now have no incentive to make their pronouncements targeted, achievable, or contain actionable content, then there is little to counteract their incentive to make strong but vague and unachievable pronouncements. By making a stronger pronouncement, agents can gain the benefits that come from appearing to have a strong commitment to the relevant values, without having to actually manifest any greater commitment. This can lead to ramping up, where, in order to avoid seeming uncommitted, agents in a community make stronger and stronger claims. This leads to another drawback, which is that too many implausibly demanding pronouncements leads audiences to become dismissive of certain categories of pronouncements altogether. It is possible for people to become desensitised to calls to action, on the basis that said calls do not provide solutions or do not grapple (or even acknowledge) the difficulties that would be faced in bringing about that change.

In contrast, requiring that agents be disposed to bear similar costs in order to make pronouncements helps identify and transmit particularly important information to themselves and others. If they are not willing to bear those costs, this is good evidence that the current approach is difficult, perhaps in ways that are not yet recognised, suggesting that alternatives may need to be considered. Perhaps sheer dint of will is not enough to comply with the value, and one’s continued failure to do so is good evidence that other forms of support are needed. But if I am disposed to bear those costs, this suggests the cause is sufficiently important to warrant bearing these costs, that others should be willing to bear them too, and that my current methods of upholding this value are feasible.

5 Hypocrisy in politics

Having made the case that we can, in principle, identify hypocrisy in political domains, it is worth considering whether other considerations might, in practice, warrant refraining from making a habit of calling out perceived hypocrisy.¹⁷ Perhaps a widely tolerant approach, one in which we give each other a strong benefit of the doubt, would be better all-things-considered. O’Brien and Whelan (2022) argue for the revisionary claim that we should not charge people with hypocrisy in the political arena. Even if hypocrisy itself is bad, they argue that it is often necessary for political success. In particular, people frequently need to act in ways that are not consistent with their values in order to gain power and create positive change. They may have to vote for a policy they despise for the sake of maintaining party unity and cohesion. More importantly, passing effective legislation often requires compromise with other parties, and insisting on purity can result in nothing improving.¹⁸

To help identify costs and benefits of different possible proposals, let’s distinguish between three kinds of standards in calling out hypocrisy: trigger-happy accusing, accurate accusing, and hesitant accusing. On the first standard, we are gung-ho about calling out hypocrisy, successfully identifying and calling out lots of hypocrites but also generating lots of false positives. On the last standard, we are reluctant to accuse, being mindful of the costs of getting it wrong. Though this avoids false positives, it also results in lots of false negatives. The accurate accusing standard is something of an ideal that avoids either extreme, by definition generating only true positives and true negatives. Let’s also distinguish between the accuracy and utility of our accusations, with utility being understood here loosely as positive or negative resulting effects. While there might be utility to accurately calling out hypocrisy, there could also be utility in refraining from calling out instances of genuine hypocrisy.

The thrust of O’Brien and Whelan’s argument is that we are mostly too trigger-happy with our hypocrisy accusations in politics, and we should instead adopt a more hesitant mindset which would produce far more utility. Even if we consciously aim at being accurate accusers, we will fail and be too trigger-happy: we are simply bad at identifying bona fide instances of hypocrisy (more specifically, instances in which it should be criticised). Additionally, even if we could be accurate accusers, there are other downstream costs which outweigh the benefits. As a result, we should have a very low prior on our ability to call out hypocrisy in a beneficial way. A default presumption against calling out hypocrisy is warranted.

Their wide-ranging argument has a number of fronts, and this acts as a useful way of considering several possible worries readers might have about identifying and calling out hypocrisy among bad feminists (environmentalists, activists etc.). To help organise classes of consideration, let’s first consider worries about accuracy, followed by worries about effects that hypocrisy accusations have on discourse, and then, finally, worries about overall utility independent of accuracy and discourse.

First, consider the worry that a habit of calling out hypocrisy leads to too many false positives. Politics requires apparent hypocrisy in a number of domains. As noted, we often (mis)identify as hypocrites people who are merely being pragmatic, compromising with others who have diverse views, or saying things they don’t believe to ensure party unity. People with power often need to occupy different roles, each with different demands. Politics can also require some amount of experimentation with new ideas and proposals that can then be learned from. These factors can all lead to perceived inconsistency. Additionally, leaders and others in the public spotlight are often under pressure to take strong positions, to be seen as doing something even when holding judgement and waiting for more information would be optimal. It can also appear hypocritical to make aspirational claims, which have as part of their nature that one will fall short of them. One can pronounce that we ought to end domestic violence in the next decade, even if it is certain that we will not succeed. But this need not be criticisable: such claims can inspire others, foster hope, and generate standards by which to hold people accountable.¹⁹

O’Brien and Whelan don’t provide a substantive account of hypocrisy, taking it to minimally involve an incoherence or mismatch between what someone says and does. Because they are mostly arguing for more hesitation on calling out, the distinction between perceived and actual hypocrisy doesn’t particularly matter for them; they are mostly concerned with the overall utility of our practices.²⁰ But I think that many of the risks they identify are less pressing once we have on the table alternative ways of reducing the number of false positives without generating a corresponding increase in false negatives.

Let’s distinguish between two ways a false positive may come about. The first occurs when one misunderstands what hypocrisy is. One might think it is merely saying one thing and doing another, for instance. The second occurs when one lacks relevant context. Though one understands what hypocrisy consists in, they lack information about a particular case that appears hypocritical but is not. These two forms of false positives call for quite different correctives. The first can be remedied through education. Simply helping people understand what is and is not hypocritical is going to substantially reduce this kind of false positive, thereby improving people’s accuracy, and it is clear that a commitment account of hypocrisy can fulfil this role. For example, most people understand that party or group cohesion is necessary to get anything done in politics. As a result, it is not too complicated to understand that favouring policies that are somewhat against one’s values in order to maintain party cohesion is not evidence of insufficient commitment, as doing otherwise would mean that one is no longer in a position to advance those values at all.

Likewise, in cases of compromise and working across the table, it is clear that these are chosen for pragmatic reasons rather than acceptance of the other group’s values and ideology. There is nothing hypocritical in saying, ‘We wanted Policy A but were unable to secure it, so we voted for Policy B which, while not perfectly in line with our values, was better than nothing at all.’ Similarly, when in a double bind, choosing the lesser of two evils is not evidence that one is committed to that kind of option. Replying, ‘But I had no other choice’ is an accurate and reasonable reply to accusations of hypocrisy in such circumstances. I suspect that most people already have an intuitive idea that values and commitments are what they are most sensitive to; even if compromises sometimes generate accusations of hypocrisy, these are typically much less frequent or severe than those that result when there were no competing considerations and the agent had the capacity and opportunity to comply with their espoused values.

Even if we can improve education about what constitutes hypocrisy, O’Brien and Whelan worry that the second cause of false positives—lacking relevant information—still results in us routinely getting things wrong. We often don’t know all the details of back-room deals. Sometimes it appears that compromise was not needed. There is something to the thought that, even if in principle we can educate, and outline exceptions, and hope that people notice the difference between merely reporting an opinion and giving a directive, they will systematically misidentify who counts as hypocritical. Similarly, people may not be attuned to the difference between merely reporting an opinion and arguing for a particular state of affairs, and so still wield accusations of a hypocrisy as weapon to silence dissent.

This is a genuine risk that comes with attempts to be accurate accusers. It is certainly true that we often lack all the details. But we should not extrapolate from this being true of some cases to assuming that we must therefore be equally as hesitant regarding all cases. At best, O’Brien and Whelan’s points favour habits of gathering more evidence before we accuse anyone of hypocrisy, or raising our standards for what counts as sufficient evidence, rather than a wholesale prohibition. Sometimes, the facts are clear: there were no competing considerations, and the politician or leader or feminist was simply inexcusably tempted by other considerations, suggesting a lack of commitment. Additionally, leaders often stay in their position for years, and this gives us quite a bit of data from which we can make inferences. One should not make too much of single actions, but as the agent’s track record of failing to uphold their professed values increases, we can increase our certainty that they are or are not hypocritical. When it comes to friends, peers, or fellow members of our cause, there is often the option of gaining more evidence: we can ask them questions, work out what their reasons were, see what kinds of incentives they respond to, compare them to others who are in similar circumstances, and, from there, using all of this evidence to make a judgement about what their commitments really are. While the problem of limited information certainly warrants humility and caution, this need not result in a broad scepticism of our abilities altogether.

Having established that worries about accuracy can be accommodated, the second class of concerns regards the effects that a general policy of calling out perceived hypocrisy has on public discourse. One effect is that accusations of hypocrisy can affect someone’s perceived right to make claims and participate in debate. This allows respondents to ignore, undermine or dismiss a claim without evaluating its content.

While this is a relevant consideration, its import seems weaker when we question whether the exclusive point of political debates and argument is to discern truth. Often there are multiple aims. When a politician or leader argues that φ should be done, this is not only an attempt to get you to believe a certain proposition. It is also an attempt to get you to endorse certain policies that ought to flow from that position. This need not be manipulative; we admire plenty of speeches throughout history and how they contributed to justice even if they achieved this through rhetoric and emotion rather than argumentative and evidential rigour.

Additionally, a common purpose of debates or arguments is to try and signal to audiences what the participants’ moral character are. As López-Guerra (2017) points out, political matters are often too complex for individuals to have informed opinions on everything. Instead, we find, elect, trust, or defer to people who we think share our values and who will act in our interests. Even if I’m an expert on a topic, I might be interested in these debates so I can assess what the participants think, not so I can update my understanding of the debate topic. Assessing what potential leaders think on a topic is often a way of assessing whether they share our concerns, rather than a means of considering what nuanced position is in fact the correct one. Sometimes audiences also use these forums to assess the strength of participants’ convictions or integrity, which cannot be evaluated through a simple evaluation of speakers’ arguments alone.²¹ O’Brien and Whelan reply that hypocrisy need not be evidence of unreliability, and the presence of reliability does not guarantee goodness of character. But the claims they are refuting are not universals, they are probabilistic, and so it will not suffice to find single examples of people who are hypocritical yet reliable. Since hypocrisy tracks insufficient commitment, and commitment is cashed out in terms of how one would behave in relevant circumstances, there is a tight connection here between commitment and reliability.

A related worry is that accusations of hypocrisy are often used to undermine standing to make claims in debates. Though this is true, the point can also be overstated. Attacks on standing don’t typically undermine one’s right to make claims simpliciter; rather, they undermine one’s right to have particular claims taken seriously. If the champagne socialist says, ‘After thinking about it, I see that I was wrong and I’ve changed my mind – capitalism is a better system’ or even simply reports, ‘I agree that caviar tastes good’, no-one has the right to accuse them of hypocrisy. It’s when they suggest something like, ‘We all ought to be doing much more to challenge capitalist injustices’ that outrage ensues. To be fair, perceived hypocrisy can sometimes undermine one’s standing to speak on a topic altogether. But that is typically because the topic of discussion has been narrowed down to a range of views taken to be incompatible with the speaker’s values. Potential hypocrites can often still make their point without criticism so long as they can acknowledge their perceived inconsistency, and thus limit the force of their pronouncement. Nuance and honesty about one’s level of commitment, even if they don’t remove people’s outrage entirely, certainly can temper it.

Providing a more nuanced look at kinds of pronouncements also helps us account for the intuition that sometimes agents who act inconsistently with their pronouncements are well-placed to advise others. On a commitment account of hypocrisy, an alcoholic is not hypocritical if they merely report what harms their alcoholism has caused them. Nor are they hypocritical if they request, or advise that others act differently. Likewise, it is not hypocritical for a bad feminist to merely report or inform others about how certain actions can be counter to feminist ends. Where these agents become hypocritical is when they do something stronger: they direct, demand, require, expect, and in the case of failures to follow such directives, they criticise, blame, or reproach. The reason for this is that the former set of speech acts do not signal anything about what standards the agent expects others to meet, or try to provide normative reasons for the target (Herstein, 2017; Piovarchy, 2020). While these responses might reveal something about my commitments, they are not communicating that I am going to uphold those commitments or am unwilling to tolerate violations of those commitments by others.

Having looked at worries about accuracy and worries about discourse, let’s finally consider the idea that being tolerant of perceived hypocrisy has some overall benefits, namely, that in some instances mere lip service is greatly preferable to wrongdoing that isn’t hypocritical. For example, it seems preferable for countries to espouse commitments to human rights even if they don’t always uphold them. The worry is that if we make hypocrisy too criticisable, we incentivise parties to abandon their good pronouncements in order to avoid appearing inconsistent.

Though I grant the point, I think this is a false dichotomy. It is rare that an agent has only two options: do something immoral and make a morally admirable but inconsistent pronouncement, or do something immoral and publicly endorse that position. And, if one has been forced into performing the immoral action no matter what, then one is in circumstances in which they already lack reasonable alternatives, and this is something which will defeat an assumption of insufficient commitment.

If we are to compare the overall benefits of various policies, we need to consider the benefits that come from accurately calling out hypocrisy (or even being trigger-happy accusers), as well as the costs that come from refraining from calling out hypocrisy. While the disutility of false positives are a worry, we need to also consider the utility gained from true positives and lost from false negatives.²²

I’ve already argued that identifying and calling out hypocrisy is particularly useful because it can help provide information about what kinds of approaches are feasible or more difficult than at first glance. But there is a second benefit that calling out hypocrisy has, which I think is particularly useful for instances where we already know what is required and have understood the costs and benefits of various proposals. This is that calling out hypocrisy helps us keep track of who has what commitments and values. Even if, in some cases, we will accidentally accuse people who are genuinely committed to the relevant values, the main idea is that over time it will be much easier for someone who is genuinely committed to change our perception than someone who is not. Calling out someone who is engaging in false pretence ensures that others are aware of this, and thus are not duped into relying on them.

This also ensures that agents do not free-ride on the reputations of others. If a political group is working towards some valuable end, the members of that group gain some credit or esteem from the broader community (Brennan & Pettit, 2005). But if agents can exhibit the markers of that group, without having to do the work, then they can benefit from that credit and esteem without having to bear any of the costs. Calling out hypocrisy helps deter agents from representing themselves as members of whatever group helps serve their selfish interests, and gaining the esteem that goes with it.

If agents know that they don’t need to have their pronouncements match their actions, this opens the door to them making whatever pronouncements will get them elected, or into positions of power, and makes it much easier for them to make excuses when they inevitably fail to deliver. Even in informal communities without hierarchy, individuals can play important roles in helping the group co-ordinate. Some people are counted on to play particular roles, some are authorities on how to weigh up competing considerations affecting the group, and some signal through their behaviour appropriate methods of achieving certain goals. Losing our ability to keep track of who is committed to what values is not only bad because uncommitted agents can get into positions they ought not hold, it makes it harder for us to collectively identify individuals who are committed to the relevant values and who can be counted on.

This state of affairs is particularly bad in settings that require a lot of ongoing co-operation. As noted, the complexity of political matters often gives us a strong interest in being able to identify leaders who share our values. And movements like feminism involve lots of relationship-building with other agents and working together, which requires knowing who can be relied upon. If I can no longer tell my Gloria Steinems from my grandstanding sycophants, it becomes very hard to get anything done. As a result, I’m more likely to be wary about being taken advantage of or wasting my time, and will stop trying certain things that require co-operation altogether. Pronouncements are not just lofty ideals, they are important signals about what someone is like, what they can be counted on to do, and thus what we can collectively achieve.²³

6 Conclusion

Many of us fall short of our political ideals, or fail to do what is required to uphold certain values. This can make us worry that we or our peers are hypocrites. In response, we might opt to instead never make any pronouncements, to ensure we’re not hypocrites, but this would radically reduce our capacity to make moral contributions. Alternatively, we might argue for a permissiveness and acceptance of our violations, but this threatens to express a lack of seriousness about our endeavours. I believe we can avoid either extreme by realising that our pronouncements are hypocritical when we are not sufficiently committed to the values our pronouncements imply or entail. Because there are many ways that agents can fail to uphold these pronouncements without this being evidence they lack sufficient commitment, many bad feminists need not worry about hypocrisy. For particular assessments, we need to take into account the importance of the value, our capacity to do what upholding that value requires, and whether other considerations are at stake. These factors are all particularly difficult to assess when it comes to pronouncements that concern political change, but we are not incapable of doing so. This framework not only applies to those with feminist values, it also helps us make sense of other political domains that require collective action.

We have seen that one particular difficulty arises in instances where creating change requires that someone take on particularly high costs. Whether one counts as a having sufficient commitment to make certain pronouncements depends upon what costs one reasonably expects other people to incur. But what costs we reasonably expect people to incur will in turn affect what kinds of feminist aims can be achieved in the first place. There is thus the possibility that some kinds of pronouncements necessarily mean that one is committed to thinking that people should be willing to incur unreasonable costs. This is as things should be: if I want many others to bear unreasonable costs, one of the best ways to solve co-ordination problems is to show that they’re the kind of costs I’m willing to bear too.