Written by Neil Levy
Originally published in Aeon Magazine
People engage in moral talk all the time. When they make moral claims in public, one common response is to dismiss them as virtue signallers. Twitter is full of these accusations: the actress Jameela Jamil is a ‘pathetic virtue-signalling twerp’, according to the journalist Piers Morgan; climate activists are virtue signallers, according to the conservative Manhattan Institute for Policy Research; vegetarianism is virtue signalling, according to the author Bjorn Lomborg (as these examples illustrate, the accusation seems more common from the Right than the Left).
Accusing someone of virtue signalling is to accuse them of a kind of hypocrisy. The accused person claims to be deeply concerned about some moral issue but their main concern is – so the argument goes – with themselves. They’re not really concerned with changing minds, let alone with changing the world, but with displaying themselves in the best light possible. As the journalist James Bartholomew (who claimed in 2015 to have invented the phrase, but didn’t) puts it in The Spectator, virtue signalling is driven by ‘vanity and self-aggrandisement’, not concern with others.
Ironically, accusing others of virtue signalling might itself constitute virtue signalling – just signalling to a different audience. Whether it should be counted as virtue signalling or not, the accusation does exactly what it accuses others of: it moves the focus from the target of the moral claim to the person making it. It can therefore be used to avoid addressing the moral claim made.
Here, though, I want to consider a different issue. In the only full treatment of the topic in the academic literature (that I know of), the philosophers Justin Tosi and Brandon Warmke accuse the ‘moral grandstander’ (their term for the virtue signaller) of perverting the function of public moral discourse. According to them, ‘the core, primary function that justifies the practice’ of such public moral discourse is ‘to improve people’s moral beliefs, or to spur moral improvement in the world’. Public moral talk aims to get others to see a moral problem they hadn’t noticed before, and/or to do something about it. But, instead, virtue signallers display themselves, taking the focus away from the moral problem. Since we often spot virtue signalling for what it is, the effect is to cause cynicism in the audience, rather than to induce them to think the signaller is so great. As a result, virtue signalling ‘cheapens’ moral discourse.
But Tosi and Warmke offer no evidence for their claim that the primary, or the justifying, function of moral discourse is improvement in other people’s beliefs or in the world. That’s certainly a function of moral discourse, but it’s not the only one (as they recognise).
Perhaps, in fact, virtue signalling, or something like it, is a core function of moral discourse.
Signalling is very common in nature. The peacock’s tail, for instance, is a signal of evolutionary fitness. It’s what biologists call an honest signal, because it’s hard to fake. It takes a lot of resources to build a tail like that, and the better the signal – the bigger and brighter the tail – the more resources must have been devoted to it. Stotting – a behaviour seen in some animals, involving leaping straight up in the air, with all legs held stiffly – is probably also an honest signal of fitness. The gazelle who stotts vigorously demonstrates to potential predators that it’s going to be hard work to run it down, which might lead the predators to look for easier prey. Humans also engage in signalling: wearing an expensive suit and a Rolex watch is a hard-to-fake signal of wealth and might help to communicate that you’re a suitable trading partner or a desirable mate.
In the cognitive science of religion, it is common to identify two kinds of signals. There are costly signals and credibility-enhancing displays. The peacock’s tail is a costly signal: it takes a lot of energy to build it and drag it around, and it gets in the way when fleeing predators. Credibility-enhancing displays are behaviours that would be costly if they weren’t honest: for example, the animal who ignores a nearby intruder not only communicates to group members its belief that the intruder isn’t dangerous, but does so in a way that certifies the sincerity of the communication because, if the intruder was dangerous, the signalling animal itself would be at risk.
Lots of religious behaviour can be understood as costly and credibility-enhancing signalling. Religions mandate many behaviours that are costly: fasting, tithing, abstinence from sex except in certain contexts, and so on. All of these behaviours are costly not only in everyday terms, but also in evolutionary terms: they reduce opportunities for reproduction, resources for offspring, and so on. Religious activities are also credibility-enhancing displays of religious belief: no one would pay these costs unless they really believed that there was a payoff.
Why, from an evolutionary point of view, would someone signal religious commitment? A likely explanation is that the function is to secure the benefits of cooperation. Cooperation with others is often a risky activity: there is the constant possibility that the other person will free-ride or cheat, making off with the benefits without paying the costs. The more complex the social group, and the easier it is to move between groups, the higher the risks: whereas in small groups we can keep track of who is honest and reliable, in a large group or when interacting with strangers, we can’t rely on reputation.
Signalling helps to overcome the problem. The religious person signals her commitment to a code, at least of cooperating with the ingroup. She signals her virtue. Her signal is, by and large, an honest signal. It is hard to fake, and religious groups can keep track of the reputation of their members if not of everyone else, since the pool is so much smaller. This kind of explanation has been invoked to explain the prominence of Quaker business people in the early years of the industrial revolution. These Quakers trusted one another, in part because involvement with the Society of Friends was an honest signal of willingness to abide by codes of ethics.
Religious signalling is already moral signalling. It is hardly surprising that, as societies secularise, more secular moral claims come to play the same role. Virtue signalling is supposed to be signalling to the ingroup: it shows that we are, by their lights, ‘respectable’ (in Tosi and Warmke’s word). That’s not a perversion of the function of morality; it is moral discourse playing one of its central roles.
If such virtue signalling is a central – and justifying – function of public moral discourse, then the claim that it perverts this discourse is false. What about the hypocrisy claim?
The accusation that virtue signalling is hypocritical might be cashed out in two different ways. We might mean that virtue signallers are really concerned with displaying themselves in the best light – and not with climate change, animal welfare or what have you. That is, we might question their motives. In their recent paper, the management scholars Jillian Jordan and David Rand asked if people would virtue signal when no one was watching. They found that their participants’ responses were sensitive to opportunities for signalling: after a moral violation was committed, the reported degree of moral outrage was reduced when the participants had better opportunities to signal virtue. But the entire experiment was anonymous, so no one could link moral outrage to specific individuals. This suggests that, while virtue signalling is part (but only part) of the explanation for why we feel certain emotions, we nevertheless genuinely feel them, and we don’t express them just because we’re virtue signalling.
The second way of cashing out the hypocrisy accusation is the thought that virtue signallers might actually lack the virtue that they try to display. Dishonest signalling is also widespread in evolution. For instance, some animals mimic the honest signal that others give of being poisonous or venomous – hoverflies that imitate wasps, for example. It’s likely that some human virtue signallers are engaged in dishonest mimicry too. But dishonest signalling is worth engaging in only when there are sufficiently many honest signallers for it make sense to take such signals into account. While some virtue signallers might be hypocritical, the majority probably are not. So on the whole, virtue signalling has its place in moral discourse, and we shouldn’t be so ready to denigrate it.
Dear Neil Levy,
Thanks for your piece. I really didn’t get your argument that, since virtue signalling is addressed to to the ingroup, then virtue signalling is moral discourse playing one of its central roles. Would you care to elaborate?
I havent’s read the Tosi & Warmke paper, but here are some points about moral discourse that immediately come to mind and I am a bit surprised you don’t mention them:
* The requirement of disinterest *
The instrumental value of virtues is that they can inspire people; however, inspiring people requires being relatable as someone genuine and authentic, as giving oneself in to a cause, project or value, and not as acting for purpose of inspiring others (Kant would say: it requires disinterest/disinterestedness); alas, virtue-signalling precisely works against giving oneself in to a cause, project, or value; hence virtue-signalling works against the instrumental value of virtue.
* The democratic usefulness of moral discourse in Western countries *
It’s not clear at all that virtues play any significant role in moral discourse. You don’t need to be an Aristotle scholar to observe that virtues are meant to be seen, felt, and hopefully acquired through imitation (with appropriate intent). Of course most of what people do happens in language, especially in our digital era, but it does not follow that the function of moral discourse is to express or convey the notion of people’s being virtuous. Instead, if discourse is just place like others where virtues can manifest, there is no reason to believe that one needs virtue signalling to see, feel or imitate people’s virtues as they manifest in discourse. In fact, the point above suggests that precisely the oppposite is the case.
To make things worse, it’s quite questionable that the moral discourse in Western countries would benefit from talk about virtues as much as it would benefit from say, talk about moral reasons (by this I mean principles, rules, norms, and values). Usually people want to know *what* they and others ought to do, individually and collectively, including States and companies, and they want to know *why* they ought to do this or that. I don’t see where talk about virtues has any role to play in this (virtues have a huge role to play in giving people’s motives, but that’s another question altogether). If that’s true you might actually want to get virtues aside and focus discourse on moral reasons. In particular when you want people to cooperate on complex and interrelated issues (climate change? animal pain? inter-human discriminations?), especially in democracies where the public is supposed to give law-makers intelligible reasons to operate with.
* Turning moral values into social values & group-based morality *
Perhaps the worse aspect of virtues-signalling is that it tries to turn moral values into social values, which has the effect of creating estranging people to one another. I am baffled you didn’t mention this when you talked about virtue-signalling as targetting the ingroup. In fact virtues signalling targets the outgroup just as much, and convey the presupposition that outsiders are less respectable. It pollutes rational discourse with biases, prejudice, and negative emotions (the insiders are likely to be perceived by outsiders as self-righteous and arrogant, while the outsiders are likely to be perceived as less respectable and epistemically or morally inferior — “they idiots really know nothing / don’t think things through / have no / a bad moral agenda”). This pollution of public discourse makes it even more difficult to have honest and rational discussion about complex stuff, and thus might actually work against getting people to cooperate and give law-makers intelligible reasons to operate with.
There’s a lot there, Géraud. I will just comment on one thing. Invoking virtue signalling doesn’t require taking a line on what is known in philosophy as virtue ethics. People don’t seriously mean that what is signalled is the possession of an Aristotelean virtue, or anything like that. I happen to agree that virtue ethics isn’t an especially useful approach to moral and political questions, but to signal virtue is just to signal a commitment to certain principles, or something along those lines.
Agreed. In fact none of the points I made above hinges on virtue-signalling being Artistotelian-virtue-signalling, I think. All that’s required is that the signalling person tries to present herself as occupying the moral high-ground.
I think this makes your position difficult to defend. Let me explain.
In the original post your write:
“Virtue signalling is supposed to be signalling to the ingroup: it shows that we are, by their lights, ‘respectable’ (in Tosi and Warmke’s word). That’s not a perversion of the function of morality; it is moral discourse playing one of its central roles.”
You didn’t really go into details but you did make the following points:
(1) signalling *in general* is valuable for membership identification: it makes the ingroup aware of one’s being a part of it, and therefore helps identify members vs. non-members;
(2) signalling in particular *by means of displaying one’s moral commitments* is valuable for ingroup status recognition: the reason is that virtue-signalling (when done honestly) often entails being a ‘core’ member — ‘core’ in the sense of adopting for herself and standing by commitments which are at the core of the group’s identity, and thus makes the virtue-signaller perceived by the ingroup as a reliable and stable member, not just as any member. That’s status right here.
You didnt’ mention a third point, which one might think also helps your position:
(3) that the effects of (1,2) might scale with the number of core members in proportion to (a) the ratio ‘core members/non-core members’ and in proportion to (b) the size of the group, so that if most members of a large group are core members, the moral commitments of the group will likely be perceived as vindicated by the outgroup (i.e. they will be perceived as worthy moral commitments for any group to adopt and stand by.)
Now even if all this is true, the problem for your position is this. Your claim that virtue-signalling is morality playing one of its central roles is premised on the following claim:
(*) The negative side effects of (1-3) on the outgroup are likely to be offset by the benefits of (1-3) for the ingroup, and perhaps, of (3) for the outgroup as well.
But (*) is not defended in your original blog post. By contrast if the points I made in the previous post together with the points made by Tosi & Warmke are on the right track, (*) does not look very promising.
I don’t see why I need the claim you label (*) at all. What role signalling is playing is an empirical claim. It may be true, regardless of whether the effects of VS are positive or negative.
The authors you are criticizing claim that the primary, or the justifying, function of moral discourse, is improvement in other people’s beliefs or in the world. This claim is normative; it specifies the sort of good that moral discourse *ought* to bring about, so as to perform the function it has (namely improve the world or other people’s beliefs). So if you want to defend the claim that virtue-signalling contributes to bringing about this good, you need to defend (*). Otherwise everything you say is compatible with what the authors say.
The function of x is y. The function of x should be y. These are different claims.
Okay, so everything you say is compatible with what these two authors wrote. They are claiming that moral grandstanding is not morally justified, so that, if the function of moral discourse is to bring good, then moral grandstanding perverts the function of moral discourse. And you are claiming that moral grandstanding is a manifestation of biological evolution, so that even if moral grandstanding is not morally justified, or perverts something, what it perverts is not the evolution-based function of moral discourse. Both claims are mutually independent. Mistake is mine — I thought you were debating with the authors.
Tosi and Warmke claim that the function of moral discourse is to point to moral features of the world, such that signalling is a perversion of its function. I claim that they’re wrong. I also claim that signalling need not have the features that would make it suspicious, like hypocrisy.
Not really. In fact, it’s an open issue whether they care about the function of moral discourse. They don’t even use the word ‘function’ in the article you are responding to — they simply note that moral discourse has a point, and that is to help people address and understand moral problems. The *point* that moral discourse has may or may not be its *function* in your sense — it’s just a norm that the authors claim moral discourse ought to meet. It is in view of that norm that they claim that moral grandstanding is morally unjustified; for them, it fails to meet the norm.
In my above remarks — premise (*) in particular — I’ve tried to show what a counter-argument to their claim would need. So it seems to me that either you are not really criticizing their claim, or you criticism lacks (*).
Comments are closed.