People often feel aggression towards things that make them feel guilt. Virtuous movements or individuals often cause people to feel guilt, by making them aware they are failing to meet their normative standards. Thus, virtuous movements are liable to be subject to guilt-motivated aggressive sentiment. Regardless of whether virtuous movements trigger guilt, they can also cause offence, by implicitly (or explicitly) questioning others’ moral rectitude.
I group the negative affect generated by both these processes under the heading “Virtue Aggression”. I suspect that Virtue Aggression has been a major hurdle for the vegetarian movement; and could prove a similarly serious issue for Effective Altruism. Whilst EA’s small size and lack of an “aura of judgement” reduce the risk it faces from Virtue Aggression; its theoretical demandingness, and the perceptions of hypocrisy such demandingness can create, are liable to generate potent Virtue Aggression towards the movement.
Finally, I explore the implications of Virtue Aggression for several key questions confronting Effective Altruism. Presuming we take the need to mitigate Virtue Aggression seriously, this will militate in favour of framing EA as an opportunity instead of an obligation, (continuing to) show openness to shallow commitments, and pursuing a narrower and more selective outreach strategy.
At that time, the Emperor was embarked on the construction of a great palace, and the land was ravaged for his ambition. Young men whose hands were needed in the fields, quarried marble instead; and villages starved from the imposts seized to feed them. So it was that a petitioner came to the Emperor, to beg an audience from his august majesty.
The Emperor listened to the man’s dreadful tales, but only laughed. “I care not” he said, “you would do better to speak to my Chancellor – he is a man of conscience, perhaps he shall take pity on you?”
So the petitioner sought audience with the Chancellor. No sooner had he begun his story: of fallow fields, and starved babes, loyal subjects brutalised and abandoned, did the Chancellor begin to cry. His eyes reddened, he stifled sobs, his fingers raked his garments in distress. At last, the petitioner reached the end of his plea, and waited hopefully for a reply.
The Chancellor wiped away tears and gasped for breath. “You wound me!” he shouted at the petitioner “can you not see the suffering you have caused, the sleepless nights, the pain that I shall carry to my very grave! Rebel! Assassin! Torturer!”. “To the gallows with him!” he said to the guards; and the petitioner was not heard from again.
A key cause of guilt (what we may cause “Moral Guilt”) results from the awareness of failure to meet normative standards.
Such Moral Guilt, like all strong negative emotions, powerfully motivates the sufferer to reduce their suffering. An individual can resolve moral guilt in multiple ways. Non-exhaustively, these include:
Taking these methods in turn, it should be noted that Methods 1 and 2 are often difficult or impossible. A normative code is ingrained by peers and education over an extended period, and can be exceptionally difficult to shift, especially at the intuitive level (it is no doubt easier to “intellectually” assume a new moral code, than to develop the emotional habits it enjoins, as every budding utilitarian discovers). Behavioural change can also be troublesome[1]: habits are difficult to disrupt, character changes slowly when it changes it all, and circumstances may compel normatively improper behaviour.
Given the difficulty of Methods 1 and 2, the sufferer of moral guilt can easily be motivated to resolve their moral guilt through Methods 3 and 4.
Method 3 can involve changes to behaviour, such as a change of job that removes the sufferer from normative dilemmas which they reliably fail. Alternatively, more dramatically (and rarely), it can involve the destruction of the source of moral guilt: either the “victim” class, who the sufferer’s moral code would ordinarily oblige them to assist; or the bearer of bad news, who draws the sufferer’s attention to their moral wrong (the “bearer”).
Method 4 can operate through distraction or avoidance. Alternatively, it can operate through motivated reasoning, which reconceptualises the sufferer’s situation into one where their actions are normatively satisfactory. This often involves stigmatisation of the victim or bearer class. Stigmatisation operates through distraction and (what I term) “moral recission”. The distraction mechanism is clear: if, as soon as you think about the victim or bearer, you think about their moral wrongdoing, you are no longer considering your moral wrongdoing. The “moral recission” mechanism can be understood as the idea that, if someone is a bad person, they cease to deserve moral regard; or cease to be able to make otherwise morally legitimate claims. To take one example; you would probably be less likely to risk your life to save a drowning Kim Jong Un, than you would a random swimmer; or to accept moral advice from Kim once you had saved him, even if such advice were impeccably reasoned.
Both these methods lend themselves to aggression towards Innocent Parties (the “victim” and the “bearer”). Taking Method 3 first, actual destruction of innocents to expiate moral guilt is thankfully rare, but does in fact happen (as we shall see). What is more common is probably the awareness that destruction would expiate guilt, with a consequent (more or less successfully suppressed) aggressive desire to effect such destruction. Method 4 operates more indirectly; but suffice it to say that people often feel aggression towards morally stigmatised individuals. The greater the stigma a morally troubled individual places on Innocent Parties, the more effective the reduction in his guilt, but also the greater the increase to his aggression.
The direct and indirect consequences of these methods for removing guilt, combine as potent generators of aggression towards Innocent Parties: what I term guilt aggression.
One fascinating example of guilt aggression derives from early modern English witch persecutions. In the words of Keith Thomas, “the overwhelming majority of fully documented witch cases fall into this simple pattern”[2]:
There are many distressing and interesting elements of this pattern. But one which leaps from the page is the link between the guilt felt by the accuser towards the accused, and their – sometimes fatally (!) – aggressive emotional response. Here both the postulated pathways of guilt aggression are at play.
Having demonstrated the link between guilt and aggression, we can now progress to establish that between virtue and aggression. This link has two prongs: guilt-based virtue aggression (which results where a person is “virtuous” by moral standards you share), and self-worth-based virtue aggression (arising where a person is “virtuous” by moral standards you do not share).
Perception of others’ virtue can cause moral guilt; by making individuals aware that they are failing to follow their normative ideals[4][5]. Crucially, people may not be aware of the implications of their normative ideals, or may just not pay attention to known implications. When a virtuous person shows them, or reminds them, of these implications, this must initially create a gulf between the convinced persons’ normative actions and beliefs, generating guilt. To take one (class of) example, people typically believe that they are not morally obliged to do the impossible. If no-one around them does a normatively demanding behaviour, they may implicitly assume that such behaviour is impossible. However, when they meet a virtuous person who does perform this behaviour (or, worse still, a movement of numerous such people), their belief in its impossibility becomes less tenable.
Thus, a virtuous person can be the cause of guilt. As a cause of guilt, we should expect them to be subject to aggressive sentiment; for the reasons elucidated in the first half of this essay. Such aggression is guilt based virtue aggression.
A seemingly similar, but importantly distinct, cause of aggressive sentiment towards the virtuous, derives from individuals’ desire to preserve their self-image (and thereby, their sense of self-worth); coupled with the fact that most individuals believe and desire themselves to be virtuous.
When someone promulgates norms which would render you unvirtuous, or reminds you that you are already unvirtuous by your own moral standards, it is easy to perceive this as a personal attack or accusation. People tend to respond poorly to such attacks, including by aggression towards the virtuous person, the author of this assault on their esteem. This is self-worth-based virtue aggression.
Notably, unlike with guilt-based virtue aggression, this cause of virtue aggression can exist alongside the absence of any commitment to the normative ideals of the “virtuous” person. All that is required is for the aggressor to perceive that their behaviour would fall short of the ideals the virtuous individual espouses (by word or deed), and that such words or deeds therefore constitute an implicit attack on their own virtue.
I have always been struck by the remarkable levels of hostility towards vegetarians and vegans. There are the jokes (“How do you tell someone is vegetarian? Don’t worry, they’ll tell you!”). There are the absurd stock characters in television and films. There are articles indicating that attitudes towards vegans are only slightly more favourable than those towards drug addicts[6]; and plenty of journalistic comment on the trend[7].
In many ways, this is an archetypal case of virtue aggression. Most people in the West have a concern for animal welfare; but also consume animals. The presence of vegetarians and vegans reminds Western omnivores of the tension between their ideals and their actions; generating guilt. Some (small) percentage of people may be motivated by this guilt to become vegetarian or vegan themselves. But behavioural change is hard. Therefore, the majority need find other ways to deal with this guilt. A common strategy is to ascribe a host of negative traits to vegetarians; both borne out of, and itself intensifying, negative affect towards them.
At the opposite end of the spectrum to the conflicted liberal omnivore is the unconflicted conservative meat-lover, for whom animal consumption is a core element of their identity. Even if such individuals have less personal concern for animal welfare, and thus are less affected by vegetarian induced guilt; the ideology of vegetarianism reclassifies a core part of their identity as morally vicious, thereby assaulting their self-worth, and triggering an aggressive emotional response.
From the vegetarian perspective, this is a terrible position for public sentiment to be in. How many people will want to become vegetarians in a culture where they are continually mocked and belittled – even in supposedly liberal circles? The answer, it seems, is not many. According to Bruce Friedrich, executive director of the Good Food Institute, the number of people who were either vegetarian or vegan “hasn’t changed in 30 years”, instead hovering around 2% of the population[8].
Three elements of vegetarian virtue aggression are especially interesting from the perspective of effective altruism. The first is that greater virtue may inspire even greater aggression. Notably, vegans attempt a more “virtuous” diet than vegetarians (in the sense that their diet is more practically demanding, and based off a more extensive application of moral concepts incipient in vegetarianism). They are also perceived more negatively than vegetarians, including by vegetarians themselves[9]. It is equally interesting to see that those following a plant-based diet for health reasons, rather than ethical (animal welfare or environmental) reasons, tend to be perceived more positively: despite the fact that the former motivation is egoistic, whilst the latter motivations are altruistic, and hence (naively) more praiseworthy on their face[10].
The second fascinating element of this case study is how aggression inspires negative stereotyping and recrimination within the community of vegans and vegetarians. One common strategy to avoid stigma is to seek to differentiate yourself from the stigmatised group you are a part of: to say in effect “I may technically be a [stigmatised person], but I’m not one of those [stigmatised people]”. Within the vegetarian community, this mostly appears to take the form of vegetarians expressing negative attitudes towards vegans, as has already been touched on above. This is an exceptionally perverse state of affairs: it is bad enough when your group is subject to stigmatising attacks from society, without having to deal with stigmatisation from co-members! This perpetuates demeaning attitudes towards vegans and vegetarians: you don’t even have to listen to conservative cable to get attacks on plant-based diets; you can get them from your local vegetarian. It also plausibly reduces the capacity of those with plant-based diets to coordinate to achieve their shared goals[11].
The relevance of virtue aggression to effective altruism should at once be clear. Many Effective Altruists act in conventionally virtuous ways, and/or in ways which explicitly or implicitly endorse a morality far more stringent than the conventional alternative[12]. On the model developed above, this should generate virtue aggression.
To the extent it does, this could prove a major problem for the movement by:
Once this potential danger is acknowledged, two crucial questions confront us. First, how large might this risk be? Second, what can be done to minimise it?
There are many “virtuous” movements in society; some of which are extremely successful (religious organisations are a historic example, and liberatory social movements are a recent one). Although such movements have triggered, and continue to trigger, virtue aggression, this has evidently failed to seriously impede their cultural reach. It is thus important to consider what factors should be expected to incline a movement to receive greater virtue aggression, and analyse the extent to which they apply to EA.
Summarising the above discussion, two factors – small movement scale, and lack of a “judgmental aura” - reduce EA’s propensity to trigger virtue aggression. However, two other factors – the theoretical demandingness of the movement, and perceptions of hypocrisy – markedly increase this propensity.
Unfortunately, the former (risk-lowering) factors are likely contingent: as a young movement, EA may well continue to grow in size and support; whilst we have already seen that an aura of judgment could be bootstrapped into existence by other grievances. Contrastively, the latter (risk-amplifying) factors are fundamental: as a movement attempting to maximise good, EA will always have a high theoretical ceiling of demands; and those high theoretical demands will always provide ample grounds for allegations of hypocrisy.[20] Therefore, EA will be (and may already be) subject to materially problematic levels of virtue aggression, which could dramatically increase over time, and pose a serious threat to the movement.[21]
There are strong theoretical reasons to be concerned about the threat posed by Virtue Aggression to Effective Altruism. If we take Virtue Aggression seriously, what implications does it have for the EA movement? What follows are my thoughts on the relevance of Virtue Aggression to several key questions within Effective Altruism:
1. Opportunity v Obligation Framing
There is some debate in the EA community over whether EA should be framed as an “Obligation” or an “Opportunity”[22][23]. Of these two framings, the “Obligation” framing is most liable to provoke virtue aggression: as it implies that ordinary people are morally blameworthy for failing to act to prevent avoidable suffering, generating guilt and/or attacking their sense of self-worth. Contrastively, an “Opportunity” framing does not imply any moral blame (instead conceptualising altruistic actions as supererogatory goods, or even as non-moral preferences people are free to take or leave – even if we would prefer for them to take them). Additionally, an “Opportunity” framing helps defuse charges of hypocrisy (as if you’re not obliged to do everything you can, your failure to do so can’t be held against you), and the “aura of judgment”, which are key amplifiers of virtue aggression. Thus, “defusing virtue aggression” is an important factor supporting an “Opportunity” framing of EA ideas.
2. Deep v Shallow Commitment
The deeper the commitment required by a movement, the greater the implicit attack mounted by the movement’s ideas against conventional morality, and the greater the virtue aggression it is liable to create. In the case of EA, focusing on smaller and more achievable goals; or large goals with fewer practical implications for the average person (e.g. existential risk and AI), could thus diminish virtue aggression. This is, of course, a double-edged sword: a corollary of smaller goals, may be smaller achievements. The ideal would be a mechanism whereby the messaging directed to each individual interacting with the movement, matches the level of commitment they are open to adopting.
3. Broad v Narrow Outreach
Another area of debate in the EA community concerns whether the community should be broader or narrower[24][25]. My natural sympathies lie towards the “broader” end of the debate; however virtue aggression is an important factor supporting a “narrower” approach. The reasons for this are as elucidated earlier in the article. Presuming that EA is the kind of movement liable to trigger virtue aggression, expanding it to a wide array of people will likely generate greater virtue aggressive blowback[26]. Moreover, broader outreach will bring EA ideas to people whose values and behaviour are less aligned with EA principles, triggering disproportionately greater virtue aggression than earlier and more targeted outreach[27].
Virtue Aggression is an easily overlooked influence on the success of moral movements. But, as the example of vegetarianism shows, it can be a crucial one. So far Effective Altruism has – for the most part – calibrated its messaging in a manner that defuses virtue aggression. However, its tiny size has also shielded it from much potential criticism. Unfortunately, the movement bears a potent catalyst for virtue aggression at its heart – a maximising drive that can easily be interpreted as demanding and hypocritical. As EA grows, it may prove increasingly difficult to deflect virtue aggression generated by this core theoretical commitment. As virtue aggression can bootstrap itself into a vicious cycle of intensifying opprobrium, it is important to consider how to minimise it (without compromising the movement’s purpose), sooner rather than later. Framing EA as an opportunity rather than an obligation, making more room for shallow commitments, and avoiding hasty outreach to unamenable constituencies; are all concrete proposals to avert Virtue Aggressive backlash. Whether these proposals are worth their cost, is a matter for each of us to consider.
[1] For an example discussion: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4931896/
[2] Religion and the Decline of Magic, by Keith Thomas, pg. 661
[3] Ibid, pg. 663
[4] Indirectly, this may be because the virtuous convince people of new and more stringent normative ideals: as they do this, the gap between peoples’ behaviour and normative ideals must initially widen, presuming that beliefs and sentiments about norms change faster than behaviour, which seems a reasonable assumption in most cases.
[5] Elsewhere, the causal mechanism has been construed as based in cognitive dissonance alone. No doubt this is a large part of the explanation. For the purposes of “virtue aggression” as a concept, both causes operate in a similar manner – the provoking of a powerful self-directed negative emotion, which the sufferer externalises/combats through motivated reasoning against the virtuous group.
[6] https://journals.sagepub.com/doi/abs/10.1177/1368430215618253
[7] https://www.theguardian.com/lifeandstyle/2019/oct/25/why-do-people-hate-vegans; https://www.vox.com/future-perfect/2018/11/2/18055532/vegans-vegetarian-research-uk
[8] https://80000hours.org/podcast/episodes/bruce-friedrich-good-food-institute/
[9] file:///C:/Users/benja/Downloads/sustainability-11-06844-v2%20(2).pdf; https://www.researchgate.net/publication/330014790_If_I_became_a_vegan_my_family_and_friends_would_hate_me_Anticipating_vegan_stigma_as_a_barrier_to_plant-based_diets (warning: small n study with WEIRD participants).
[10] https://journals.sagepub.com/doi/10.1177/1368430215618253
[11] Followers of plant-based diets who make negative differentiating statements about other members of their group, may internalise those statements, thus reducing their desire to work with other group members (and perhaps making them resentful of their co-members’, whose supposed stigma-generating behaviour is causing them to also be subject to stigma). Conversely, followers of plant-based diets who do not make such negative differentiating statements may reasonably be leery of cooperation with those who do (for instance, vegans may be wary of cooperating with vegetarians who disparage them).
[12] For instance, effective altruists may donate a large portion of their salary to charity; choose careers based on what will help humanity; choose charitable work based on what is most “effective”, rather than what is most “affective”; etc.
[13] Although, N.B., this is not to say that it is not important – quite the reverse.
[14] A useful analogy here concerns the discourse about vegans and vegetarians “banning meat” in the United States. This is not something seriously proposed or considered by the Democratic party or its outriders. Nevertheless, right wing media vigorously propagates the meme that followers of plant-based diets (and, by extension, the left wing who they are associated with) want to ban meat. To steelman this meme (which in its vulgar form is patent nonsense), consider the following conversation between a vegan and an omnivore:
V: “I don’t eat meat because I believe animal suffering is wrong. Farmed animals are treated in the cruellest ways imaginable, and I can’t be complicit in that evil.”
O: “What, you want to ban my meat?”
V: “When did I say that? Of course I don’t want to ban meat. Everyone has a personal choice about what they eat. I’ve made my choice, but I won’t force it on others.”
O: “Hmmmm… I’m not so sure. You say you don’t want to ban meat; and to be fair to you, I haven’t seen you going round attempting to shut down farms or close my butcher. But if I thought something was a terrible, overriding moral evil, I’d absolutely want to ban it. What’s more, my feelings would reflect a deep-seated element of human psychology, one that people can’t get rid of. You might say you wouldn’t ban meat, and you might even believe it (right now, whilst you’re in the minority). But if your movement ever came to political prominence, I bet your reticence would fall away in no time. So I still regard you as a threat to my way of life, and I’m still going to treat you accordingly.”
The omnivore has a point.
[15] Real utilitarians do not practically spend every minute of their lives acting with the end of maximising happiness, or working through elaborate calculations about the consequences of wearing a jumper or a jacket.
[16] https://journals.sagepub.com/doi/10.1177/1948550611415695
[17] The argument that vegans exploit bees, because bees are needed to pollinate crops they consume, for instance. This ignores the fact that vegans may not believe insects are conscious, that the bees may do their pollination freely and without exploitation, and that pollinating bees living in the wild are not subject to the same cruel practices as e.g. factory farmed chickens.
[18] Imagine you are a stock social democrat, who believes that taxes should be higher, kids from poorer areas should be better educated, and there should be greater racial equity. Theoretically one could argue they are a hypocrite for not personally paying the government more money, for not spending their evenings attempting to better educate poor children, and for not donating their remaining untaxed income to marginalised communities. However, such arguments are rare.
[19] The following 2019 survey estimates a total of 5,000 – 10,000 EAs globally; of which 2,000 – 3,000 are “highly engaged” (although the reach of EA ideas, and “EA sympathisers” is no doubt higher): https://forum.effectivealtruism.org/posts/zQRHAFKGWcXXicYMo/ea-survey-2019-series-how-many-people-are-there-in-the-ea
[20] The above discussion covers theoretical reasons for EA to be vulnerable (or resistant) to virtue aggression. But this leaves open the question: do we already have practical evidence of virtue aggression directed towards EA? One tool for diagnosing virtue aggression could be to examine the gap between how much we would naively expect a belief to offend people, and how much it does in fact offend them. Taking vegetarianism as an example, on its face there is little to protest (who could object to a personal choice taken to reduce the deaths and suffering of animals?). However, in actuality, people bear vegetarians substantial negative affect, and justify that based on fairly tendentious reasoning (nebulous allegations of militancy, asceticism, hypocrisy, etc.). In theory, the core of Effective Altruism (encouraging people to take personal choices to maximise the good they do with their lives) seems similarly unoffensive. However my experience is that there is a surprising amount of opposition to EA, much of which seems strangely impassioned (for a good example of the genre, see: https://ssir.org/articles/entry/the_elitist_philanthropy_of_so_called_effective_altruism). This gap between naïve expectations of aversion levels, and actual aversion levels, could evidence virtue aggression against EA. However, it is important to note that there are many other factors that can account for opposition to EA: from simple disagreement with non-commonsensical philosophical principles, to poor messaging and runaway memes (e.g. “earning to give”), to the low quality of public discourse generally.
[21] It must be said that blaming ones’ failure to convince on others’ moral inadequacy, is far more gratifying (and thus undeservedly easy to accept) than blaming oneself! We should be careful to avoid “Principal Skinner-ism”.
[22] http://lukemuehlhauser.com/effective-altruism-as-opportunity-or-obligation/
[23] https://forum.effectivealtruism.org/posts/CFjvywYNWkqGtQiZh/altruistic-motivations
[24] https://www.1fortheworld.org/blog/2020/2/26/effective-altruism-broad-or-narrow
[25] https://80000hours.org/2020/02/anonymous-answers-effective-altruism-community-and-growth/
[26] As a more popular movement represents a psychologically and materially greater challenge to the morality of those who reject it.
[27] Presuming EA outreach coordinators are rational and informed actors, who first target their outreach towards the most EA-value aligned individuals, and only subsequently target less EA-value aligned individuals.
