'Essays on Longtermism' Competition 

(For TL:DR, see Summary and conclusions) 

Introduction

 This essay is written in response to Essays on Longtermism: Present Action for the Distant Future. For succinctness, I will refer to this volume as ‘the Book’ and the essay chapters collectively as ‘the Essays’. Any references in this essay to ‘chapters’ refer to chapters in the Book. 

I seek to engage with arguments raised in the Book, using perspectives from certain chapters and some themes from across the Essays as lenses for considering whether present action to increase compassion may be a worthwhile action (or at least worthy of further discussion and exploration) as a potential means of improving outcomes in the distant future – and / or for it’s potential nearer-term benefits. 

I note here at the outset that this in my first essay in >10 years and my first ever in the realm of either philosophy or ethics, so this may not read like a typical EA forum post or essay on long-termism. However, as a strong believer in substance over form, I hope readers will look past the style of writing (and lower density of references^[1]) and consider seriously the potential benefits of present action to increase global compassion as an effective strategy for improving outcomes in the both the distant future and the nearer term. With apologies in advance to readers better-read than I for any errors or oversimplification in my discussion of moral arguments.

In this essay, I will first set out my definition of compassion, why compassion is important (for humanity, EA and discussion of longtermism) and why, in my view, there should be more discussion of increasing compassion. I then consider increasing compassion as a present action for the distant future in the context of some of the arguments raised in chapters of the Book, and provide some discussion of the neglectedness and tractability of action to increase compassion. 

I am anticipating readers will have prior knowledge of the Book and the wider discussion on longtermism that the Essays are part of. For any readers interested in this essay who are not familiar with the Book, the full text can be found here. 

A need for greater discussion of action to increase compassion

Defining compassion

The Latin roots of the word compassion, literally mean “to suffer (together) with” and although there is no globally agreed definition of compassion, most definitions refer to sympathy for others' distress together with a desire to alleviate it. As per the Greater Good Science Center, “Among emotion researchers, it is defined as the feeling that arises when you are confronted with another’s suffering and feel motivated to relieve that suffering.”

For the purposes of this essay, I am defining compassion as empathy-driven action, and when I refer to compassion, I am including within that the component of self-compassion, in which the action is focused on extending the same kindness to yourself as you would to others in the same situation. It is important to be clear that what I refer to compassion, this is distinct from empathy in that compassion includes acting on feelings of empathy. 

In line with the requirement for action, I am not viewing compassion as an emotion, but as a skill which can be taught, practiced and strengthened. With this context, in referring to ‘increasing compassion’, I am referring to strengthening the skill of compassion in the human population via training and practice.

The importance of compassion for Effective Altruism, humanity and discussions of the far future

In Zachary Robinson’s opening talk at EA Global: London 2025, he explained “how EA principles serve as tools for putting compassion intro practice, translating our feelings about the world’s problems into effective action.” And within this speech, he makes explicit that a distinguishing feature of the EA community is “that we don’t just stop at feeling. We act.” Compassion – acting on feelings of empathy – is a key driver of the EA movement. It compels people to act to reduce suffering and inequality; and to seek out the most effective ways to improve the lives of other beings. Without compassion, there seems little reason to consider our potential impact, or whether we can improve outcomes in the far future. Compassion drives donations of money and time, both to EA organisations and more generally, motivating giving pledges and volunteers.  

Compassion is also argued to be a key factor in the success and survival of humankind; Charles Darwin argued in the Descent of Man that human cooperation and compassion was key to human survival; Dr Chris Kukk, an author and political science professor, summarises in his 2017 article:

Darwin’s research shows that “survival of the kindest” is more correct for explaining which species climb the evolutionary ladder efficiently and effectively.

…According to biologists from Darwin to E. O. Wilson, cooperation has been more important than competition in humanity’s evolutionary success. Compassion is the reason for both the human race’s survival and its ability to continue to thrive as a species.

Darwin was very clear about the weakness of the survival-of-the-fittest argument and the strength of his “sympathy hypothesis” when he wrote: “Those communities which included the greatest number of the most sympathetic members would flourish best and rear the greatest number of offspring. What Darwin called “sympathy” in the words of Paul Ekman, “today would be termed empathy, altruism, or compassion.”

Considering moral progress specifically, we can see the advantages that compassion brings humanity and it’s role in progress. Significant moral trajectory changes of the past such as the abolition of slavery, could not happen without compassion – specifically compassionate action, not empathy alone. Ongoing moral challenges where we seek shifts, such as animal welfare, global poverty and climate change require more compassion to make progress. Problems – such as AI governance, nuclear disarmament and biosecurity – that require cooperation and agreement between parties (nations or otherwise) with differing interests are easier to progress on if all parties show compassion to each other. 

Compassion – perhaps unsurprisingly as an underpinning driver of morality discussions – is also baked into discussion on the future of humanity, including within the Book. However, this can be easy to overlook as assumptions on compassion aren’t always explicitly drawn out. Considering chapter 11 for example, Powell talks of a ‘moral void’ in the absence of humans, but this presumes that humans retain an ethical stance, which I would argue requires humans that retain the skill of compassion. If the values that some of the moral arguments place on human survival rely on the assumption that future populations will be compassionate, surely we need to take action to ensure that future populations are compassionate? Although humans have an innate predisposition to compassion, it’s plain to see from current wars and political discourse around the world, that this predisposition can be overridden and we shouldn’t take for granted that future humans (or digital beings influenced by human values) will behave in a compassionate way.  

Considering action to increase compassion in the context of essays on longtermism

Is increasing compassion a longtermist good? 

In chapter 19, consideration is given to what a longtermist society might look like and the expectation that more resources would be allocated to “longtermist goods: broadly understood, activities which aim to directly protect the potential value of the longterm future.”, with recognition that in practice there are longtermist foundations for goods that are usually understood as purely neartermist. 

I suggest that increasing compassion is one of those investments in the welfare of present-day people that though might be traditionally viewed as a purely neartermist endeavour, would actually be a longtermist good for the following reasons:

• Increasing compassion has been shown to increase happiness and increased happiness leads to increased productivity;
• This productivity benefit may be more pronounced in relation to activities performed by those motivated by moral causes, as training in compassion (including self-compassion) is likely to reduce burnout in this sub-population whose dedication can typically put them at higher risk of burnout or overwhelm;
• Increased compassion is also likely to lead to increased cooperation, increased giving (both donations and volunteering time) and increased interest in moral considerations;
• Increased compassion is likely to make expanding the moral circle easier and to generate more support for the specific moral consideration of how to improve the lives of future generations;
• The combination of increases in productivity, cooperation; donated resources; interest in morality and future outcomes, is likely to accelerate progress on work dedicated to improving longterm outcomes.
• Additionally, if increased compassion can become part of the culture and is sustained through generations, it is likely to have a positive impact on what the longterm future looks like.
• In the context of a period of time where we may be more susceptible to values lock-in, the present era may be a particularly beneficial time to promote the skill and culture of compassion. 

If the push for compassion increase is seen as an EA/longtermist drive, you could even argue that there may be additional fringe benefit of challenging the narrative of EA as cold and uncaring, potentially mobilising more hearts, minds and resources for longtermist (or wider EA) causes.

Even if increasing compassion failed to motivate any additional support for longtermist goals (which I think is unlikely); the reduction in current suffering that could result from a generalised compassion increase is likely to result in resources freed up in future for other causes (and create a compassion virtuous cycle e.g. of those thankful for donations they received). 

Further, disillusionment and threats to civil society breakdown today are driven by inequality and by those on the sharp end of inequality (by their local comparators) feeling left behind and uncared for – as compassion becomes more and more prevalent, every individual should be feeling more cared for; compassion promotes – and requires - caring actions. If far more of these are taking place in every community, it’s very possible that the ‘what about me and mine?’ objections to helping those further afield (whether in distance or time) dissipate.

Linked to it’s potential noted above to accelerate progress on solving large problems through increased resources, increased engagement of individuals and more productive individuals, in response to Ord’s discussion of Shaping Humanity’s Longterm Trajectory in chapter 13, I’d also posit that increasing compassion may useful in shaping humanity’s longterm trajectory as both a ‘speed up’ and an ‘enhancement’ through it’s potential to increase wellbeing.

Increasing compassion in the context of uncertainty about predictions for the far future and whether present action is worthwhile

Deliberation of long-term problems

On consideration of how we should address long-term problems, in Chapter 12 Kitcher argues for a system of well-informed deliberation among all affected parties aiming at a mutually agreeable solution, with parties seeking to represent to interests of future generations (to the extent that this is possible). Considering this approach, compassion is likely to be a pre-requisite for both good-faith efforts to reach mutually agreeable solutions, and having any hope of meaningfully representing the views of future generations. Without sufficient levels of compassion, parties cannot empathise with and act on this empathy for either future generations or the others around the table. Moreover, keeping the chain of deliberation going, requires ongoing compassion for future generations. 

Flexibility of action as an advantage

At various points in the Book, authors recommend epistemic humility about whether or not we can reliably make very long-term forecasts, with concerns including humans’ lack of predictive power. Given the limited evidence available, this seems a sensible rule of thumb - especially when it comes to specific scenarios or impacts. (Though I do wonder whether – albeit with lower confidence than in the present context – we could reasonably say that in most imaginable far future scenarios, a high-compassion society would likely result in better outcomes than a low-compassion society?) In the context of needing epistemic humility about forecasting the future, various contributors to chapters in the Book express uncertainty or a negative view on whether it is worthwhile to take present action to mitigate risks in the far future. Given the high levels of uncertainty, it seems beneficial for any present action for the distant future that we do pursue to be flexible in nature and something that could foreseeably have a positive impact in a broad range of scenarios. With this in mind, and the knowledge that humans are effective as passing knowledge and skills down to future generations, a skill like compassion rather which could be flexibly applied to many different scenarios, may be more useful than less adaptable interventions.

Population ethics 

As noted above, the flexible skill of compassion may be especially beneficial in delivering more ethical or equitable decision-making. Throughout the Book, ‘Population Ethics’ pops up, and, recognising this is a notoriously complex area, with risks of Repugnant Conclusions, decision-making in this area is important. In the future it may be the domain of governments as much as ethicists and it seems likely that a high-compassion society is more likely than a low-compassion society to favour population ethics approaches that longtermists would find palatable. 

Other thoughts on compassion through the lens of chapters in the Book 

Alignment of longtermist and non-longtermist aims

A central debate across multiple chapters in the book is whether or not any action for the long-term is the moral thing to do, with dissenting opinions criticising longtermism for ignoring present suffering or requiring sacrifice from present agents for a benefit they will not see. 

A real benefit of increasing compassion as an action for the distant future is that it is flexible enough in it’s utility that I don’t believe support for it requires a strong conviction one way or the other in this debate as increasing compassion now could improve outcomes in both the present and the distant future. It doesn’t require choosing between present and future, nor significant sacrifice of present individuals for a payoff that they will never see. 

Similarly, considering Curran’s discussion of the implications of anti-aggregationism for prioritisation between different ways of trying to affect the very long run, it strikes me that the reason to have the discussion in the first place (beyond the academic interest of the arguments) is a practical consideration of where it’s best to spend limited resources, from a moral standpoint, what are we obligated to prioritise? In considering this, the benefit of an action that can combine high likelihood of short-term impact in addition to potential benefit in the future (albeit with a much lower level of confidence) sets apart action to increase compassion from many of the other potential actions discussed in the Book. 

In such scenarios where there is a debate over prioritising actions to benefit one cause or another, finding an action that provides benefit to both can often involve a compromise whereby in order to provide some addressing of both scenarios, the action taken is adapted to be less well suited to one or both scenarios. In this regard, I again think increasing compassion has an advantage in that it is not obvious that the actions you’d take to increase compassion in the short-term would be different to or detract from efforts to increase compassion in the long-term and vice-versa (though much more exploration is required to give any real epistemic weight to this intuitive statement).

Seeking survival of humanity

Some chapters of the Book express a view that the extinction of humans would or could be a catastrophe – for reasons including moral voids and loss of perfectionist goods. Thinking back to the view of Darwin and more recent evolutionary biologists that compassion was integral to the survival and thriving of the human race, if we seek to prevent human extinction and increase the chances of (far) future generations thriving, it makes sense that we should consider actions that could increase or promote compassion. 

Should increasing compassion be a priority?

So then, let’s say we choose to buy in to the arguments above that increasing compassion would be beneficial for distant future outcomes, making future flourishing more likely, or that we decide it’s an approach we want to prioritise because even if – with the huge uncertainty over our ability to influence long-term outcomes – our efforts don’t have the specific impact we aim for in the long-term, they would still have significant positive outcomes in the near-term and are likely to increase available resource for and interest in efforts to improve the long-term. Is it actually possible to increase human compassion and do we have any reason to believe that investment in this would provide good value or could result in significant progress? 

Neglectedness

At present, increasing compassion is not a stated priority of key EA organisations. On review of cause prioritisation statements and research agendas, references to increasing compassion are conspicuously absent. Outside of EA, there are various organisations such as Charter for Compassion, Coalition for Compassionate Schools, and the Global Compassion Coalition focused on spreading compassion in the present, but these organisations are not well-known outside of the positive psychology field; receive little funding; and are largely reliant on volunteers. There are also some networks focused on positive psychology (education) more broadly such as the Greater Good Science Center which provide resources related to teaching compassion, but also have relatively limited reach, public awareness and funding. 

Despite a growing evidence base demonstrating the effectiveness of compassion interventions for increasing compassion, and significant personal benefits to doing so (including improved happiness, resilience, mental health and even reduced risk of heart disease), compassion interventions and research have not yet been recognised as a priority for governments, national health systems, institutional health funders (e.g. Wellcome Trust, Gates Foundation) or global authorities on health (e.g. WHO). Consequently, funding for compassion interventions is limited and typically comes from relatively small philanthropic organisations or funds for social sciences research such as the Economic and Social Research Council.  

Tractability 

Increasing global compassion is a more tractable problem than you might intuitively think. Culture change is difficult, yes. Very difficult in fact – especially if you’re seeking to influence global decision-makers. However, this may be a case where a bottom up approach (starting with individuals) is actually a better approach – especially as adults seeking happier, more successful lives, and schools and parents aiming to increase child wellbeing and performance may actively seek out programmes that increase compassion. 

Noting that the science of compassion, and positive psychology more widely, is a relatively young field, and (as noted above) there are relatively few funding sources, there are many existing knowledge gaps and there is significant potential for furthering our understanding significantly through provision of additional research funding. 

And existing research evidence is promising. Not only do neuroscience studies suggest that the brain structures involved in positive emotions like compassion are highly plastic, but compassion training programmes have demonstrated that psychologists can reliably train people in compassion, with multiple benefits to the individual. These compassion training programmes are ripe for scaling up and translating for different contexts (requiring consideration of both language and cultural context). Though it would require effort at the policy level as well as investment in research, teacher training and translation, there’s no obvious reason that teaching of compassion (ideally paired with promotion of compassionate culture in education settings) couldn’t be rolled out on a very large scale; perhaps in a similar way to how sex education in schools has gone from non-existent to being highly prevalent, and even part of the national curriculum in some jurisdictions. Though any interventions involving children and young people can be controversial, researchers studying such compassion interventions have demonstrated various positive outcomes including improved classroom behaviour, attitudes to learning, higher academic outcomes, and reduction of suicidal ideation. However, the impacts of limited funding is reflected in the relatively small scale of most of these studies and in the lack of longitudinal studies.

Risks and criticisms of increasing compassion

Compassion as a weakness - A persistent criticism of promoting compassion is the risk of compassion or self-compassion creating weakness. What if some countries behave compassionately and others do not (or even other worlds/intelligent lifeforms over a long timeline) – won’t the compassionate populations be taken advantage of and possibly be dominated by the less compassionate ones? Arguments like this are based on an understandable fear but they imply that logic and reasoning are impaired by compassion, when there’s no evidence supporting this fear – and studies show that humans adapt their compassion behaviours for individuals that they know might cheat or try to take advantage. Having the capacity to show care for someone doesn’t make you blind to their actions or motivations. A visible real world example of this in extremis is the Dalai Lama; he is often described as one of the most compassionate people in the world, but his vast propensity for compassion has not resulted in him yielding to the will of China. Similarly, studies by self-compassion authority, Kristen Neff, have demonstrated that self-compassion doesn’t make people too easy on themselves or less productive.

Culture wars - A bigger risk, to the success of implementation of interventions to increase compassion may be culture wars. In the same way that activities promoting equality, diversity and inclusion (EDI) have been targeted, it is possible compassion could be targeted and you could argue that the ‘America First’ mantra, and resulting US government interventions such as the dismantling of USAID are in direct contradiction to the principles of compassion. I have some hope that compassion – as a concept more established than EDI, and more explicitly integral to world religions including Christianity, is harder to demonise in the same way. And may also have greater resilience to this type of attack as there is more personal incentive for individuals to increase compassion in themselves and their children due to the proven benefits for their happiness, resilience and academic success. Nevertheless, many of the political parties around the world promoting ‘anti-woke’ agendas and politics of division, will be aware that promoting compassion over division would reduce their appeal and therefore it is highly plausible that movements such as MAGA in the USA would seek to smear or eradicate programmes to increase compassion.

Empathic collapse / compassion fatigue - Another criticism of programmes promoting compassion is the risk of individuals being overwhelmed by feelings of empathy for others’ suffering and this resulting in not only a retreat from compassion and resistance to engaging with suffering, but potentially a feeling of hopelessness. This has been given various names, including ‘empathic collapse’, ‘compassion fatigue’ and ‘compassion collapse’. However, as noted in this discussion of compassion collapse “Compassion cultivation techniques have been shown to increase positive emotions and social support, reduce negative distress at human suffering, and reduce people’s fears of feeling compassion for others. Such training programs may prevent the collapse of compassion, by letting people overcome fears of fatigue and accept their own compassion.” This suggests that compassion collapse is not inevitable and can be actively avoided with the right training in compassion. 

Risks of iatrogenic harm and ineffectiveness - There have recently been various criticisms of classroom mental health and mindfulness interventions and concerns about the risk off iatrogenic harm have led to some to call for an end to all school interventions related to psychology and wellbeing. However, Prof Frances Maratos, an expert in the field of compassion education, points out in her recent article that such harms can be avoided so long as we (i) avoid conflating mental health treatment and social and emotional learning (ii) ensure only sufficiently robust and evidence-based interventions are rolled out in schools and (iii) restrict delivery of interventions addressing mental health issues (rather than developing social skills such as compassion) to delivery by qualified mental health professionals.

 In addition to the right content, compassion programmes require the right teaching and delivery. More research is required to determine the required level of training for teachers to optimise outcomes of compassion training programmes, and to answer wider pedagogical effectiveness questions such as whether compassion training is more effective when delivered as part of a wider programme of positive psychology/science of happiness training or does it work better as a standalone intervention? And if we view it compassion a skill that needs to be practiced, is the benefit sustained if schools (and later workplaces/wider society) do not provide a compassionate culture in which compassionate actions are the norm? 

Summary and conclusions

I view the absence of discussion on increasing compassion in the Book represents a missed opportunity and a wider weakness in the discussions of longtermism. Whilst I’ve made some attempt at discussing the relative merits of in the context of Essays in the Book – and hope to have illustrated the relevance of compassion to many of the discussions in the book, this essay is undeniably rough around the edges and I’d encourage experts in the field of longtermism and EA more generally to explore and elaborate on the points made in this essay. 

Key ideas that I’d encourage further research and discussion on include:  

• Some moral arguments for long-termism or preventing human extinction seem to be contingent on future humans being compassionate.
• Increasing compassion may be a longtermist good and / or a ‘speed up’ and ‘enhancement’ for shaping humanity’s longterm trajectory.
• Issues championed by longtermists would be easier solved in a world where the average human is more compassionate than now.
• Compassion is a highly flexible intervention that could have positive impacts in a broad range of (far) future circumstances.
• The need for high-quality decision making, including in relation to population ethics, to secure positive long-term outcomes may be an argument for increasing compassion.
• There would be more funding available for altruistic causes, including longtermist and other EA priority causes, in a more compassionate society.
• Increasing compassion is also likely to have a positive impact on near-term EA goals (for example, it could make achieving aims related to biosafety, nuclear disarmament, and AI governance easier).
• (With lower epistemic confidence) Achieving near-term and far future benefits from increased compassion may not require radically different approaches which could be a further flexibility-based reason to prioritise action to increase compassion.
• This – and the potential for positive impact for both near-term and long-term outcomes – means that increasing compassion as a priority could create some alignment of priorities for addressing current/near-term and long-term suffering – addressing one of the criticisms of longtermism – that actions for the distant future come at the expense of addressing current/near-term suffering.
• Despite the potential benefits above, there is not currently any prioritisation of actions to increase compassion either amongst longtermists or EAs more generally.
• As far as I can tell, (and maybe I’m wrong/looking in the wrong places?) there has been very little exploration of increasing compassion as a potential action for the distant future.
• It seems likely that increasing compassion is both highly neglected and relatively tractable.
• So long as interventions for increasing compassion are evidence-based – i.e. by utilising proven compassionate mind training programmes to avoid iatrogenic harm and empathic burnout, there seems to be limited downside risk to increasing compassion. Though additional consideration needs to be given to the risks culture wars pose to scaling up such interventions.

In closing, I want to recognise the hopefulness of the growing evidence base for the benefits of increasing compassion and effectiveness of compassion interventions, as well as the current limitations in the existing research, especially when it comes to large-scale or longitudinal studies. 

In my view there is strong enough reason in the existing evidence base to support investing in present actions to increase compassion as a response to concerns about the distant future – and the possibly unique potential that increasing compassion has as a present action for the distant future that could not only improve long-term outcomes, but could also (i) act as a multiplier for activities in existing EA priority cause areas; (ii) enable increased altruistic activity through increases in giving and volunteering; and (iii) providing alignment between desired actions for the far future and near-term, with highly likely benefits in the near-term, even if longtermist objectives are not realised. 

However, for those not as convinced as I am, I’d argue that anyone advocating for actions to promote (far) future flourishing or for the expansion of the moral circle, should also be advocating for field-building or at least some foundational work in this area to assess the tractability of actions to increase compassion and their potential impact on both the distant future and nearer-term EA goals.

^[1] Am I the most qualified person to write an essay on this topic? Absolutely not! I am an enthusiast - perhaps being generous an amateur - when it comes to both EA and positive psychology, and a time-constrained one at that. However, as no one else has written it, and the idea of increasing compassion feels to me both important and conspicuously absent from the present discourse, I’ve decided to have a go. Please be compassionate in your criticisms. I do appreciate the writing is clumsy, lacking references and depth at points, and not in the lexicon of a professional ethicist. I encourage an actual philosophy professor to write their own essay exploring these ideas more robustly.