I actually just linked to part 3 because I’m too lazy to collect them all into one post. Also, part 1 might not really be all that necessary for some EAs to read, especially if you’ve already read stuff by people like Brian Tomasik or Magnus Vinding. Anyway, I’m hoping some of what I’ve written is interesting/helpful :)

Oh yeah, warning that this discusses extreme suffering and I’ve been told it can be quite depressing.

Part 1

Why it’s bad

Life is suffering.

Many people know the quote. Everyone has suffered, to one degree or another. But few humans have a deep or broad understanding of it. I am not one of them, at least on the dimension of depth; for this, I am very thankful.

But I have suffered, and I can use that as a reference point of sorts. When I was around 15 years old, I developed an anxiety disorder. Ever since then, it’s come and gone in waves. Fortunately, medication and therapy, along with general self-development, have generally made these episodes more sparse and bearable. But still. For those who may not be able to understand, those episodes feel like being stalked alone through a dark alleyway or menacing forest, except for hours, days, or even months on end. The suffering bleeds into every corner and crevice of experience. At its worst, it feels like it has lasted forever, and that it will last forever. It feels unbearable, and sometimes, it is. Sometimes I am surprised by how much I suffer from it. Yet suffering of this sort is commonplace, and is only the tip of an incomprehensible, terrible iceberg.

A terrifying example is that of cluster headaches. Several people describe their experiences with cluster headaches. One account that really struck me was this one:

A few years ago I had a life — a husband, children, an interesting job and a lovely home. One day my life became a nightmare. I lost everything. I had no life anymore just pain and suffering all the time with just a few hours’ relief every day. I couldn’t eat, sleep, listen or play with my kids — they were too loud. I couldn’t go to the theatre — too loud. I couldn’t sleep, couldn’t work, I was suffering too much and no one could understand me. I had lost everything (Kalina, France)

Suffering, especially that of this and similar magnitudes, should be relieved when possible, full stop. We do not think twice about giving anesthetic to those who undergo surgery, and we do not (or at least should not) question the need for pain relief for many mothers who undergo childbirth. More than that, the relief of intense suffering is of paramount importance. If the patient begins to scream from suffering on the operating table, relieving the patient’s suffering becomes a top priority (and for the patient, it is certainly the top priority). Similarly, if a person or animal is in a state of intense suffering, our first response should be one of urgent compassion and not detached apathy.

In case you’re inclined to believe that these reports are exaggerated, or that it would be different for you, you can check your pride here. You aren’t special, and neither are those who’ve experienced intense suffering. The only difference is that of luck. In another world, you are the one who is in Hell.

Okay. Get ready.

In my view, suffering is immense in the three key dimensions: scale/importance (depth and breadth), tractability, and neglectedness.

I already said that everyone suffers, but I say it again to clarify the definition of everyone. Every human that has ever lived, each of the 100+ billion of them, has suffered. Many have experienced severe suffering of one kind or another. One episode that demonstrates the wantonness of suffering was (according to Wikipedia) recounted by Sima Guang:

… The Song deaths and injuries were innumerable. When Wei forces encountered Song young men, the forces quickly beheaded them or cut them in half. The infants were pierced through with spears, and the spears were then shaken so that the infants would scream as they were spun, for entertainment. The commanderies and counties that Wei forces went through were burned and slaughtered, and not even grass was left. When sparrows returned in the spring, they could not find houses to build nest on, so they had to do so in forests. Wei soldiers and horses also suffered casualties of more than half, and the Xianbei people were all complaining.

Any decent scholar of history will tell you that such episodes are not the exception, but the rule. Even if a small fraction of such recountings reflect reality, what the Mirror shows is a very grim picture indeed. And it gets worse. For of course, pain is not limited to that which is deliberately caused. Disease, starvation, childbirth, and accidents were (and in some ways still are) cruel instruments used to great effect by an arbitrary and uncaring world.

And then it gets worse still when one expands the scope beyond mere humanity and considers all beings who are capable of suffering. Though the boundaries of sentience are somewhat fuzzy, current science definitively identifies most animals as sentient (this declaration represents the general scientific consensus). This is extremely bad news.

This is what it’s like to be boiled alive. If you don’t want to read the article, the gist of it is that it’s really bad. If you’ve ever been burned by steam or hot water or some other fluid, a good intuition might be to imagine that happening all over your body and all within your body for an agonizingly long time. Now imagine you’re a chicken in a factory farm. You’re shackled by your feet and hung upside down before a blade slices through your neck. But you’re still alive and conscious, and you watch with increasing terror as you’re lowered into a vat of boiling-hot water. Now imagine that happening a million times every year.

In fact, over forty billion farmed land animals (cows, pigs, and chickens) and many more farmed/wild-caught marine animals (including fish and crustaceans) are slaughtered every year and, in the case of farmed animals (land and marine), are forced to live in atrocious and inhumane conditions until the point of slaughter. These are numbers too large for our puny brains to comprehend. For context, a rough conservative estimate I came up with is that, were the Holocaust to be carried out on farmed land animals rather than humans, it would need to be repeated over 2,000 times to kill the number of farmed land animals that are killed worldwide in a single year. Imagine a 10,000 year Holocaust condensed into the span of a single year, year after year, and you start to imagine the scale which I’m speaking of.

We’re not done. For we have not even begun to consider the welfare of wild animals. It’s easy to fall prey to the Myth of Bambi (or as I like to think of it, the Lion King Effect) and either ignore or rationalize the suffering of wild animals. Gary Paulsen, in his book Woodsong, writes:

Wolves do not kill “clean.” (If there can be such a thing.) It is a slow, ripping, terrible death for the prey and only those who have not seen it will argue for that silly business about the prey actually selecting itself. Two wolves held the doe by the nose, held her head down to the ice, and the other wolves took turns tearing at her rear end, pulling and jerking and tearing, until they were inside of her, pulling out parts of her and all this time she was still on her feet, still alive. She was still on her feet though they had the guts out of her now, pulled back on the ice, eating and pulling, and I wanted it to end, wanted it to be over for her. And she sank. She somehow did not die then and still does not die in my mind. She just sinks. Over and over I can see her sinking as they pull at her. When I could stand it no longer, when I was sick with it and hated all wolves for the horror of it, I yelled. “Leave her…”

One may argue that the wolves had no choice, that they needed to eat her or they would starve. One may argue that there is a certain harmony in the order of nature, that the deer herself pulled up plants by their roots and devoured them. One may argue that it is in the service of the balance of the ecosystem, which we do not understand and should not try to change. One may argue.

Yet if there’s one thing I want to get across, it’s this:

Suffering brooks no arguments

Imagine you are the doe. Imagine the sharp teeth tearing into your flesh, ripping skin from organs and sinew from bones. Imagine your suffering in that eternal moment, frightened and alone and in unbearable agony. What argument could I make to you that would justify that?

Now imagine this happening on a scale so large it dwarfs even that of factory farming. From the Great Apes to the tiny ants, and everywhere in between. Trillions of trillions of sentient beings subjected to the same horrible ends.

Brian Tomasik writes about the horror of suffering:

On one afternoon, while walking outside, I saw many winged ants on the sidewalk, some of which had been stepped on. I wondered whether I should try to move them out of the way of pedestrians, but there were too many. The sheer number of ants was such that I felt awful thinking about the pain of each of their deaths, multiplied by how many there were. Of course, this is a drop in the bucket relative to nature as a whole, but seeing it up close made the horror sink in

That is nature, and that is our world. Simply indifferent and full of extreme suffering.

The beauty and harmony of nature is no consolation to its sufferers. In fact, it is the worst perversion. That an exterior so beautiful and wondrous should conceal a bottomless Hell of suffering and indignity is a level of perversity only a demon could orchestrate.

I could go on and I could go further, considering speculations on the suffering of future people, of artificial intelligences, of alien civilizations, and of even wilder things. But I don’t need to. I hope.

The good news is that because there’s so much intense suffering, there’s so much opportunity to reduce it, both at individual and collective levels.

Even tweaking our daily habits can reduce intense suffering immensely. For example, joining the growing plant-based or plant-forward movement directly corresponds to a reduction in the number of animals tormented and slaughtered for food. Through this alone, you could spare thousands of animals immense suffering and painful deaths. Imagine the joy you might feel if you could save even one dog from a puppy mill. Now imagine getting to do that day after day after day. That’s the sort of impact you could have just by changing what you consume.

Donating to the right charities is also a great way to reduce intense suffering. It’s true that many charities are ineffective, but not for the reasons commonly assumed. Effective charities are effective not only because they are managed well but because they use pragmatic interventions and focus on effective cause areas. Choosing the right charity is important, and there are many methods you can use to ensure your donations are used effectively.

A person can also use their career to reduce intense suffering. Of course, this is what many people already do with their careers (e.g. many healthcare professionals, humanitarian aid workers, and activists). If you’re earlier on in your career, or you’re (open to) transitioning to another career, then your choice of career path will be important to determining the impact you can have. Even if you’re later on in your career, or you’re set on a specific discipline, you can still make decisions that will maximize your impact. To take just one example, the fight against extreme poverty can and has drawn people from many disciplines. Norman Borlaug, an agronomist, is credited for spearheading efforts to increase food production that saved an estimated billion people from starvation. Peter Singer, the patron saint of animal welfare activists, used his expertise in philosophy to argue for greater efforts to reduce extreme poverty. Rob Mather, founder of the Against Malaria Foundation, comes from a background in business. These are just a few of the most prominent and exceptional individuals who have done much to reduce this one specific slice of suffering. Even ordinary folks like us can use the skills and interests we have effectively, for example by employing them in the service of an effective organization.

On the collective level, things can seem to unfold more slowly from the perspective of the individual. However, within our lifetimes, we’ve seen massive victories for civil rights, LGBTQ+ rights, and even farmed animal protections (the biggest example being California’s Proposition 12, which was recently upheld by the Supreme Court). These were all massive efforts that were made possible by the hard work of coalitions of people. There is no guarantee of progress except that which we make for ourselves, but there is also nothing more powerful than a collective, well-directed effort continually striving across time.

Another reason there’s so much opportunity to reduce intense suffering is that many causes regarding immense suffering are highly neglected. The biggest example is in animal welfare: over 99% of all animals in the US are farmed animals, but they receive only 0.8% of all funding for non-human animals in the US. The situation for wild animals is even more dismal, especially when one excludes funding that is not in their interests but rather in the interests of “nature” or conservation (which are often neutral to or even cause suffering in wild animals; e.g. reintroducing predators). Effort or resources expended for a more neglected cause area will be more beneficial, all else equal, than the same amount of effort or resources expended for a less neglected cause area. To put this in perspective, even if the number of farmed animals was the same as the number of shelter animals and farmed animal suffering was equal to that of shelter animals, it would still be more effective to invest one’s effort and resources into helping farmed animals because their cause is more neglected.

The bottom line is that many of us are capable to one degree or another of reducing suffering. Indeed, many of us already do so in our day-to-day lives. We might help an old lady cross a street, give money to a homeless person, or volunteer at the children’s hospital. These acts that help to alleviate the suffering of others are some of the most meaningful, both to the helper and the one helped. This is no different from that. It’s just that we now realize how much greater the scope of suffering was than we had previously thought, and we should adjust how we think and act accordingly. If you’ve made it through this essay, thank you; I imagine it might not have been easy. But you might have motivated yourself with the thought that, hard as it was for you, it was infinitely harder for the individuals who underwent the torments I described. For them, the least we can do is to learn and try not to repeat the terrors of the past.

Part 2

What makes it a priority

Imagine that you are creating a fabric of human destiny with the object of making men happy in the end, giving them peace and rest at last, but that it was essential and inevitable to torture to death only one tiny creature—that baby beating its breast with its fist, for instance—and to found that edifice on its unavenged tears, would you consent to be the architect on those conditions? Tell me, and tell the truth. (The Brothers Karamazov, Chapter 4)

Truly immerse yourself in this situation. Imagine as if it were real that you were there, making the choice between two possible worlds. In one, you have a world full of billions of blissful and joyful beings, but to achieve it, you must use the worst forms of torture on an infant. Burning, beating, stabbing, flaying, and so forth. The infant will survive for a long while with pain undiminished; as long as any other person. In another world, you have nothing. Which world would you choose?

Bad Is Stronger Than Good

The very fact that we struggle with this question at all should make us seriously question some underpinnings of our worldviews. To drive the point home, consider the inverted question: you have a choice between two worlds; in one, billions of beings endure the worst suffering while an infant enjoys the greatest bliss and joy, and in the other, there is nothing. Which world would you choose? It seems far more obvious in this case that the better choice would be nothing. Yet if pleasure and pain were equally important to us, we should expect the same amount of hesitancy to arise in answering both questions.

A number of scientific results seem to converge with the intuition that Bad Is Stronger Than Good. Individuals tend to react more strongly to bad events than good ones (for example, it is relatively routine for someone to experience intensely painful episodes even years after a traumatic event, but much less common for someone to experience similarly intensely pleasurable episodes years after a transcendent event). Relationships tend to be more affected by bad experiences, to the point where that it was estimated that stable relationships need at least a 5:1 ratio of good to bad experiences to remain stable. We have more words for bad emotions than good ones, and more strategies for avoiding bad states than for seeking good ones. We react more and are more averse to (perceived) losses than they are to (perceived) gains. Fear-inducing events can make practically unremovable changes to our brains, which remain even after a behavioral responses have been extinguished through unlearning techniques (the analogy would be cutting only the visible stem of a weed plant; its roots remain intact and ready to regrow another plant when an opportunity emerges). When a pair of “objectively” equally good and equally bad things happen (e.g. someone gives you $100 vs. someone steals $100 from you), it seems that any aggregation of the two produces not net neutrality but net badness. This would seem to apply even in the case that marginal utility was not significantly diminishing; for example, in the case that you were a millionaire

One could argue that—like loss aversion—the tendency to be disproportionately averse to bad things is irrational, and that we should even decrease the consideration we give to suffering based on this information (since suffering is “not as bad” as it feels). I hope there are few who would sincerely believe such an argument, but I will address it just in case. Of course, we should seek to remind ourselves where appropriate and possible that our feelings may not be adaptive. But anyone who has suffered to a significant degree knows that suffering—especially extreme suffering—is not something that can just be argued away. Until technology that can rewire our biology at the deepest level is developed, there will be no solution to the Asymmetry.

There Is More Bad Than Good

This is something that may have come across in Part 1, but it’s probably still worth covering here. Fortunately, when just considering humans, aggregate subjective well-being seems to be around neutral (in the least developed countries) and positive (in more developed countries). This might unfortunately imply that our ancestors (at least back until the agricultural revolution) lived at best neutral and at worst net negative lives. Though of course it also may imply that we can continue to improve human well-being, which is what matters. So hooray for small victories.

Okay, we’re done with the feel-good part, because whether or not human well-being is net positive or not pretty much doesn’t matter. The obvious place to turn to next is the non-human animals in our custody, and it’s pretty obvious that their lives are net bad. If you don’t think so, lock yourself in a dark room too small to turn around in and live in your own piss and shit for three weeks and come back and tell me whether you think that life was worth living. Even in the ultra-conservative case where you assign moral weights by neuron count or brain mass, the number of farmed (land) animals is such that together they would almost certainly tip the scale of aggregate well-being to be net bad (for example, in aggregate farmed land animals have about ¹⁄₉ the total brain mass that humans do, which is still significant). But you probably shouldn’t assign moral weights by neuron count or brain mass. You wouldn’t think that your baby or your dog deserved less moral consideration because they had less neurons or smaller brains… I hope. You also (probably) think you have greater moral worth than a single elephant, even though their neuron count and brain mass is far greater than yours. Though it’s unclear exactly how much suffering animals feel relative to humans, it seems that their suffering must be broadly comparable to our own. Perhaps they suffer more because they cannot understand the context of the situation they are in, or because they do not have all the cognitive tools that humans have to soothe themselves (though of course they have some). If, for many species, individual animals’ capacities for suffering are comparable to that of individual humans’, then their total suffering dwarfs both our total suffering and total pleasure.

And then, of course, there are the wild animals, whose numbers make even those of farmed animals look small. If we hold the assumptions established in the previous paragraph (that capacity to feel pleasure and suffering among sentients is similar), then their suffering or pleasure will predominate. Though it’s true that there’s a lot of suffering out in the wild, does this mean that the suffering must outweigh the pleasure?

We don’t know as much as we’d like to about wild animals, so any answer is going to be speculative. Nevertheless, the answer seems to be yes. In wild animals, suffering outweighs pleasure, too. If both our models for pleasure and suffering are such that there are increasing costs/diminishing returns to greater pleasure/suffering, which seems to be a reasonable assumption, then for any species in which more members are unsuccessful (die from starvation, predation, etc. and/or live frustrated lives) than successful (reproduce), there will be greater suffering. Take the example of a species where, for every successful member, two members will be unsuccessful; this model predicts that unsuccessful members will expend a cost (say 1) in producing suffering (punishment for not being successful) while successful members will expend a cost 2 in producing pleasure (reward for being successful). We would assume that a greater likelihood of success/failure would result in greater resources expended to reward/punishment. But expending twice as many resources into either pleasure or suffering will not yield twice as much pleasure/suffering due to diminishing returns, so it follows that the suffering in a scenario with more offspring who fail rather than succeed is greater than the pleasure. This does not even rely on the assumption that bad is stronger than good; even in the case where pleasure and suffering functions are equal but opposite, the model still holds true. In a case where there were more successful members than unsuccessful ones, we might expect pleasure to predominate, but there are few, if any, examples where this is true (humans might be one). In contrast, we know of many species (e.g. fruit flies) that produce hundreds or even thousands of offspring per female; this is far above the replacement rate, and so many more offspring will fail than succeed.

To give some intuitions to support the above analysis, one might imagine that for every frog that is successful in mating, dozens of frogs will die before reproduction, often in painful ways (predation, starvation, disease, etc.). To take an example closer to home, for every wild dog that is successful, around five or six others will not be and will once again die, often in painful ways. And then there’s the possibility that even “successful” members will die in painful ways or live frustrated lives after reproducing, whereas there’s no posthumous pleasure for an “unsuccessful” member who has already died (unless you believe dogs go to heaven).

Badness Can’t Be Offset By Goodness, Especially One’s Badness By Another’s Goodness

This is probably one of the hidden intuitions that troubles people when they run into classical utilitarianism. An example of this is a problem that arises where atrocities like slavery and genocide become permissible if a sufficiently large majority derives sufficient pleasure from them and a sufficiently small minority derives sufficiently little suffering from them. A major component of these thought experiments that are focused on are principles like justice, rights, and dignity. These, however, may obscure another problem that these thought experiments expose, which is that it is not obvious that goodness can offset badness. Suffering, specifically extreme suffering, may just be categorically more important than pleasure.

Let’s go through a toy exercise. Say child A makes fun of child B. A derives pleasure from this, while B derives suffering. Even if A is rolling with laughter while B merely sags in their seat, we (or at least many of us, I think) would still say that it would be better had this situation not occurred. This becomes especially true to us if B’s level of suffering reaches a sufficient level; for example, if B is crying from what A said, then there’s practically no level of pleasure A could derive that would somehow make up for that. One might argue that this would imply that we would need to accept something like enforced political correctness, but this is not necessary because the suffering caused by censorship might plausibly outweigh the suffering caused by collateral damage from stupid expressions of free speech.

But badness might not be offsetable by goodness, not only for comparisons between self and others, but even within a single person. Imagine you are given a choice: you can either choose to have two hours of a neutral experience, or you can choose to have one hour of incredible torture and one hour of incredible bliss. No bodily harm will come to you in either case, and you will not know which hour will be served first. Which will you choose? In my view, you should choose the two hours of neutrality. A follow-up question to this would be how many hours of incredible bliss you would need to experience to make up for an hour of incredible torture. I am not sure what the average person’s answer to this would be (though I’d guess it would be at least five or six hours of bliss for one hour of torture), but I know my answer is that there is no number of hours of incredible pleasure I would take in exchange for one hour of incredible pain. We can then compare this to a condition where you choose either between one hour of incredible pain or two hours of incredible pain. Obviously we would all choose one hour of incredible pain. So the intuition that becomes clear here is that, even in oneself, pleasure might not compensate for pain. This is demonstrated in a real-life case where a man named Dax Cowart endured an extremely painful event where he suffered severe burns and was forced to live through extreme pain while being treated.

After his treatment, Dax managed to recover and live what he considered a happy life — he successfully sued the oil company responsible for the pipeline leak, which left him financially secure; he earned a law degree; got married; and reported to have had “some very, very good experiences” after the accident. Yet even from this position of an accomplished and self-reportedly happy life, he still wished that he had been killed rather than treated. In Dax’s own view, no happiness could ever compensate for what he went through (Vinding, pp 69).

Implications

Cases like these are troublesome for classical utilitarianism, which sees suffering and pleasure as interchangeable both within one person and between different people. By contrast, negative utilitarianism (which has three different “levels”), which focuses on suffering, has no issues with them. I think the results of these experiments, both philosophical and real, are sufficient to demonstrate that we should at least accept a weak, lexical-threshold form of negative utilitarianism. I think we should very seriously consider a strong lexical-threshold form of negative utilitarianism, and I think we should still consider a strong form of negative utilitarianism. In any case, reducing suffering (especially extreme suffering) should be our greatest priority, whether you identify as a negative utilitarian, classical utilitarian, or something else.

Part 3

How to reconcile with it

In the first two parts of this series, I talked a lot about Suffering (with a big S), taking a really abstract and general kind of view. And then there were two interludes where I talked about suffering on an individual, visceral level. Here, I want to look at it from an individual, cerebral level. This is for the long haul, for answering questions like “why is the world like this?” and “how can I cope with me and everyone I know being murderers and manslaughterers?”. I want to be clear: I can’t exactly be of much help in dealing with most forms of one’s own suffering, especially physical suffering. What I aim to do here is to help to mitigate any suffering caused by the realization of ideas such as the ones I’ve discussed so far. I don’t think there’s any silver bullet, just like there isn’t any silver bullet for suffering in general, but there are things that can help.

The Dialectic

“Dialectic (Greek: διαλεκτική, dialektikḗ; German: Dialektik), also known as the dialectical method, refers originally to dialogue between people holding different points of view about a subject but wishing to arrive at the truth through reasoned argumentation”

— Wikipedia

The Dialectic has been, without a doubt, one of the most useful concepts in my life. I think many of us use dialectics to some extent in an unconscious manner, and I think there’s so much we could gain from utilizing them to their fullest extent (this sentence is something of a dialectic). The dialectic technique I’d like to use, which I’ll refer to as synthesis dialecticism, is different both from classical logic and from naive dialecticism. Classical logic does not allow for middle positions and doesn’t provide a good framework for evolving dialogue, while naive dialecticism tends to blindly take the average of competing positions and/or to ignore seeming contradictions altogether. There are many cases in which another tool is needed to make sense of seeming contradictions that may be very difficult to resolve with classical logic and which may go unnoticed by naive dialecticism.

An example of how synthesis dialecticism can be helpful is in the case of veganism. Someone coming from a framework of classical logic might reason:

Consumption of any amount of animal products contributes to slaughter and exploitation of animals
Contributing to slaughter and exploitation makes a person a killer
Killers should be shamed, punished, and even attacked
All people who are not strictly vegan should be shamed, punished, and even attacked

This sort of reasoning, I think, can lead to outcomes that are not good (or not as good as they could be) from a larger perspective. Indeed, it’s possible that applying a framework of classical logic may result in counterproductive or even dangerous fanaticism because of its focus on elimination of seeming contradictions.

At the other extreme, a person who uses a naive dialectical framework may simply try to average the two viewpoints, or to accept both of them. So, for example, one might say that both a vegan diet and a non-vegan diet are equally okay. And while it may be pragmatic to say this to defuse tensions at the dinner table, it’s dangerously complacent to hold too sincerely to this attitude. The analogue of 200 years ago is a person who sincerely believes that slavery is an issue of state’s rights, not individual ones.

For me, the synthesis of these attitudes is to recognize that factory farming is perhaps the largest injustice we have ever directly perpetrated and that the most effective methods to combat it may be ones that we don’t feel would be most authentic or just. That many people do things that have bad consequences which we should not be complacent about and that they are not horrible people with the intention to do harm. We ought to make compromises where it is pragmatic to do so and continue to work towards animal liberation wherever we can.

Here are a few that I think have been really useful to me:

You’ve made mistakes and you are not a bad person

“I don’t like the terms “good person” or “bad person” because it is impossible to be entirely good to everyone.”

— Armin Arlert, Attack on Titan

I believe that very few people are capable of thinking they are bad people.

For if one placed any significant level of importance on being a good person, as many of us do, and if one defined the goodness or badness of one’s person by the acts they engaged in, then it would be impossible for one to appraise one’s actions as bad, because that would be in contradiction to one’s belief that one was good.

Even for those of us who can recognize mistakes, we tend to distance them from ourselves. What we did was actually something our past self did, and we criticize that person, who we are no longer. By shunting our sins onto our past selves, we remain clean. Think about it: when was the last time that, while you were doing something, you decided resolutely it was immoral and then continued to do it anyway? I don’t know if I can come up with anything recent. The closest most of us can come is that we have some moral struggle with ourselves, some doubt about what is permissible. Only later do we condemn ourselves as having done wrong.

Maybe something that contributes to this is how we internalize our actions. My belief is that, when we take some action of great consequence and define ourselves with it, it can be both difficult to deal with and counterproductive. Feelings of guilt and shame can be overwhelming, and shaming oneself and others can backfire (because of the possibility that one will distance oneself from the thing that provoked shame rather than changing behavior; see page 11). I think it’s more accurate to recognize that our actions were bad, or that they had bad consequences, or that there were better actions that could have been taken. Not only that, but at least in theory, this framing is more helpful because it leaves open the possibility of change. To define oneself as a bad person implies somewhat that there is an inability to change, that one will be tainted by this mark of badness no matter what one does. And that is counterproductive, because the point is to change what one does in the future, not to change an inaccessible past. To say that one has done bad things or acted in ways that had bad outcomes not only leaves open the possibility for future actions to be better, but implies that they will be.

Silver bullets don’t exist and you can do good without them

“There is no single, simple answer to the question of how we can best reduce suffering”

— Magnus Vinding

Raoul Wallenberg, a Swedish diplomat during WWII, used his position to save at least hundreds to thousands of Jews in Hungary, and may have been instrumental to a plot that saved the lives of an estimated 70,000 Hungarian Jews. Whatever the case may be, neither he nor anyone else saved all the Jews in Hungary, let alone all the Jews in Nazi-occupied Europe. And it would have been highly unrealistic to expect them to do so. Even for a single train of Jews about to be deported to Auschwitz, there was only so much he could do.

After Wallenberg had handed over the last of the passports he ordered all those who had one to leave the train and walk to the caravan of cars parked nearby, all marked in Swedish colours. I don’t remember exactly how many, but he saved dozens off that train, and the Germans and Arrow Cross were so dumbfounded they let him get away with it.

Undoubtedly many more dozens of Jews were not able to get a passport, and were deported to Auschwitz. And it must have been terrible if one were on that train, seeing the passports handed through but not being able to get one. Wallenberg himself may have felt regret that he could not have printed more passports, that he could not offer one to every person on that train, that he could not have saved them all.

I’ve personally gone through periods where I implicitly felt that there were silver bullets, and more than that, that I had one. I definitely held martyr and savior complexes, and I have a sneaking suspicion that I’m not the only one, as these phenomena already have names. Maybe these things never completely go away, though I’ve begun to recognize them for what they are. It was personally painful to recognize that there was really so little I, or anyone else for that matter, could do on the scale of trillions. At the same time, to (relatively) concretely rather than abstractly grasp the idea that I could probably work for the benefit of thousands, and possibly even for the benefit of millions, is something that’s been incredibly motivating.

Humans tend to engage in binary, black-and-white thinking. One is either good or bad. One is either effective or not. One can either save everyone or fail everyone. The reality is, of course, different. Though we cannot save everyone, we can save some. Many, even. And that’s important. Each one of those people who received a passport must have been overjoyed. And perhaps, even if a parent did not receive a passport, they were overjoyed that their child received one. Each one of these people who was saved was one person who was spared the torments of Auschwitz. So even though Wallenberg and others could not save everyone, they still did an enormous amount of good, and are rightly celebrated for their actions.

You have more than one goal, and that’s fine good

“...if you just ease off a little bit on the maximising, then you’ve got a strategy that’s much more robust”

— Toby Ord

Perhaps because Effective Altruists (and activists more broadly) have high goals and lofty standards, it can be disillusioning to learn that they (we) are also homo sapiens. There are internal and external pressures to live up to ideals and defer to common wisdom (or more cynically, dogma).

Many EAs (and activists more broadly) are so deeply convinced (and rightly so) of the importance of their work that they strive for the minimum of self-care, which is dangerous. This ends up compromising your ability to be productive and effective in the long-run (and this is long-run enough that even if you have pretty short TAI timelines, this should still be a concern!). Even just calculating from expected value, it would be better for one to err on the side of caution and accept some loss of potential productivity than to push to the absolute limit and risk long-lasting or permanent loss of potential productivity. How much buffer one should give oneself is something each individual must decide for themselves; I think this worksheet is helpful in providing a framework for individuals to improve effective self-caring. In this way, what appears at first to be merely irrational or wasteful—when inspected more carefully—might be an important protective factor (like how charities with low overhead are sometimes less effective).

I think this may apply even more broadly. Even if you are a moral realist and you subscribe to one moral system (and even if that system is utilitarianism), what the most moral thing is for you to do is more complicated than you are able to represent to yourself. If you’re a hard-core utilitarian, for example, you might believe that all people in Effective Altruism (EA) ought to work to solve the alignment problem, work as hard as possible, donate all their money to MIRI, and always aim for maximization in everything they do. Thinking through such a world, we can imagine a situation in which everyone in EA working on the alignment problem creates a technical solution, and the world is still destroyed by superintelligent AI because no one was creating policies and institutional changes to properly implement the technical solution. It would have been better had even a few EAs who were interested and competent in policy gone into the policy and institutional work rather than focusing on solving alignment. This is a pretty strawman example, and unfortunately I think models of similar naive simplicity are held by some people out there. I think more complex/less naive optimization models are still brittle in that they likely have unforeseen failure modes and these failure modes could be disastrous.

The Four Brahmaviharas

“In the same way, in this kind of time, you can let this equanimity to grow; compassion to grow. First, you will have a fraction of compassion, let it grow. A fraction of unconditional friendliness. Let it grow. A fraction of compassion. Maybe you are just having a small action. Just one action. Let it grow.”

— Triple Gem of the North

Perhaps some who read this will know of the concept of loving-kindness, or metta. And of course, people are aware of the concept of compassion, though they may not have the same understanding of compassion or karuna as how I would define it here. Likely less well-known are the concepts of sympathetic joy (mudita) and equanimity (upekkha). The path to developing a dialectical understanding of these four abodes has been immensely helpful to me, so I hope it can be helpful to you, too.

Loving Kindness (Metta)

“Darkness cannot drive out darkness: only light can do that. Hate cannot drive out hate: only love can do that.”

— Martin Luther King, Jr.

Loving kindness can be described as an opening of the heart and the radiating of a warm and light feeling, and can be defined as the wish to bring happiness and well-being to oneself and others. It’s most readily felt towards people close to us, especially those such as young children and animal companions. The point of practicing metta is to take that affection that we naturally feel to the ones close to us and let it blossom. We learn to love acquaintances, then strangers, and even villains. This may seem strange; how could one love Hitler, for example? And why would we want to love someone who has done so much harm and injustice?

The far enemy of love is hatred or ill-will. Hatred is an integral part of human nature and of our moralities (indeed, people regularly rank “villains” such as Hitler as being outside their moral circles, as far as or further out than plants). It is natural to hate those who have done evil or those who are callous or sadistic. In some cases, it may even be adaptive (in a limited sense) to do so; for example, the honor culture of the American South likely arose due to a need to protect one’s herds (which could be easily stolen) and to protect oneself (as law enforcement was weak). At the same time, the risks and harms that can come from hateful acts are enormous. In a culture of honor, even single slights can quickly escalate into massive feuds, such as the famous Hatfield-McCoy feud; in a world of billions of people and weapons of mass destruction, such feuding is simply not an option. As John Green once said, an eye for an eye leaves the whole world monocular.

The near enemy of love is conditional love. True love is not conditional on traits, actions, or familial relations. Hitler did terrible things, and it would’ve been better had he drowned in a river when he was a kid (assuming, of course, no one worse took his place). We can readily acknowledge that many of his desires, intentions, and actions were incredibly wrong. At the same time, we can recognize the once-living being behind the name, and wish he had had the upbringing and environment (and even the innate dispositions) that would’ve allowed him to self-actualize in a benevolent and peaceful way.

We can extend this notion of unconditional love to ourselves and our relations. Even if one is not a utilitarian, it is hard to deny the argument that we fall short of the moral standards we set for ourselves. I think the ideal response is not to deny this or rationalize it away, nor is it to modify our ethical standards to fit the status quo; rather, it is to hold the fact that we cause suffering and death in our minds while also holding that by virtue of existing we are worthy of love, just as any other being is.

Though it’s not listed as a near enemy, I think another near enemy of love is indulgence, by which I mean fulfilling another’s superficial desires. Loving Hitler would not mean allowing him to kill all the Jews in the world. Instead, we might wish that he might be freed from the burdens of hatred and narrow-mindedness, and that he might become a better person. Similarly, loving ourselves and our relations does not equate to indulgence; we want to continue to challenge ourselves to do better; to become the people we truly are.

Practicing loving-kindness is not always easy, and it may be that you are not yet at a place where you can love Hitler. I don’t think I am. The principle is to simply allow our love to grow by not smothering it with feelings of judgment or ill-will. By giving the proper conditions and focus to loving kindness, we can let it sprout from a tiny seed into a beautiful flower. The result is that we feel lighter in our hearts and we are more effective in our interactions with others.

Compassion (Karuna)

“No matter who the organism is or what he/she has done, his/her happiness and suffering still count just the same”

— Brian Tomasik

Compassion, from my perspective, is very similar to loving kindness. The main difference is that loving kindness can be defined as focusing on increasing happiness and well-being, while compassion can be defined as focusing on alleviating suffering. Compassion may be the trickiest brahmavihara to define, and the most consequential to understand.

Part of the difficulty of defining compassion is the word itself and the cultural context it is in. Compassion is from the Latin com + passion, which literally means “to suffer with”. The word that actually has this definition in modern English is empathy, and more specifically visceral/emotional empathy. The distinction between compassion and empathy is unclear but very important (important enough that there’s an entire book written about it).

So what is the definition of compassion, and how is it different from empathy? Fortunately, some people are doing scientific research on compassion, which requires a (relatively) rigorous definition. One definition is that compassion is composed of five factors:

Recognizing suffering
Understanding the universality of suffering
Feeling empathy for the sufferer (emotional resonance)
Tolerating uncomfortable feelings aroused in response to the suffering and remaining open to and accepting of the sufferer
Motivation to act/acting to alleviate suffering.

The important parts of this definition to me are recognizing suffering, understanding its universality, and being motivated to act to alleviate the suffering in whatever way is possible.

The far enemy of compassion is cruelty, and this is something we are very capable of. Even for someone like me, who is high in agreeableness and low in dark traits, there have been times where I have felt the urge to be cruel. When we see cruelty, and see its victims, the first thought that should come to our mind is compassion for the sufferer. So often in social movements, we become caught up in who we are fighting against; we feel hatred towards those who perpetrate cruelty. For whatever reason, we rarely stop to feel compassion for the actual ones who suffer. The degree to which we hold compassion for the ones we seek to help in our minds and hearts is precisely the degree to which we stay true to the mission of doing good.

The second thought should be this: there but for the Grace of God go I. For it is as Aleksandr Solzhenitsyn declared: “The line between good and evil runs not through states, nor between classes, nor between political parties either — but right through every human heart”. To some extent, all of us have caused great suffering, and all of us have suffered. We can be grateful that we have not caused or received as much suffering as many others have.

The near enemy of compassion is pity. Pity is the recognition of suffering without the connection provided by common being/universality of suffering. This results in a feeling of “standing above” or condescension which makes pity irritating to the one pitied.

Though it’s not often defined as a near enemy of compassion, I think an even more important one is despair which arises from empathy. Empathy has its part to play in compassion, for without an understanding of the suffering that is occurring, and without an understanding of the badness of suffering, compassionate action is not possible. At the same time, empathy alone is insufficient because while there is an understanding of the suffering, there is no action to alleviate it. I think, especially for EAs, confusion of compassion and empathy is the larger problem and can lead to more dangerous outcomes.

Emotions might be usefully thought of as measurement devices that detect certain stimuli across a range of situations. (Visceral) empathy might then be thought of as one’s “care-o-meter”; an indicator of how much one ought to care in a certain situation. In many ordinary situations in life, the care-o-meter is useful and functions well enough. If we see a little duckling struggling to move because their feathers are saturated with crude oil, our care-o-meters (hopefully for most of us, at least) register that and in a way we suffer with them (our mirror neurons may literally make it so we actually feel some of what we conceive their suffering to be).

The problem arises when we try to use our care-o-meter to gauge situations it was not built for. Increasing the number of individuals we’re trying to care about from one to several paradoxically decreases our care rather than increasing it. And when we reach scales of thousands, millions, or billions of individuals, our care-o-meters cease to be able to reliably tell the difference (see this example, section 5.2 or page 66). It’s like trying to measure the circumference of the Earth with a standard ruler; impossible, because it’s the wrong tool. The implication is that we cannot rely on empathy alone to tell us what we ought to do on a large scale, and that visceral empathy may even lead us in the wrong direction (i.e. by driving us to focus on smaller and more catchy causes, like preserving giant pandas, at the expense of larger and less glamorous causes, like distributing anti-malaria nets). I claim that this holds true for any philosophy of ethics, potentially with the exception of ethics of care.

In addition to its unreliability on large scales, empathy can be dangerous to the empath. I know this from personal experience. After reading an article by Brian Tomasik one night, I became viscerally aware of the massive amount of suffering that was happening all around me. I felt more than I’d ever felt before the suffering of people and animals and wild animals around the world, from the battery-caged chicken many miles away to the worker ant right outside my window. It felt like I was a vessel for molten fire, and I prayed to God to make all the suffering go away. It was the most painful experience of my life. And I knew that even then, what I felt was a hopelessly small fraction of the suffering my mind was trying to represent. I’m sure many others in helping professions (such as activists and healthcare workers) have similar experiences. And logically, it can be no other way. Imagine someone who is probably already high in empathy being exposed (to one degree or another) to thousands, millions, or billions of beings in massive suffering. To proportionally feel even a “normal” amount of empathy for each individual would be unbearable. Again, our care-o-meters just aren’t built for that. Trying to empathize with the suffering of the world is like trying to use one’s body to put out a volcano. All you’ll succeed in doing is destroying yourself.

I’m glad that Effective Altruism has shined a spotlight on effectiveness, and that utilitarianism has motivated people to increase their impartial impact. At the same time, it’s important to recognize that tying your care-o-meter to proportional radical empathy is literally impossible, and that trying to achieve this will leave you miserable and incapacitated while also impairing your decision-making. The mistake we make when we learn about effectiveness is to try to tie what we rationally believe to be true to our care-o-meters. Using them to get ourselves to care at all might be a useful first step, but it’s important that we don’t let ourselves become beholden to them.

So then, what’s the solution? Well, you can probably guess that I think it’s compassion as I’ve defined it here. One thing that I haven’t mentioned yet is that compassion is the lightly joyful wish to alleviate suffering. This may seem strange and even sacrilegious. At the same time, it’s the recognition and intention that count. Consider this: what might one feel when an infant is crying in distress? The empathetic response would be to get on the ground and start bawling with them, or perhaps more realistically, to feel anger in response to their anger. We know this is not helpful. A more helpful response is the heartfelt and joyful wish to soothe the suffering of the infant, to smile compassionately and say “shhhhh, it’s alright, I’m here now”. This light and beautiful feeling is compassion, and it is my belief that compassion can guide and protect us through the inferno of suffering.

Sympathetic Joy (Mudita)

“To help his friends, Frederick shared what he collected. Frederick told the field mice about the yellow corn, the green leaves, and the blue and purple flowers. Frederick warmed and cheered his friends with his words”

— Frederick, by Leo Lionni

I love children and non-human animal companions.

And yes, I know I’m falling right into the idealization of them as angels and cherubs. I know they’re capable of uncaring selfishness and vicious cruelty as well. At the same time, there’s something about them that is irreplaceable; a glimpse of gold in a world of green. Because a child or animal companion whose material, relational, and spiritual needs are taken care of has a joy that is unmatched in this world. It’s the joy of bright eyes and big smiles, of feet pattering on the sidewalk, of a new discovery in every moment. It’s the joy of innocence, both in the spiritual and in the moral sense.

I always smile when I see children, because I see this joy. Even when they must be quiet and well-behaved, I can see it hiding just below the surface, waiting for a chance to come out. And I feel joy for their joy. I feel sympathetic joy.

I also feel great joy when I see farm animals cared for in sanctuaries. Knowing that they will never be hurt (again), and knowing that they’ll spend their lives happy and loved, and knowing that THIS IS WHAT I’M FIGHTING FOR is what makes the fight worth it.

It is important to know that there is suffering, and that this may even predominate in the world. It is also vitally important to keep in mind that there is still joy and good in the world. This joy is like a candle that illuminates the darkness, for this joy is the joy we aim for in all sentient beings. We need causes we can fight for. We need something to remind us that hope—in the sense of not giving up in trying to do better—is always rational, no matter how dire the circumstances.

The far enemy of joy is envy. It’s very easy to feel envious; I know I’ve felt envious of people ranging from my friends to complete strangers. It’s also very difficult to feel envious, both because of the unpleasantness of the feeling and because of the judgment we bring upon ourselves for feeling envy. We may want to feel happy at someone else’s joy, and we may not be able to bring ourselves to that in that moment, and that’s okay. I think the best response to envy is self-compassion; once one validates and attends to one’s own suffering, one has the capability to turn outward and naturally feel joy for joy.

The near enemy of joy is intoxication or greed; in essence, it is craving. I believe this to be a subtler version of suffering, especially when considering the craving alone (and not the fulfillment of what is craved, though that could be argued to simply be a temporary release from the craving). Where joy is relaxed and contented, craving tends to be tight, tense, and seeking. This also gives an answer to the question of why one ought not to feel joy towards sadistic joy, for this is not joy but craving; compassion should be felt instead.

Personally, I also feel joy towards my own joy, in particular towards the joy I had in my younger years. I was very fortunate to have had a life that, up until the beginning of middle school, was relatively pleasant and which had many good experiences. Of course, it wasn’t all good, and I have many happy memories from that time. I’m happy I had that time of innocence, and I’m happy others can have it, too.

Equanimity (Upekkha)

“The Oak stood proudly and fought against the storm, while the yielding Reeds bowed low. The wind redoubled in fury, and all at once the great tree fell…”

— The Oak and the Reed

Equanimity is a sense of relaxed acceptance of what is. It is the ceasing of all resistance, as one might see in some in the moments before a death they know is coming.

The far enemy of equanimity is exertion and rigidity. Life appears to require exertion; when we are faced with threats, we become tense. We feel we must adhere to strict rules and schedules. When we are faced with a problem, our first instinct is to push back directly against it. We glorify this straining, this stretching of the self to the limit. Both the need for and the efficacy of this exertion are questionable. One might have had the experience of choking in a sport, where either overthinking or stress/tension resulted in poor performance in crucial moments. The grasping for control leaves us powerless, and the quest for strength leaves us fragile. Only when we surrender control can we perform at our peak. In general, forcing is going to be less effective because it’s pitting two opposing forces against one another, destroying or weakening both of them. You can strike a rock, and it will break, but strike water with the same force, and soon it will be as if nothing had happened.

The near enemy of equanimity is indifference. Indifference is why it’s infuriating when someone tries to console you by saying that “this was part of God’s plan” or “everything happens for a reason”. My perspective is that indifference is itself a form of exertion, because it is the resistance against connection with self and others, and a denial of what is true. When someone you love is dying, and you’ve made peace with it, that doesn’t equate to being indifferent to their plight. It doesn’t mean that, if a doctor suddenly came in with a miracle cure, you would refuse. It simply means that, when all is said and done, you will bear what must be borne. Perhaps it is well-expressed in the Serenity Prayer: “God, grant me the serenity to accept the things I cannot change, the courage to change the things I can, and the wisdom to know the difference”.

To fully unpack the concept of equanimity, and such rich philosophical concepts as radical acceptance, Kierkegaard’s despair, Camus’s defiance, and Nietzsche’s amor fati, will require a lot more writing than I can fit here. So… sorry for the cliffhanger. Hopefully you can accept it ;)

You can rest assured that I’ll come back to this topic.

Conclusion

I know I’m not the most qualified to speak on these things, and I’ve only really seriously engaged with suffering-focused ideas for around a year. Nevertheless, I think it’s worth putting out there because I believe that it’s really important to maintain stability and to have the correct motivations. I also hope these reflections will be helpful to some people, if only because they’ll know someone else other than themselves is thinking about these things. Knowing about massive suffering, in addition to being immensely burdensome in itself, can also be incredibly isolating. For one, because many others either cannot or will not understand it. And for another, because it’s easier to react to suffering by isolating ourselves. And then there’s the personal motivation; I know that it can be really difficult and depressing to bear with anything approximating the reality of this world, so I wanted to try my best to give people the tools to bear with that rather than just slapping them with the truth and leaving them to deal with it on their own. I wish only the very best for you, and I hope together we can alleviate the sufferings of many many sentient beings.

My Thoughts On Suffering