The post Toby Ord on the perils of maximising the good that you do appeared first on 80,000 Hours.

Remarkably, Foote saw what this momentous discovery meant.

“An atmosphere of that gas would give our earth a high temperature,” she wrote in 1857.²

Though Foote could hardly have been aware at the time, the potential for global warming due to carbon dioxide would have massive implications for the generations that came after her.

If we ran history over again from that moment, we might hope that this key discovery about carbon’s role in the atmosphere would inform governments’ and industries’ choices in the coming century. They probably shouldn’t have avoided carbon emissions altogether, but they could have prioritised the development of alternatives to fossil fuels much sooner in the 20th century, and we might have prevented much of the destructive climate change that present people are already beginning to live through — which will affect future generations as well.

We believe it would’ve been much better if previous generations had acted on Foote’s discovery, especially by the 1970s, when climate models were beginning to reliably show the future course of warming global trends.³

If this seems right, it’s because of a commonsense idea: to the extent that we are able to, we have strong reasons to consider the interests and promote the welfare of future generations.

That was true in the 1850s, it was true in the 1970s, and it’s true now.

But despite the intuitive appeal of this moral idea, its implications have been underexplored. For instance, if we care about generations 100 years in the future, it’s not clear why we should stop there.

And when we consider how many future generations there might be, and how much better the future could go if we make good decisions in the present, our descendants’ chances to flourish take on great importance. In particular, we think this idea suggests that improving the prospects for all future generations is among the most morally important things we can do.

This article will lay out the argument for this view, which goes by the name longtermism.

We’ll say where we think the argument is strongest and weakest, respond to common objections, and say a bit about what we think this all means for what we should do.

We’d like to give special thanks to Ben Todd, who wrote a previous version of this essay, and Fin Moorhouse, who gave insightful comments on an early draft.

The case for longtermism

While most recognize that future generations matter morally to some degree, there are two other key premises in the case for longtermism that we believe are true and underappreciated. All together, the premises are:

1. We should care about how the lives of future individuals go.
2. The number of future individuals whose lives matter could be vast.
3. We have an opportunity to affect how the long-run future goes — whether there may be many flourishing individuals in the future, many suffering individuals in the future, or perhaps no one at all.⁴

In the rest of this article, we’ll explain and defend each of these premises. Because the stakes are so high, this argument suggests that improving the prospects for all future generations should be a top moral priority of our time. If we’re able to make an exceptionally big impact, positively influencing many lives with enduring consequences, it’s incumbent upon us to take this seriously.

This doesn’t mean it’s the only morally important thing — or that the interests of future generations matter to the total exclusion of the present generation. We disagree with both of those claims.

There’s also a good chance this argument is flawed in some way, so much of this article discusses objections to longtermism. While we don’t find them on the whole convincing, some of them do reduce our confidence in the argument in significant ways.

If we’re able to make an exceptionally big impact, positively influencing many lives with enduring consequences, it’s incumbent upon us to take this seriously.

However, we think it’s clear that our society generally neglects the interests of future generations. Philosopher Toby Ord, an advisor to 80,000 Hours, has argued that at least by some measures, the world spends more money on ice cream each year than it does on reducing the risks to future generations.⁵

Since, as we believe, the argument for longtermism is generally compelling, we should do a lot more compared to the status quo to make sure the future goes well rather than badly.

It’s also crucial to recognise that longtermism by itself doesn’t say anything about how best to help the future in practice, and this is a nascent area of research. Longtermism is often confused with the idea that we should do more long-term planning. But we think the primary upshot is that it makes it more important to urgently address extinction risks in the present — such as catastrophic pandemics, an AI disaster, nuclear war, or extreme climate change. We discuss the possible implications in the final section.

But first, why do we think the three premises above are true?

1. We should care about how the lives of future individuals go

Should we actually care about people who don’t exist yet?

The discussion of climate change in the introduction is meant to draw out the common intuition that we do have reason to care about future generations. But sometimes, especially when considering the implications of longtermism, people doubt that future generations matter at all.

Derek Parfit, an influential moral philosopher, offered a simple thought experiment to illustrate why it’s plausible that future people matter:

Suppose that I leave some broken glass in the undergrowth of a wood. A hundred years later this glass wounds a child. My act harms this child. If I had safely buried the glass, this child would have walked through the wood unharmed.

Does it make a moral difference that the child whom I harm does not now exist?⁶

We agree it would be wrong to dispose of broken glass in a way that is likely to harm someone. It’s still wrong if the harm is unlikely to occur until 5 or 10 years have passed — or in another century, to someone who isn’t born yet. And if someone else happens to be walking along the same path, they too would have good reason to pick up the glass and protect any child who might get harmed at any point in the future.

But Parfit also saw that thinking about these issues raised surprisingly tricky philosophical questions, some of which have yet to be answered satisfactorily. One central issue is called the ‘non-identity problem’, which we’ll discuss in the objections section below. However, these issues can get complex and technical, and not everyone will be interested in reading through the details.

Despite these puzzles, there are many cases similar to Parfit’s example of the broken glass in the woods in which it’s clearly right to care about the lives of future people. For instance, parents-to-be rightly make plans based around the interests of their future children even prior to conception. Governments are correct to plan for the coming generations not yet born. And if it is reasonably within our power to prevent a totalitarian regime from arising 100 years from now,⁷ or to avoid using up resources our descendants may depend on, then we ought to do so.

While longtermism may seem to some like abstract, obscure philosophy, it in fact would be much more bizarre and contrary to common sense to believe we shouldn’t care about people who don’t yet exist.

2. The number of future individuals whose lives matter could be vast.

Humans have been around for hundreds of thousands of years. It seems like we could persist in some form for at least a few hundred thousand more.

There is, though, serious risk that we’ll cause ourselves to go extinct — as we’ll discuss more below. But absent that, humans have proven that they are extremely inventive and resilient. We survive in a wide range of circumstances, due in part to our ability to use technology to adjust our bodies and our environments as needed.

How long can we reasonably expect the human species to survive?

That’s harder to say. More than 99 percent of Earth’s species have gone extinct over the planet’s lifetime,⁸ often within a few million years or less.⁹

It’s possible our own inventiveness could prove to be our downfall.

But if you look around, it seems clear humans aren’t the average Earth species. It’s not ‘speciesist’ — unfairly discriminatory on the basis of species membership — to say that humans have achieved remarkable feats for an animal: conquering many diseases through invention, spreading across the globe and even into orbit, expanding our life expectancy, and splitting the atom.

It’s possible our own inventiveness could prove to be our downfall. But if we avoid that fate, our intelligence may let us navigate the challenges that typically bring species to their ends.

For example, we may be able to detect and deflect comets and asteroids, which have been implicated in past mass extinction events.

If we can forestall extinction indefinitely, we may be able to thrive on Earth for as long as it’s habitable — which could be another 500 million years, perhaps more.

As of now, there are about 8 billion humans alive. In total, there have been around 100 billion humans who ever lived. If we survive to the end of Earth’s habitable period, all those who have existed so far will have been the first raindrops in a hurricane.

If we’re just asking about what seems possible for the future population of humanity, the numbers are breathtakingly large. Assuming for simplicity that there will be 8 billion people for each century of the next 500 million years,¹⁰ our total population would be on the order of forty quadrillion. We think this clearly demonstrates the importance of the long-run future.

And even that might not be the end. While it remains speculative, space settlement may point the way toward outliving our time on planet Earth.¹¹ And once we’re no longer planet-bound, the potential number of people worth caring about really starts getting big.

In What We Owe the Future, philosopher and 80,000 Hours co-founder Will MacAskill wrote:

…if humanity ultimately takes to the stars, the timescales become literally astronomical. The sun will keep burning for five billion years; the last conventional star formations will occur in over a trillion years; and, due to a small but steady stream of collisions between brown dwarfs, a few stars will still shine a million trillion years from now.

The real possibility that civilisation will last such a long time gives humanity an enormous life expectancy.

Some of this discussion may sound speculative and fantastical — which it is! But if you consider how fantastical our lives and world would seem to humans 100,000 years ago, you should expect that the far future could seem at least as alien to us now.

And it’s important not to get bogged down in the exact numbers. What matters is that there’s a reasonable possibility that the future is very long, and it could contain a much greater number of individuals.¹² So how it goes could matter enormously.

There’s another factor that expands the scope of our moral concern for the future even further. Should we care about individuals who aren’t even human?

It seems true to us that the lives of non-human animals in the present day matter morally — which is why factory farming, in which billions of farmed animals suffer every day, is such a moral disaster.¹³ The suffering and wellbeing of future non-human animals matters no less.

And if the far-future descendants of humanity evolve into a different species, we should probably care about their wellbeing as well. We think we should even potentially care about possible digital beings in the future, as long as they meet the criteria for moral patienthood — such as, for example, being able to feel pleasure and pain.

We’re highly uncertain about what kinds of beings will inhabit the future, but we think humanity and its descendants have the potential to play a huge role. And we want to have a wide scope of moral concern to encompass all those for whom life can go well or badly.¹⁴

When we think about the possible scale of the future ahead of us, we feel humbled. But we also believe these possibilities present a gigantic opportunity to have a positive impact for those of us who have appeared so early in this story.

The immense stakes involved strongly suggest that, if there’s something we can do to have a significant and predictably positive impact on the future, we have good reason to try.

Arnold Paul, CC BY-SA 2.5 (cropped)

3. We have an opportunity to affect how the long-run future goes

When Foote discovered the mechanism of climate change, she couldn’t have foreseen how the future demand for fossil fuels would trigger a consequential global rise in temperatures.

So even if we have good reason to care about how the future unfolds, and we acknowledge that the future could contain immense numbers of individuals whose lives matter morally, we might still wonder: can anyone actually do anything to improve the prospects of the coming generations?

It’d be better for the future if we avoid extinction, manage our resources carefully, foster institutions that promote cooperation rather than violent conflict, and responsibly develop powerful technology.

Many things we do affect the future in some way. If you have a child or contribute to compounding economic growth, the effects of these actions ripple out over time, and to some extent, change the course of history. But these effects are very hard to assess. The question is whether we can predictably have a positive impact over the long term.

We think we can. For example, we believe that it’d be better for the future if we avoid extinction, manage our resources carefully, foster institutions that promote cooperation rather than violent conflict, and responsibly develop powerful technology.

We’re never going to be totally sure our decisions are for the best — but often we have to make decisions under uncertainty, whether we’re thinking about the long-term future or not. And we think there are reasons to be optimistic about our ability to make a positive difference.

The following subsections discuss four primary approaches to improving the long-run future:

• Reducing extinction risk
• Positive trajectory changes
• Longtermist research
• Capacity building

Reducing extinction risk

One plausible tactic for improving the prospects of future generations is to increase the chance that they get to exist at all.

Of course, if there was a nuclear war or an asteroid that ended civilization, most people would agree that it was an unparalleled calamity.

Longtermism suggests, though, that the stakes involved could be even higher than they first seem. Sudden human extinction wouldn’t just end the lives of the billions currently alive — it would cut off the entire potential of our species. As the previous section discussed, this would represent an enormous loss.

And it seems plausible that at least some people can meaningfully reduce the risks of extinction. We can, for example, create safeguards to reduce the risk of accidental launches of nuclear weapons, which might trigger a cataclysmic escalatory cycle that brings on nuclear winter. And NASA has been testing technology to potentially deflect large near-Earth objects on dangerous trajectories.¹⁵ Our efforts to detect asteroids that could pose an extinction threat have arguably already proven extremely cost-effective.



So if it’s true that reducing the risk of extinction is possible, then people today can plausibly have a far-reaching impact on the long-run future. At 80,000 Hours, our current understanding is that the biggest risks of extinction we face come from advanced artificial intelligence, nuclear war, and engineered pandemics.¹⁶

And there are real things we can do to reduce these risks, such as:

• Developing broad-spectrum vaccines that protect against a wide range of pandemic pathogens
• Enacting policies that restrict dangerous practices in biomedical research
• Inventing more effective personal protective equipment
• Increasing our knowledge of the internal workings of AI systems, to better understand when and if they could pose a threat
• Technical innovations to ensure that AI systems behave how we want them to
• Increasing oversight of private development of AI technology
• Facilitating cooperation between powerful nations to reduce threats from nuclear war, AI, and pandemics.

We will never know with certainty how effective any given approach has been in reducing the risk of extinction, since you can’t run a randomised controlled trial with the end of the world. But the expected value of these interventions can still be quite high, even with significant uncertainty.¹⁷

One response to the importance of reducing extinction risk is to note that it’s only positive if the future is more likely to be good than bad on balance. That brings us onto the next way to help improve the prospects of future generations.

Positive trajectory changes

Preventing humanity’s extinction is perhaps the clearest way to have a long-term impact, but other possibilities may be available. If we’re able to take actions that influence whether our future is full of value or is comparatively bad, we would have the opportunity to make an extremely big difference from a longtermist perspective. We can call these trajectory changes.¹⁸

Climate change, for example, could potentially cause a devastating trajectory shift. Even if we believe it probably won’t lead to humanity’s extinction, extreme climate change could radically reshape civilisation for the worse, possibly curtailing our viable opportunities to thrive over the long term.

There might even be potential trajectories that could be even worse. For example, humanity might get stuck with a value system that undermines general wellbeing and may lead to vast amounts of unnecessary suffering.

How could this happen? One way this kind of value ‘lock-in’ could occur is if a totalitarian regime establishes itself as a world government and uses advanced technology to sustain its rule indefinitely.¹⁹ If such a thing is possible, it could snuff out opposition and re-orient society away from what we have most reason to value.

We might also end up stagnating morally such that, for instance, the horrors of poverty or mass factory farming are never mitigated and are indeed replicated on even larger scales.

It’s hard to say exactly what could be done now to reduce the risks of these terrible outcomes. We’re generally less confident in efforts to influence trajectory changes compared to preventing extinction. If such work is feasible, it would be extremely important.

Trying to strengthen liberal democracy and promote positive values, such as by advocating on behalf of farm animals, could be valuable to this end. But many questions remain open about what kinds of interventions would be most likely to have an enduring impact on these issues over the long run.

Grappling with these issues and ensuring we have the wisdom to handle them appropriately will take a lot of work, and starting this work now could be extremely valuable.

Cobija, CC BY-SA 3.0, via Wikimedia Commons

Longtermist research

This brings us to the third approach to longtermist work: further research.

Asking these types of questions in a systematic way is a relatively recent phenomenon. So we’re confident that we’re pretty seriously wrong about at least some parts of our understanding of these issues. There are probably several suggestions in this article that are completely wrong — the trouble is figuring out which.

So we believe much more research into whether the arguments for longtermism are sound, as well as potential avenues for having an impact on future generations, is called for. This is one reason why we include ‘global priorities research’ among the most pressing problems for people to work on.

Capacity building

The fourth category of longtermist approaches is capacity building — that is, investing in resources that may be valuable to put toward longtermist interventions down the line.

In practice, this can take a range of forms. At 80,000 Hours, we’ve played a part in building the effective altruism community, which is generally aimed at finding and understanding the world’s most pressing problems and how to solve them. Longtermism is in part an offshoot of effective altruism, and having this kind of community may be an important resource for addressing the kinds of challenges longtermism raises.

There are also more straightforward ways to build resources, such as investing funds now so they can grow over time, potentially to be spent at a more pivotal time when they’re most needed.

You can also invest in capacity building by supporting institutions, such as government agencies or international bodies, that have the mission of stewarding efforts to improve the prospects of the long-term future.

Summing up the arguments

To sum up: there’s a lot on the line.

The number and size of future generations could be vast. We have reason to care about them all.

Those who come after us will have to live with the choices we make now. If they look back, we hope they’ll think we did right by them.

But the course of the future is uncertain. Humanity’s choices now can shape how events unfold. Our choices today could lead to a prosperous future for our descendants, or the end of intelligent life on Earth — or perhaps the rise of an enduring, oppressive regime.

We feel we can’t just turn away from these possibilities. Because so few of humanity’s resources have been devoted to making the future go well, those of us who have the means should figure out whether and how we can improve the chances of the best outcomes and decrease the chances of the worst.

We can’t — and don’t want to — set our descendants down a predetermined path that we choose for them now; we want to do what we can to ensure they have the chance to make a better world for themselves.

Those who come after us will have to live with the choices we make now. If they look back, we hope they’ll think we did right by them.

A protostar is embedded within a cloud of material feeding its growth. Credit: NASA, ESA, CSA, STScI

Objections to longtermism

In what follows, we’ll discuss a series of common objections that people make to the argument for longtermism.

Some of them point to important philosophical considerations that are complex but that nonetheless seem to have solid responses. Others raise important reasons to doubt longtermism that we take seriously and that we think are worth investigating further. And some others are misunderstandings or misrepresentations of longtermism that we think should be corrected. (Note: though long, this list doesn’t cover all objections!)

This is a lot to think about. So what are our bottom lines on how we think we’re most likely to be wrong about longtermism?

Here are a few possibilities we think are worth taking seriously, even though they don’t totally undermine the case from our perspective:

• Morality may require a strong preference for the present: There might be strong moral reasons to give preference to existing people and individuals over future generations. This might be because something like a person-affecting view is true (described above) or maybe even because we should systematically discount the value of future beings.
  • We don’t think the arguments for such a strong preference are very compelling, but given the high levels of uncertainty in our moral beliefs, we can’t confidently rule it out.
• Reliably affecting the future may be infeasible. It’s possible that further research will ultimately conclude that the opportunities for impacting the far future are essentially non-existent or extremely limited. It’s hard to believe we could ever entirely close the question — researchers who come to this conclusion in the future could themselves be mistaken — but it might dramatically reduce our confidence that pursuing a longtermist agenda is worthwhile and thus leave the project as a pretty marginal endeavour.

• Reducing extinction risk may be intractable beyond a certain point. It’s possible that there’s a base level of extinction risk that humans will have to accept at some point and that we can’t reduce any further. And if, for instance, there were an irreducible risk of an extinction catastrophe at 10 percent every century, then the future, in expectation, would be much less significant than we think. This would dramatically reduce the pull of longtermism.

• A crucial consideration could change our assessment in ways we can’t predict. This falls into the general category of ‘unknown unknowns,’ which are always important to be on the watch for.

You could also read the following essays criticising longtermism that we have found interesting:

• A review of The Precipice written by Theron Pummer
• A blog post called “Against Longtermism” by Eric Schwitzgebel
• A post on the Effective Altruism Forum by Denise Melchin called “Why I am probably not a longtermist“

If I don’t agree with 80,000 Hours about longtermism, can I still benefit from your advice?

Yes!

We want to be candid about what we believe and what our priorities are, but we don’t think everyone needs to agree with us.

And we have lots of advice and tools that are broadly useful for people thinking about their careers, regardless of what they think about longtermism.

There are also many places where longtermist projects converge with other approaches to thinking about having a positive impact with your career. For example, working to prevent pandemics seems robustly good whether you prioritise near- or long-term benefits.

Though we focus as an organisation on issues that may affect all future generations, we would generally be really happy to also see more people working for the benefit of the global poor and farmed animals, two tractable causes that we think are unduly neglected in the near term. We also discuss these issues on our podcast and list jobs for them on our job board.

Credit: Yen Chao CC2.0

What are the best ways to help future generations right now?

While answering this question satisfactorily would require a sweeping research agenda in itself, we do have some general thoughts about what longtermism means for our practical decision making. And we’d be excited to see more attention paid to this question.

Some people may be motivated by these arguments to find opportunities to donate to longermist projects or cause areas. We believe Open Philanthropy — which is a major funder of 80,000 Hours — does important work in this area.

But our primary aim is to help people have impactful careers. Informed by longtermism, we have created a list of what we believe are the most pressing problems to work on in the world. These problems are important, neglected, and tractable.

As of this writing, the top eight problem areas are:

1. Risks from artificial intelligence
2. Catastrophic pandemics
3. Building effective altruism
4. Global priorities research
5. Nuclear war
6. Improving decision making (especially in important institutions)
7. Climate change
8. Great power conflict

We’ve already given few examples of concrete ways to tackle these issues above.

The above list is provisional, and it is likely to change as we learn more. We also list many other pressing problems that we believe are highly important from a longtermist point of view, as well as a few that would be high priorities if we rejected longtermism.

We hope more people will challenge our ideas and help us think more clearly about them. As we have argued, the stakes are incredibly high.

We have a related list of high-impact careers that we believe are appealing options for people who want to work to address these and related problems and to help the long-term future go well.

But we don’t have all the answers. Research in this area could reveal crucial considerations that might overturn longtermism or cast it in a very different light. There are likely pressing cause areas we haven’t thought of yet.

We hope more people will challenge our ideas and help us think more clearly about them. As we have argued, the stakes are incredibly high. So it’s paramount that, as much as is feasible, we get this right.

Learn more

• Toby Ord discussed these arguments in his book, The Precipice, and he discussed the ideas with us on our podcast.
• Will MacAskill also made the argument in his book, What We Owe the Future, and we interviewed him about it on our podcast.
• Benjamin Todd and Arden Koehler discussed varieties of longtermism in this podcast.
• Hilary Greaves presented the case for longtermism at Oxford University.
• In this podcast, Holden Karnofsky talked about the case that we’re living in the most important century.
• Article: The case for reducing existential risks
• Podcast: Carl Shulman on the common-sense case for existential risk work and its practical implications
• Podcast: Anders Sandberg on war in space, whether civilisations age, and the best things possible in our universe

Read next

This article is part of our advanced series. See the full series, or keep reading:

• 

  Is it ever OK to take a harmful job in order to do more good? An in-depth analysis

  Read more
• 

  Why the problem you work on is the biggest driver of your impact

  Read more
• 

  The case for reducing existential risks

  Read more

The post Longtermism: a call to protect future generations appeared first on 80,000 Hours.

Your career is really ethically important, but it’s not a single, discrete choice. To build a high-impact career you need to make thousands of smaller choices over many years — to take on this particular project, to apply for that internship, to give this person a positive reference, and so on.

How do you make all those little decisions?

If you want to have an impact, you hope to make the decisions that help you have a bigger impact rather than a smaller one. But you can’t go around explicitly estimating the consequences of all the different possible actions you could take — not only would that take too long, you’d probably get it wrong most of the time.

This is where the idea of virtues — lived moral traits like courage, honesty, and kindness — can really come in handy. Instead of calculating out the consequences of all your different possible actions, try asking yourself, “What’s the honest thing to do? What’s the kind thing to do?”

A few places I find ‘virtues thinking’ motivating and useful:

• When I am facing a difficult work situation, I sometimes ask myself, “What virtue is this an opportunity to practise?” For example, maybe now is a great opportunity to practise being honest — even when it’s difficult or embarrassing.
• Sometimes I get socially anxious. When that happens I often find it helpful to ask what virtues I can bring to a conversation — how would a kind, gracious, and curious person act?

(Some people think that acting in line with virtues is just what it is to live a moral life, but even if you don’t think that, they are still useful mental shortcuts.)

Honesty and kindness are very commonly thought of as virtues. But it’s also worth asking whether there are other, non-standard virtues you might want to cultivate to help you accomplish what you think is most important.

In my case, I think ‘scope sensitivity‘ — caring much more about something that affects many more individuals or affects them to a much greater degree — is a virtue. Everyone has to prioritise in their work, and I think scope sensitivity helps people prioritise in ways that make the world better.

For example, becoming a doctor lets you help a lot of people. But if you’re also well-suited to a career that could help prevent the next catastrophic pandemic, you might be able to help even more people than you would in other areas of medicine. If you’re scope sensitive, you may notice this opportunity, and working to prevent or mitigate pandemics might become very appealing. I think that’s an admirable quality.

The main questions virtues bring up are:

1. How do you decide which character traits are virtues?
2. What do you do when two virtues conflict? For example, should you tell someone the brutal truth (honesty) or spare their feelings (kindness) by obscuring it?

I think to answer the first question it makes sense to look at the long-term consequences of acting in a certain way. If you act honestly as a general rule throughout your life, is that likely to make things go better or worse? I think there’s a good case that it will go much better; people will trust you, because you’ll be trustworthy — and this can be a big help in many areas of life.

You can also look at role models — what kinds of lived moral traits do the people who seem to have had a big positive impact have? One answer is courage: for example, Benjamin Lay stood up over and over for the idea that slavery was wrong, even when he faced severe social sanctions for it. To do a lot of good you might have to face down tough situations with courage.

And you can put plausible-sounding virtues to the test by asking what would happen if you applied them consistently over time. For example, you might think that agreeableness is a virtue because it makes people feel comfortable and supported. But if you’re practically always agreeable, people might not trust you to tell them what you really think, and you might go along with others’ plans even when they are bad. So it seems like a mixed bag.

Of course you can also go overboard with honesty; most people probably agree it’s a bad idea to tell the truth if a murderer asks for your friend’s location. But what makes honesty a virtue in my book is that it’s much better to go through life erring on the side of being too honest than not honest enough — so it’s a good general guide for behaviour. I think that’s less true of agreeableness.

Deciding what to do when virtues are in conflict is tough. But here are a few strategies:

• Is there something about the situation that makes one of the virtues more important? If the context makes trust especially important, you might have extra reason to lean into honesty in this case even at the expense of other virtues.

• Is there a commonly thought of hard and fast ethical rule that says what you should do? Even if you don’t philosophically believe in absolute rules, they usually point to very important considerations.

• Look at the consequences: in cases where heuristics fail us, it can be worth it to think explicitly about the consequences of each action — especially if the consequences are dramatically better or worse depending on what you choose.

The web team at 80,000 Hours has a few ‘virtues for 2023’ to help us navigate unexpected situations:

• Openness: be curious, transparent, and open to changing our minds.
• Patience: go for what’s best in the long-run, and don’t get unduly distracted by what’s salient now.
• Boldness: don’t be afraid to prioritise what we think is most important and push for it.

Learn more:

• Scope-sensitive ethics
• Why it’s a bad idea to break the rules, even if it’s for a good cause
• What is social impact? A definition

The post Why you should think about virtues — even if you’re a consequentialist appeared first on 80,000 Hours.

By getting clearer about your definition, you can better target your efforts. So how should you define social impact?

Over two thousand years of philosophy have gone into that question. We’re going to try to sum up that thinking; introduce a practical, rough-and-ready definition of social impact; and explain why we think it’s a good definition to focus on.

This is a bit ambitious for one article, so to the philosophers in the audience, please forgive the enormous simplifications.

A simple definition of social impact

If you just want a quick answer, here’s the simple version of our definition (a more philosophically precise one — and an argument for it — follows below):

Your social impact is given by the number of people¹ whose lives you improve and how much you improve them, over the long term.

This shows that you can increase your impact in two ways: by helping more people over time, or by helping the same number of people to a greater extent (pictured below).

We say “over the long term” because you can help more people either by helping a greater number now, or taking actions with better long-term effects.

This definition is enough to help you figure out what to aim at in many situations — e.g. by roughly comparing the number of people affected by different issues. But sometimes you need a more precise definition.

A more rigorous definition of social impact

Here’s our working definition of “social impact”:

“Social impact” or “making a difference” is (tentatively) about promoting total expected wellbeing — considered impartially, over the long term.

We don’t think social impact is all that matters. Rather, we think people should aim to have a greater social impact within the constraints of not sacrificing other important values – in particular, while building good character, respecting rights and attending to other important personal values. We don’t endorse doing something that seems very wrong from a commonsense perspective in order to have a greater social impact.

In fact, we even think that paying attention to these other values is probably the best way to in fact have the most social impact anyway, even if that’s all you want to aim for.

In the rest of this article, we’ll expand on:

1. Why we think social impact is primarily about promoting what’s of value — i.e. making the world better
2. Why we think making the world a better place is in large part about promoting total expected wellbeing
3. How social impact fits with other values
4. How we can assess what makes a difference in the face of uncertainty

We believe that taking this definition seriously has some potentially radical implications about where people who want to do good should focus, which we also explore.

Two final notes before we go into more detail. First, our definition is tentative — there’s a good chance we’re wrong and it might change. Second, its purpose is practical — it aims to cover the most important aspects of doing good to help people make better real-life decisions, rather than capture everything that’s morally relevant.

Social impact is about making the world better

What does it mean to act ethically? Moral philosophers have debated this question for millennia, and have arrived at three main kinds of answers:

1. Being virtuous — e.g. being honest, kind, and just
2. Acting rightly — e.g. respecting the rights of others and not doing wrong
3. Making the world better — e.g. helping others

These correspond to virtue ethics, deontology, and consequentialism, respectively.

Whatever you think about which perspective is most fundamental, we think all three have something to offer in practice (we’ll expand on this in the rest of the article). So we could say that a simple, general recipe for a moral life would be to:

1. Cultivate good character
2. Respect constraints, such as the rights of others
3. And then do as much good as you can

In short, character, constraints and consequences.

When our readers talk about wanting to “make a difference,” we think most interested in the third of these perspectives — changing the world for the better.

We agree focusing more on doing good makes sense for most people. First, we don’t just want to avoid doing wrong, or live honest lives, but actually leave the world better than we found it. And more importantly, there’s so much we can do to help.

For instance, we’ve shown that by donating 10% of their income to highly effective charities, most college graduates can save the lives of over 40 people over their lifetimes with a relatively minor sacrifice.

From an ethical perspective, whether you save 40 lives or not will probably be one of the most significant questions you’ll ever face.

In our essay on your most important decision, we argued that some career paths open to you will do hundreds of times more to make the world a better place than others. So it seems really important to figure out what those paths are.

Even philosophers who emphasise moral rules and virtue agree that if you can make others better off, that’s a good thing to do. And they agree it’s even better to make more people better off than fewer. (More broadly, we think deontologists and utilitarians agree a lot more than people think.)

John Rawls was one of the most influential (non-consequentialist) philosophers of the 20th century, and he said:

All ethical doctrines worth our attention take consequences into account in judging rightness. One which did not would simply be irrational, crazy.²

Since there seem to be big opportunities to make people better off, and some seem to be better than others, we should focus on finding those.

This might sound obvious, but most discussion of ethical living is very focused on reducing harm rather than doing more good.

For instance, when it comes to fighting climate change, there’s a lot of focus on our personal carbon emissions, rather than figuring out what we can do to best fight climate change.

Asking the second question suggests radically different actions. The best things we can do to fight climate change probably involve working on, advocating for, and donating to exceptional research and advocacy opportunities, rather than worrying about plastic bags, recycling, or turning out the lights.

Why is there so much focus on our personal emissions? One explanation is that ethical views originate from before the 20th century, and sometimes from thousands of years ago. If you were a medieval peasant, your main ethical priority was to help your family survive, without cheating or harming your neighbors. You didn’t have the knowledge, power, or time to help hundreds of people or affect the long-term future.

The Industrial Revolution gave us wealth and technology not even available to kings and queens in previous centuries. Now, many ordinary citizens of rich countries have enormous power to do good, and this means the potential consequences of our actions are usually what’s most ethically significant about them.

But this isn’t to say that we can ignore harms. In general, we think it’s vital to avoid doing anything that seems very wrong from a commonsense perspective, even if it seems like it might lead to a greater social impact, and to cultivate good character

In general, we see our advice as about striving to have a greater impact, within the constraints of your other important values.

So, we think ‘social impact’ or ‘making a difference’ should be about making the world better. But what does that mean?

What does it mean to “make the world better”?

We imagine building a world in which the most beings can have the best possible lives in the long term — lives that are free from suffering and injustice, and full of happiness, adventure, connection, and meaning.

There are two key components to this vision — impartiality and a focus on wellbeing — which we’ll now unpack.

Impartiality: everyone matters equally

When it comes to ‘making a difference,’ we think we should strive to be more impartial — i.e. to give equal weight to everyone’s interests.

This means striving to avoid privileging the interests of anyone based on arbitrary factors such as their race, gender, or nationality, as well as where or even when they live. We also think that the interests of many nonhuman animals should be given significant weight, although we’re unsure of the exact amount. Importantly, we’re also concerned about potentially sentient future digital beings, which could exist in very large numbers and whose welfare could be in part determined by how we design them.

The idea of impartiality is common in many ethical traditions, and is closely related to the “Golden Rule” of treating others as you’d like to be treated, no matter who they are.

Acting impartially is an ideal, and it’s not all that matters. As individuals, we all have other personal goals, such as caring for our friends and family, carrying out our personal projects, and having our own lives go well. Even considering only moral goals, it’s plausible we have other values or ethical commitments beyond impartially helping others.

We’re not saying you should abandon these other goals and strive to treat everyone equally in all circumstances.

Rather, the claim is that insofar as your goal is to ‘make a difference’ or ‘have a social impact,’ we don’t see good reason to privilege any one group over another — and that you should therefore have some concern for the interests of strangers, nonhumans, and other neglected groups.

(And even if you think that the ultimate ideal is to have equal concern for all beings, as a matter of psychology, you probably have other, competing goals, and it’s not helpful to pretend you don’t.)

In Peter Singer’s essay, Famine, Affluence, and Morality, he imagines you’re walking and come across a child drowning in a pond. Everyone agrees that you should run in and save the child, even if it would ruin your new suit and shoes.

This illustrates a principle that many people can get behind: if you can help a stranger a great deal with little cost to yourself, that’s a good thing to do. This shows that most people give some weight to the interests of others.

If it also turns out that you have a lot of power to help others (as we argued above), then it would imply that social impact should be one of the main focuses of your life.

Impartiality also implies that you should think carefully about who you can help the most. It’s common to say that “charity begins at home,” but if everyone’s interests matter equally, and you can help more people who are living far away (e.g. because they’re without cheap basic necessities you can provide), then you should help the more distant people.

We’re convinced that a degree of impartiality is reasonable, and that many people should try thinking harder about impartiality than they are used to. But there remains huge questions about how impartial to be.

The trend over history seems to have been towards a greater impartiality and a wider and wider circle of concern, but we’re unsure where that should stop. For instance, compared to people today, how exactly should we weigh the interests of nonhuman animals, people who don’t exist yet, and potential digital beings? This is called the question of moral patienthood.

Here’s an example of the stakes of this question: we don’t see much reason to discount the interests of future generations simply because they’re distant from us in time. But because there could be so many people in the future, the main focus of efforts to do good should be to leave the best possible world for those future generations. This idea has been called ‘longtermism,’ and is explored in a separate article. We think longtermism is an important perspective, which is part of why we say “over the long term” in our definition of social impact.

This section was about who to help; the next section is about what helps.

Wellbeing: what does it mean to help others?

When aiming to help others, our tentative hypothesis is that we should aim to increase their wellbeing as much as possible — i.e. enable more individuals to live flourishing lives that are healthy, happy, fulfilled; are in line with their wishes; and are free from avoidable suffering.

Although people disagree over whether wellbeing is the only thing that matters morally, almost everyone agrees that things like health and happiness matter a lot — and so we think it should be a central focus in efforts to make a difference.

Putting impartiality and a focus on wellbeing together means that, roughly, how much positive difference an action makes depends on how it increases the wellbeing of those affected, and how many are helped — no matter when or where they live.

What wellbeing consists of more precisely is a controversial question, which you can read about in this introduction by Fin Moorhouse. In brief, there are three main views:

• The hedonic view: wellbeing consists in your degree of positive vs negative mental states such as happiness, meaning, discovery, excitement, connection, and equanimity.
• The preference satisfaction view: wellbeing consists in your desires being fulfilled.
• Objective list theories: wellbeing consists in achieving certain goods, like friendship, knowledge, and health.

In philosophical thought experiments, these different views have very different implications. For instance, if you support the hedonic view, you’d need to accept that being secretly placed into a virtual reality machine that generates amazing experiences is better for you than staying in the real world. If instead you support (or just have some degree of belief in) the preference satisfaction view, you don’t have to accept this implication, because your desires can include not being deceived and achieving things in the real world.

In practical situations, however, we rarely find that different views of wellbeing drive different decisions, such as about which global problems to focus on. The three notions correlate closely enough that differences in views are usually driven by other factors (such as the question of where to draw the boundaries of the expanding circle discussed in the previous section).

What else might matter besides wellbeing? There are many candidates, which is why we say promoting wellbeing is only a “tentative” hypothesis.

Preserving the environment enables the planet to support more beings with greater wellbeing in the long term, and so is also good from the perspective of promoting wellbeing. However, some believe that we should preserve the environment even if it doesn’t make life better for any sentient beings, showing they place intrinsic value on preserving the environment.

Others think we should place intrinsic value on autonomy, fairness, knowledge, and many other values.

Fortunately, promoting these other values often goes hand in hand with promoting wellbeing, and there are often common goals that people with many values can share, such as avoiding existential risks. So again, we believe that the weight people put on these different values has less effect on what to do than often supposed, although they can lead to differences in emphasis.

We’re not going to be able to settle the question of defining everything that’s of moral value in this article, but we think that promoting wellbeing is a good starting point — it captures much of what matters and is a goal that almost everyone can get behind.

How good is it to create a happy person?

We’ve mostly spoken above as if we’re dealing with potential effects on a fixed population, but some decisions could result in more people existing in the long term (e.g. avoiding a nuclear war), while others mainly benefit people who already exist (e.g. treating people who have parasitic worms, which rarely kill people but cause a lot of suffering). So we need to compare the value of increasing the number of people with positive wellbeing with benefiting those who already exist.

This question is studied by the field of ‘population ethics’ and is an especially new and unsettled area of philosophy.

We won’t try to summarise this huge topic here, but our take is that the most plausible view is that we should maximise total wellbeing — i.e. the number of people (again including all beings whose lives matter morally) who exist in all of history, weighted by their level of wellbeing. This is why we say “total” wellbeing in the definition.

That said, there are some powerful responses to this position, which we briefly sketch out in the article on longtermism. For this reason, we’re not certain of this ‘totalist’ view, and so put some weight on other perspectives.

Expected value: acting under uncertainty

How do you know what will increase wellbeing the most?

In short, you don’t.

You have to weigh up the different likelihoods of different outcomes, and act even though you’re uncertain. We believe the theoretical ideal here is to take the action with the greatest expected value compared to the counterfactual. This means taking into account both how much wellbeing our actions could result in, and how likely those outcomes are, and adding them together.

In practice, we try to approximate this with rules of thumb, like the importance, neglectedness, tractability framework.

Going into expected value theory would take us too far afield, so if you want to learn more, check out our separate articles:

• Expected value: how to act when you’re uncertain what will help?
• Counterfactuals and how they change our view of what does good
• Cluelessness: can we know the effects of our actions?

Why do we emphasise respecting other values?

This is to remind us of how much could be left out of our definition, and how radical our uncertainty is.

Many moral views that were widely held in the past are regarded as flawed or even abhorrent today. This suggests we should expect our own moral views to be flawed in ways that are difficult for us to recognise.

There is still significant moral disagreement within society, among contemporary moral philosophers, and, indeed, within our own team.

And past projects aiming to pursue an abstract ethical ideal to the exclusion of all else have often ended badly.

We believe it’s important to pursue social impact within the constraints of trying very hard to:

• Not do anything that seems very wrong from a commonsense perspective
  • E.g. to not violate important legal and ethical rights
• Cultivate good character, such as honesty, humility and kindness
• Respect other people’s important values and to be willing to compromise with them
• Respect your other important personal values, such as your family, personal projects and wellbeing

First, following these principles most likely increases your social impact in the long-term, once you take account of the indirect benefits of following them and the limitations of your knowledge.

Second, many believe following these principles is inherently valuable. We’re morally uncertain so try to consider a range of perspectives and do what makes sense on balance.

You can read more about why we think following principles like the above makes sense no matter your ethical views in these additional articles:

• Is it ever okay to take a harmful job in order to do more good?
• Moral uncertainty: how to act when you’re uncertain about what’s good
• Doing good together: how to coordinate effectively and avoid single-player thinking)
• Ways people trying to do good accidentally make things worse, and how to avoid them

Considering these principles, if we had to sum up our ethical code into a single sentence, it might be something like: cultivate a good character, respect the rights of others, and promote wellbeing for the wider world.

And we think this is a position that people with consequentialist, deontological, and virtue-based ethics should all be able to get behind — it’s just that they support it for different reasons.

Conclusion

We’re not sure what it means to make a difference, but we think our definition is a reasonable starting point that many people should be able to get behind:

“Social impact” or “making a difference” is (tentatively) about promoting total expected wellbeing — considered impartially, over the long term.

We’ve also gestured at how social impact might fit with your personal priorities and what else matters ethically, as well as many of our uncertainties about the definition — which can have a big effect on where to focus.

We think one of the biggest questions is whether to accept longtermism, so we’ve dedicated a whole separate article to that question.

From there, you can start to explore which global problems are most pressing based on whatever definition of social impact you think is correct.

You’ll most likely find that the question of which global problems to focus on is more driven by empirical or methodological uncertainties than moral ones. But if you find a moral question is crucial, you can come back and explore the further reading below.

In short, if you have the extraordinary privilege to be a college graduate in a rich country and to have options for how to spend your career, it’s plausible that social impact, as defined in this way, should be one of your main priorities.

Learn more

Top recommendations

• Why not to take a harmful career even if you think it’ll do more good.
• Podcast: Will MacAskill on moral uncertainty, utilitarianism, and how to avoid being a moral monster
• Podcast: Toby Ord on the perils of maximising the good that you do

Read next

This article is part of our advanced series. See the full series, or keep reading:

• 

  Longtermism: a call to protect future generations

  Read more
• 

  Is it ever OK to take a harmful job in order to do more good? An in-depth analysis

  Read more
• 

  Counterfactuals and how they change our view of what does good

  Read more

The post What is social impact? A definition appeared first on 80,000 Hours.

We’ve argued that you should approach your career like a scientist doing experiments: be willing to test out many different paths and gather evidence about where you can have the most impact.

More generally, this approach of open truth-seeking — being constantly, curiously on the lookout for new evidence and arguments, and always being ready to change our minds — is a virtue we think is absolutely crucial to doing good.

One of our first-ever podcast episodes was an interview with Julia Galef, author of The Scout Mindset (before she wrote the book!).

Julia argues — in our view, correctly — that it’s easy to end up viewing the world like a soldier, when really you should be more like a scout.

Soldiers have set views and beliefs, and defend those beliefs. When we are acting like soldiers, we display motivated reasoning: for example, confirmation bias, where we seek out information that supports our existing beliefs and misinterpret information that is evidence against our position so that it seems like it’s not.

Scouts, on the other hand, need to form correct beliefs. So they have to change their minds as they view more of the landscape.

Acting like a scout isn’t always easy:

• There’s lots of psychological evidence suggesting that we all have cognitive biases that cloud our thinking.
• It can sometimes be really painful to admit you were wrong or to come to think something unpleasant, even if that’s what the evidence suggests.
• Even if you know you should change your beliefs, it’s difficult to know how much they should change in response to new evidence — the subject of our interview with Spencer Greenberg.
• Having good judgement is actually just a difficult skill that needs to be practiced and developed over time.

But if we want to form correct beliefs about the world and how it works, we have to try.

And if we want to do good, forming correct beliefs about how our actions will impact others seems pretty crucial.

Why are we talking about this now?

Up until a few weeks ago, we’d held up Sam Bankman-Fried as a positive example of someone pursuing a high-impact career, and had written about how we encouraged him to use a strategy of earning to give.

Sam had pledged to donate 99% of his earnings to charity — and a year ago his net worth was estimated to be more than $20 billion. We were excited about what he might achieve with his philanthropy.

On November 11, Sam’s company, FTX, declared bankruptcy, and its collapse is likely to cause a tremendous amount of harm.

Sam appears to have made decisions which were, to say the least, seriously harmful.

If newspaper reports are accurate, customer deposits that were meant to be safely held by FTX were being used to make risky investments — investments which left FTX owing billions of dollars more than it had.

These reported actions are appalling.

We failed to see this coming.

So this week’s thoughts on the scout mindset are as much a reminder for us at 80,000 Hours as anyone else.

In the coming weeks and months, we want to thoughtfully examine what we believed and why — and in particular, where we were wrong — so we can, where needed, change our views.

There are so many questions for us to consider, in order to shape our future actions — here are a few:

• To what extent is the collapse of FTX evidence that it was always going to be hard to earn to give on a very large scale without causing harm?
• The FTX Future Fund was a major philanthropic funder in many of our top problem areas. Is the resignation of the Fund’s staff, and the collapse of the Fund, evidence that we should be more suspicious of very rich donors by default? In what ways might funding sources affect the approaches taken in attempting to solve problems? In particular, how should we make sure our research findings aren’t unduly influenced by our funders?
• How does the loss of potential funding for some of the world’s most pressing problems affect their neglectedness?
• How should this impact our views of which career paths seem highest impact?
• Should we be conducting more research into topics like how people trying to do good can end up making things worse?
• How should the effective altruism community change in response to this moving forward?

There aren’t yet clear answers to these questions.

As we learn more about what happened and the wider effects of the collapse of FTX, we’re going to do our best to act like scouts, not soldiers: to defend beliefs only if they’re worthy of defence, and to be prepared and ready to change our minds.

We’ve released a statement regarding the collapse of FTX, and hope to write more on the topic soon.

We’re optimistic that the work of identifying, prioritising, and pursuing solutions to some of the world’s most pressing problems will continue. We hope you’ll be with us in that project.

Learn more

• Rob Wiblin shares his personal reaction to the FTX bankruptcy on The 80,000 Hours Podcast (November 23)
• Why “scout mindset” is crucial to good judgment — Julia Galef’s TEDx Talk
• Podcast (September 2017): Julia Galef on making humanity more rational, what EA does wrong, and why Twitter isn’t all bad
• Ways people trying to do good can make things worse, and how to avoid them
• Considering considerateness: why communities of do-gooders should be exceptionally considerate
• Doing good together: how to coordinate effectively, and avoid single-player thinking
• From Open Philanthropy’s co-CEO Holden Karnofsky: Some comments on recent FTX-related events (November 10) and My takes on the FTX situation will (mostly) be cold, not hot (November 18)
• A job matching service for affected FTX Future Fund grantees (November 21)

The post Why being open to changing our minds is especially important right now appeared first on 80,000 Hours.