Someone recently cross-posted this to the main EA Facebook group, where it got a lot of attention from people who came to EA from sources other than Less Wrong. I think the essay is a classic, so I'm cross-posting it to the Forum, where I expect at least a few people to see it who hadn't seen it before.

Yesterday:

There is this very, very old puzzle/observation in economics about the lawyer who spends an hour volunteering at the soup kitchen, instead of working an extra hour and donating the money to hire someone...

If the lawyer needs to work an hour at the soup kitchen to keep himself motivated and remind himself why he's doing what he's doing, that's fine.  But he should also be donating some of the hours he worked at the office, because that is the power of professional specialization and it is how grownups really get things done.  One might consider the check as buying the right to volunteer at the soup kitchen, or validating the time spent at the soup kitchen.

I hold open doors for little old ladies.  I can't actually remember the last time this happened literally (though I'm sure it has, sometime in the last year or so).  But within the last month, say, I was out on a walk and discovered a station wagon parked in a driveway with its trunk completely open, giving full access to the car's interior.  I looked in to see if there were packages being taken out, but this was not so.  I looked around to see if anyone was doing anything with the car.  And finally I went up to the house and knocked, then rang the bell.  And yes, the trunk had been accidentally left open.

Under other circumstances, this would be a simple act of altruism, which might signify true concern for another's welfare, or fear of guilt for inaction, or a desire to signal trustworthiness to oneself or others, or finding altruism pleasurable.  I think that these are all perfectly legitimate motives, by the way; I might give bonus points for the first, but I wouldn't deduct any penalty points for the others.  Just so long as people get helped.

But in my own case, since I already work in the nonprofit sector, the further question arises as to whether I could have better employed the same sixty seconds in a more specialized way, to bring greater benefit to others.  That is: can I really defend this as the best use of my time, given the other things I claim to believe?

The obvious defense—or perhaps, obvious rationalization—is that an act of altruism like this one acts as an willpower restorer, much more efficiently than, say, listening to music.  I also mistrust my ability to be an altruist only in theory; I suspect that if I walk past problems, my altruism will start to fade.  I've never pushed that far enough to test it; it doesn't seem worth the risk.

But if that's the defense, then my act can't be defended as a good deed, can it?  For these are self-directed benefits that I list.

Well—who said that I was defending the act as a selfless good deed?  It's a selfish good deed.  If it restores my willpower, or if it keeps me altruistic, then there are indirect other-directed benefits from that (or so I believe).  You could, of course, reply that you don't trust selfish acts that are supposed to be other-benefiting as an "ulterior motive"; but then I could just as easily respond that, by the same principle, you should just look directly at the original good deed rather than its supposed ulterior motive.

Can I get away with that?  That is, can I really get away with calling it a "selfish good deed", and still derive willpower restoration therefrom, rather than feeling guilt about it being selfish?  Apparently I can.  I'm surprised it works out that way, but it does.  So long as I knock to tell them about the open trunk, and so long as the one says "Thank you!", my brain feels like it's done its wonderful good deed for the day.

Your mileage may vary, of course.  The problem with trying to work out an art of willpower restoration is that different things seem to work for different people.  (That is:  We're probing around on the level of surface phenomena without understanding the deeper rules that would also predict the variations.)

But if you find that you are like me in this aspect—that selfish good deeds still work—then I recommend that you purchase warm fuzzies and utilons separately.  Not at the same time.  Trying to do both at the same time just means that neither ends up done well.  If status matters to you, purchase status separately too!

If I had to give advice to some new-minted billionaire entering the realm of charity, my advice would go something like this:

• To purchase warm fuzzies, find some hard-working but poverty-stricken woman who's about to drop out of state college after her husband's hours were cut back, and personally, but anonymously, give her a cashier's check for $10,000.  Repeat as desired.
• To purchase status among your friends, donate $100,000 to the current sexiest X-Prize, or whatever other charity seems to offer the most stylishness for the least price.  Make a big deal out of it, show up for their press events, and brag about it for the next five years.
• Then—with absolute cold-blooded calculation—without scope insensitivity or ambiguity aversion—without concern for status or warm fuzzies—figuring out some common scheme for converting outcomes to utilons, and trying to express uncertainty in percentage probabilitiess—find the charity that offers the greatest expected utilons per dollar.  Donate up to however much money you wanted to give to charity, until their marginal efficiency drops below that of the next charity on the list.

I would furthermore advise the billionaire that what they spend on utilons should be at least, say, 20 times what they spend on warm fuzzies—5% overhead on keeping yourself altruistic seems reasonable, and I, your dispassionate judge, would have no trouble validating the warm fuzzies against a multiplier that large.  Save that the original, fuzzy act really should be helpful rather than actively harmful.

(Purchasing status seems to me essentially unrelated to altruism.  If giving money to the X-Prize gets you more awe from your friends than an equivalently priced speedboat, then there's really no reason to buy the speedboat.  Just put the money under the "impressing friends" column, and be aware that this is not the "altruism" column.)

But the main lesson is that all three of these things—warm fuzzies, status, and expected utilons—can be bought far more efficiently when you buy separately, optimizing for only one thing at a time.  Writing a check for $10,000,000 to a breast-cancer charity—while far more laudable than spending the same $10,000,000 on, I don't know, parties or something—won't give you the concentrated euphoria of being present in person when you turn a single human's life around, probably not anywhere close.  It won't give you as much to talk about at parties as donating to something sexy like an X-Prize—maybe a short nod from the other rich.  And if you threw away all concern for warm fuzzies and status, there are probably at least a thousand underserved existing charities that could produce orders of magnitude more utilons with ten million dollars.  Trying to optimize for all three criteria in one go only ensures that none of them end up optimized very well—just vague pushes along all three dimensions.

Of course, if you're not a millionaire or even a billionaire—then you can't be quite as efficient about things, can't so easily purchase in bulk.  But I would still say—for warm fuzzies, find a relatively cheap charity with bright, vivid, ideally in-person and direct beneficiaries.  Volunteer at a soup kitchen.  Or just get your warm fuzzies from holding open doors for little old ladies.  Let that be validated by your other efforts to purchase utilons, but don't confuse it with purchasing utilons.  Status is probably cheaper to purchase by buying nice clothes.

And when it comes to purchasing expected utilons—then, of course, shut up and multiply.

In this post, some ACE staff members discuss the contributions they’ve made to charities in 2019. This is intended to give interested readers examples of different approaches to charitable giving, from some of the people who are most familiar with our materials.

Heather Herrell, Director of Philanthropy

I like to think of philanthropy as an investment in a better future. While I have given to a variety of charities over the decades (primarily animal organizations), this year I took the Try Giving Pledge to officially commit to donating at least 5% of my income, which I have happily exceeded. I relied significantly on ACE’s annual recommendations to help direct my 2019 giving.

This year, I continued my monthly support of ACE because I genuinely value the contributions our organization makes to the animal advocacy movement. I also made several donations to a number of our effective Top and Standout Charities that are doing impactful work around the globe to reduce animal suffering. On Giving Tuesday I took advantage of Facebook’s $7M in matching funds, successfully organized by the EA Giving Tuesday group (thank you!). The Double Up Drive and Effective Animal Advocacy Fund Matching Challenge were additional incentivizing giving opportunities to increase the amount and potential impact of my donations.

I’m looking forward to giving more in 2020! Imagine how much good we could do if everyone gave something.

Erika Alonso, Director of Communications

My partner Adam and I share finances, but he largely defers to my judgment on where we donate. Our donations throughout 2019 were very similar to prior years and along the same lines of reasoning. Similar to last year, we allocated most of our donations to funds as opposed to individual charities—this year we donated to both ACE’s Effective Animal Advocacy Fund and Recommended Charity Fund. We’re excited to diversify our giving portfolio based on ACE’s recommendations, not only across Top and Standout Charities but to small but promising charities working within the animal advocacy space as well. We also made small one-time and monthly donations to ACE and ACE’s Top Charities.

We also continued our sponsorship of a chicken with Farm Sanctuary. This donation is a gift to my niece in lieu of other gifts, and we hope it inspires her to think differently about farmed animals. Additionally, we donated a small amount to nonprofits working on empowering women throughout the world via the new Momentum Giving app.

Our last donation of the year will go to Strong Minds, as they are working in a cause area that resonates with me on a personal level, and one that we believe is high impact and largely neglected. In the future, I hope to see more mental health-focused charities recommended within the effective altruism community as it’s a cause area that I strongly support, but I am uncertain about the effectiveness of organizations working in this space.

Gina Stuessy, Director of Operations

I consider myself a part of the effective altruism (EA) movement, and each year I aim to contribute at least 10% of my income to high-impact causes (I took the Giving What We Can Pledge). This year I made one of my two main donations on Giving Tuesday to take advantage of the Facebook matching opportunity: A few EAs worked very hard on the Giving Tuesday match to collect pledges from donors and provide best tips for getting donations matched as well as statistics afterward (thank you!). All together, EAs donated over $1.2 million as part of this single-day effort. My donation went to ACE to help meet our funding needs.

A couple of weeks later I realized I had enough to make another donation, and this time I gave to ACE’s Effective Animal Advocacy Fund—I’m excited to help support many animal charities working around the world with programs that our team finds to be quite promising. Additionally, donations to that fund are currently being matched by a generous, anonymous donor!

Jamie Spurgeon, Researcher

The majority of my giving this year has been split between both of ACE’s funds, the Recommended Charity Fund and the Effective Animal Advocacy Fund. Between the two funds, my contributions can help fund the most effective charities in the movement and also support a healthy pluralistic movement by ensuring funding is allocated across effective charities of all sizes. While I have a good sense of how I would allocate the funding myself within each Fund, I find that going through a team decision process consistently yields better results, and I greatly value the viewpoints of my colleagues when we make these funding decisions. In particular, I think that our second round of granting from the Effective Animal Advocacy Fund was a substantial improvement from the first round, and I look forward to seeing continual improvements as we move forward.

I have also donated specifically to The Good Food Institute (GFI) following their renewed status as a Top Charity. Aside from being continually impressed with the work that GFI does, I suspect that they are playing a crucial role in the move away from factory farming. Their work is both institutional in focus and serves to reduce the barrier non-vegans face in reducing their animal product consumption, which I think will be crucial for wider-scale societal change and moral circle expansion.

I have also donated a smaller amount to Rethink Priorities. I have been impressed with the work they’ve done in the last year and found it useful for informing my own work, particularly on estimating cost-effectiveness. I’m excited to see what the coming years bring for them.

Melissa Guzikowski, Managing Editor

I always directed the majority of my donations to animal charities. However, since I spend a large portion of my day-to-day life advocating for animals, every year I also direct a smaller amount of monetary resources to non-animal charities. I believe that helping to address human suffering is valuable both intrinsically and instrumentally as a tool to a brighter future for all species. This year, I donated directly to GiveWell, an effective altruism organization that conducts research to find the most effective giving opportunities.

For animal charities, I donated to The Good Food Institute (GFI) and The Humane League (THL), two of ACE’s four 2019 Top Charities. I support GFI because they promote the development of cellular agriculture. Though I believe that humanity’s moral circle is expanding, I am skeptical that the world is going to take on a vegan ethic in the near future. As such, I believe it’s essential that we create attractive alternatives to traditional animal products that can appeal to many people in the meantime. I’m also particularly impressed by THL’s work; I admire their strategic approach to advocacy and their commitment to collaborating with other groups in the movement. I donated to THL on Giving Tuesday and had my donation matched.

This year, I also donated to ACE’s Effective Animal Advocacy Fund (EAA Fund). The EAA Fund supports a wide range of groups that might not otherwise be eligible for ACE’s recommended charity status due to factors such as their small sizes or short track records. This donation was also doubled through an ACE matching challenge.

Cash Callaghan, Field Research Associate

I gave most of my donations this year to ACE’s Recommended Charity Fund and Effective Animal Advocacy Fund. I think it may be more effective to donate to funds at ACE’s discretion because the marginal value of donations to charities may change depending on how much of charities’ funding gaps have been filled at the time of disbursement. In general, I have high fidelity in my colleagues who deliberate together to determine how the money from those funds is dispersed. Their expertise and the group decision-making process makes me confident that the money from these funds will be allocated as effectively as possible. Rather than donating to one or a few charities, I hope that by giving to these funds my donations are spread out between a range of highly effective and promising charities. I especially hope that the Effective Animal Advocacy Fund will help expand and diversify the animal advocacy movement by supporting younger organizations that are doing promising work.

I also donated through Facebook’s Giving Tuesday Donation Match Campaign, in hopes that my donations will be counterfactually matched by Facebook (thank you to the EA Giving Tuesday team who helped organize this effort!). Specifically, I donated to Wild Animal Initiative, Fish Welfare Initiative, Aquatic Life Institute, and Rethink Priorities. Wild Animal Initiative researches effective ways to help wild animals, a relatively high scale and neglected cause area within animal advocacy. I am hopeful that their work will help inform and empower efforts to improve the welfare of wild animals. Fish Welfare Initiative and Aquatic Life Institute are both new nonprofits dedicated to improving fish welfare, which is also a relatively high scale and neglected cause area within animal advocacy. I am excited to see more from these two organizations in the coming year and to support them in their efforts to expand the growing field of fish advocacy. Rethink Priorities conducts foundational research on neglected causes (including causes related to animal advocacy). In the past year, Rethink Priorities has published articles related to animal advocacy that I have used in my own work, including their work related to corporate campaigns, invertebrate welfare, wild animal welfare, accuracy issues with FAO animal estimates, and a collection of effective animal advocacy resources.

I am very grateful for the work these organizations do to help animals, and I look forward to seeing what they accomplish over the next year.

Marianne van der Werf, Effective Animal Advocacy Fund Program Officer

In general, I think that the largest amount of donations should go towards each of ACE’s Top Charities, a large amount should go to each of the ACE’s Standout Charities, and some amount should go to each promising newer or smaller organization, such as ACE’s Effective Animal Advocacy Fund (EAA Fund) grantees. For deciding my own donations I take this “ideal” distribution of additional resources into account as well as what I believe the real distribution will be, based on ACE’s giving metrics reports from 2018, 2017, and 2016. Based on this, I decided to give half of my donations this year to three of ACE’s Standout Charities, for which I selected Faunalytics, the Sociedade Vegetariana Brasileira, and the Federation of Indian Animal Protection Organisations (FIAPO).

I am giving the other half of my donations to ACE’s EAA Fund, for which I’m program officer. I’m excited that the Fund allows us to support a more diverse range of programs that complement ACE’s Recommended Charities. This donation stands out from my other donations because it can either result in one EAA Fund grantee receiving a larger donation, or it can result in us being able to support an additional applicant.

Lastly, I donated a smaller amount to two EAA Fund grantees that I was especially impressed with after receiving their grant updates: Essere Animali and Christopher Sebastian McJetters. I believe McJetters’ work contributes to a more sustainable, equitable, and effective movement, and I support him via his Patreon. I appreciate Essere Animali’s work on fish welfare, as well as the fact that they used part of the EAA Fund grant towards improving their internal structures and communications between staff, which I believe is key to building an effective organization.

My notes on the article

• I think people have generally internalized "move along quickly" within EA orgs, but that may be the most underutilized advice in the article among people working in non-EA roles (many of whom are members of this community)
• In my experience, many EA organizations give new hires a lot of room to define their roles if those hires have the necessary skills/experience for their desired job descriptions. The "roles that don't exist yet" advice has been relevant to me at multiple times in my career.
• I strongly second the suggestion that becoming world-class at something has a lot of value that can't necessarily be matched by being "okay" at a high-priority position. You can see more on the subject in my response to a world-class composer of obscure music who asked about how that work might compare to a traditional EA career. This is also one reason that I continue to put work into being a world-class Magic: the Gathering player; even if I don't win much money, being a respected author and streamer within that community could open up a lot of opportunities for public messaging and connections to interesting people in other fields.

The article

The following are excerpts from interviews with people whose work we respect and whose answers we offered to publish without attribution. This means that these quotes don’t represent the views of 80,000 Hours, and indeed in some cases, individual pieces of advice explicitly contradict our own. Nonetheless, we think it’s valuable to showcase the range of views on difficult topics where reasonable people might disagree.

The advice is particularly targeted at people whose approach to doing good aligns with the values of the effective altruism (EA) community, but we expect most of it is more broadly useful.

This is the seventh in this series of posts with anonymous answers. Other entries have asked:

1. “Is there any career advice you’d be hesitant to give if it were going to be attributed to you?”
2. “How have you seen talented people fail in their work?”, and
3. “What’s the thing people most overrate in their career?”
4. “If you were 18 again, what would you do differently this time around?” And other personal career reflections.
5. How risk averse should talented young people be?
6. What bad habits do you see among people trying to improve the world?

What mistakes do people most often make when deciding what work to do?

Doing something that they don’t enjoy at all

I don’t think you should just do what you’re passionate about, but I also don’t think you should do whatever has the highest expected value to the world if it’s something that you’re going to hate.

When I started out in my career, the things I was told to work on needed to be done — but they were incredibly draining for me personally. And even though there were other things I could have been working on that I love to do, things that others might find unbearable — there was no attempt to work out what I enjoyed or what I was good at.

I think organisations generally need to become more receptive to finding specific tasks for specific people — something that they’ll be good at and at least somewhat motivated by.
For individuals, it’s hard to get sample data on what a job will actually be like — but not impossible. You can talk to a whole bunch of people working in that area, find out which specific things they’re working on that sound exciting to you. Read as much as you can about the area, and figure out what you think the most effective positions are.

Critically, once you do take a new job — immediately start thinking “is there something else that’s a better fit?” There’s still a taboo around people changing jobs quickly. I think you should maybe stay 6 months in a role just so they’re not totally wasting their time in training you — but the expectation should be that if someone finds out a year in that they’re not enjoying the work, or they’re not particularly suited to it, it’s better for everyone involved if they move on. Everyone should be actively helping them to find something else.

Doing something you don’t enjoy or aren’t particularly good at for 1 or 2 years isn’t a tragedy — but doing it for 20 or 30 years is.

You can get a sense for what the most effective role is from outside, but you’ll only find out if you enjoy it by actually doing it — so people should be way more willing to jump ship.

Not focusing on becoming really good at something

One of my biggest disagreements with effective altruism (EA) conventional wisdom relates to my belief in how incredibly valuable it is to achieve a high level of general status and career accomplishment.

If you can be an ‘okay’ AI safety researcher, assuming that’s the best thing to be, versus being truly world-class and remarkable at some other job that might not be very important in itself, I think the second one is probably the better call.

When you really do well at something you become in demand. You get a lot of opportunities. I see a lot of people right now who have incredible opportunities to do important things in the AI world, and those opportunities will never exist for an ‘okay’ AI safety researcher. The reason they have those opportunities is because they got to the top of something. And now, they know people. And they have impressive abilities that are in demand and hard to find. And so OpenAI, or DeepMind, or the government is interested in them, they want their help.

You also get friends and connections that no one else has. I think in a lot of cases, the best way to get truly outstanding people interested in doing effective altruism-focused things is to actually be friends with them. And having something you’re extremely good at can put you in a good position to make those kinds of connections.

So one side is: how valuable is it to be really good at something? And the other side is: how hard is it?

I think people underestimate how hard it is. I wonder if people get a weird, misleading model of this from college. Because in college, everything is on a four-year clock at most, often more like a few-month clock. So a lot of times you can get your A in your class, or you can be president of your club, by just working unsustainably hard for a short period of time.

But in the real world, I think to be really great at something, to be better than other people — you probably need to be a crazily good fit, and you need to put in a really crazy number of total hours that don’t have any relationship to how many hours you put into anything for a few months in college. As in you might need something like 30-80 good hours a week for 45-50 weeks a year for at least 5 years, often 10-20 years, in order to get to this world-class level.

This is why I often put more weight than other people put on things like being really excited about a job, working on things you naturally like to work on, and picking things you think play to your strengths. I think that generally those kinds of things are really important, and so usually my advice to someone early in their career is to really focus on learning about yourself, and what you like, and becoming better at things, and finding something that you can do really well. Don’t pick a job because a website told you to pick the job.

Following the paths of similar people

I think people often try to figure out the question “where do people like me work?” Rather than asking “where could my skills be most useful?”

This means that collectively we can be slow to get new things going.

Say there’s some social science problem, and what’s really needed to help solve the problem is people with really strong technical skills.

But the technically skilled people look at different fields and think “the people who seem impressive to me are professors in maths, or physics, or people going into AI”. Then they look at the social scientists and think “these people don’t seem like they’re good technically, I don’t want to be like them” — not noticing that if they went into that field they wouldn’t be like that.

Not thinking about roles that don’t exist yet

Being really focused on what roles are available now. Take the example of someone who’s an excellent communicator — maybe you tell them they should aim to be a communication officer at an effective altruism themed org. But they say “well, there aren’t any roles open for that”. Okay, that might be literally true. But it’s the wrong mindset; if you become excellent at a broad skill, like writing, public speaking or management, a skill that’s in demand in the effective altruism community — people will fight over you. They’ll fight over you like crazy.

Just thinking about what roles are open right now is only relevant if you literally need to get a job in the next few months — and even then, you should try to find out whether there’s a role that could be created for you, or non-public openings.

If you’re a young person starting out, just go and develop the skills you need outside of the effective altruism community for a few years. And then come back once you’re a superstar writer, speaker, or manager.

Working on the most interesting puzzles

I don’t know if this is a mistake, but it’s something that a lot of people do — they work on whatever happens to pique their interest. I think if you are naturally a problem solver, there are lots of super interesting puzzles in the world, and it can be really hard to resist these. But a lot of the most interesting puzzles in the world are not that critical to solve, while many of the largest problems in the world aren’t necessarily the most interesting puzzles.

I think if you have a mind of someone who likes proving things, you also want a definitive solution to a problem that was very difficult to solve. And yet a lot of the biggest problems in the world don’t come in that form; they’re partially empirical and involve a lot of uncertainty, and so you can’t always get a definitive solution.

At the same time, it’s very hard to work on something that you’re not intellectually interested in. So the advice is probably: try to find the intersection of problems that you find intellectually stimulating, and problems that are really important.

I see a lot of people optimising only for things that are intellectually interesting. A lot of the most brilliant minds I know are working in super abstract areas of logic and mathematics.

I totally see how these are interesting problems, and that they have great minds for these, but I wish I could see what kind of progress they could make on one of these much messier, more important problems.

But the opposite is also true. If you find that your work isn’t intellectually stimulating, then it’s going to be really hard to work on it, and it’s probably going to make you sad — you haven’t found the intersection in those cases.

Not thinking enough about developing skills

There’s insufficient attention paid to the question of “what skills will I learn at this job?”

E.g. if you think you have the seeds of being a great manager, is this a place where you’ll develop that skill? People can be really vague about career capital… think about what skills you actually want to develop, check if you will develop them in the job you’re considering.

Not spreading out among many different fields

For effective altruism to be successful as a community, it needs to have work happening in lots of different areas. You might keep hearing about how AI is the cool thing to work on, but it would be a pretty big mistake if everyone actually worked on AI.

Doing a variety of work is not only important for those different fields, but it’s also really valuable in terms of building a healthy and functional intellectual community — compared to a group of people having one swing at one particularly important problem, assuming their world-view turns out to be right.

Assuming direct work is best

I think there’s too much binary thinking around direct work. Often either people think they’re just automatically doing the right thing by working directly, or if they’re not working at an ‘EA org’ — they think their work is useless to the community.

There’s also too much of a focus on “what cause area am I working on?” and not enough on “what kinds of skills am I developing as an early career person trying to do good in the long-term?”

Valuing breadth over depth

I think people put too much emphasis on breadth rather than depth. People are generally not going to reward you for breadth — compared to specialising in an area and going really in depth. It makes sense to think that going for breadth is going to be more interesting, but people probably have the wrong expectations about who else might value it. Others will generally value expertise in a very specific area.

Thinking that their cause is “the one true cause”

There are a lot of people who think that their cause area is “the one true cause area.” Strategically, this isn’t a good place to start a conversation with someone who has different goals. It also just lacks perspective.

There are existential risk focused people who think “how could you work on animal welfare instead of the millions of future generations of people who might exist?.” There are animal welfare people who think “how could you work on global poverty when there are 9 billion chickens who suffer every day?”.

But some people aren’t going to be attracted to saving millions of future generations, and some people aren’t going to be attracted to preventing the suffering of factory farmed animals. And that should be okay.

Does the following seem like a reasonable brief summary of the key disagreements regarding AI risk?

Among those experts (AI researchers, economists, careful knowledgeable thinkers in general) who appear to be familiar with the arguments:

• Seems to be broad (but not universal?) agreement that:
• Superintelligent AI (in some form, perhaps distributed rather than single-agent) is possible and will probably be created one day
• By default there is at least a decent chance that the AI will not be aligned
• If it is not aligned or controlled in some way then there is at least a decent chance that it will be incredibly dangerous by default
• Some core disagreements (starred questions are at least partially social science / economics questions):
• * Just how likely are all of the above?
• * Will we have enough time to see it coming, and will it be obvious enough, that people will react appropriately in time to prevent bad outcomes?
• Still might be useful to have some people keeping tabs on it (Robin Hanson thinks about 100), but not that many
• How hard is it to solve?
• If easy then less time needed to see it coming, or inventors more likely to incorporate solutions by default
• If really hard then may need a long time in advance
• * How far away is it?
• Can we work on it profitably now given that we don't know how AGI will work?
• If current ML scales to AGI then presumably yes, otherwise disagreement
• * Will something less than superhuman AI pose similar extreme risks? If yes: How much less, how far in advance will we see it coming, when will it come, how easy is it to solve?
• * Will we need coordination mechanisms in place to prevent dangerous races to the bottom? If yes, how far in advance will we need them?
• If it's a low probability of something really catastrophic, how much should we be spending on it now? (Where is the cutoff where we stop worrying about finite versions of Pascal’s Wager?)
• * What about misuse risks, structural risks, or future moral risks?
• Various combinations of these and related arguments result in anything from "we don't need to worry about this at all yet" to "we should be pouring massive amounts of research into this"

You can find the PDF version of this report here.

Introduction

This report lays a foundation for future research into projects to improve wild insect welfare by promoting more humane insect pest management practices. I summarize recent insect sentience literature and estimate the number of insects impacted by agricultural insecticide use in the United States. I then examine the usage and modes of action of a variety of common insecticides and non-insecticidal pest management practices, highlighting knowledge gaps and suggesting directions for future inquiry. As a companion resource to this report, I have compiled a database of insecticidal compounds^[1], their modes of action, and their insecticidal mechanisms. This database is under development and may be expanded to include pests targeted, brand names, and chemical fact sheets where available. I have also developed a rough impact estimate table, outlining a method for using a pest control literature review to calculate the minimum number of insects affected by U.S. agricultural insecticide use.

Insect Sentience

The expected value of an intervention to improve the welfare of wild insects will depend on the confidence we have in insects being sentient, and how intense we think their experiences are. For the purpose of this report, I define sentience as primary (or phenomenal) consciousness: the capacity to have mental experiences (Allen & Trestman 2017). I refer to positive and negative mental experiences as affective states, which typically occur in response to pleasant or aversive stimuli. So defined, sentience can occur in the absence of secondary (or access) consciousness: the ability to reflect upon or report those experiences. While other mental experiences also contribute to a sentient individual’s welfare, the type of experience that is particularly relevant to the use of insecticides is pain, or an “unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage” (IASP 2017). Determining whether or not an organism is likely to feel pain is important for understanding where we should focus our welfare efforts.

What kinds of organisms might be sentient?

If we assume that sentience is a fundamentally biological phenomenon, then it will be subject to processes such as natural selection (Mellor 2016, Dawkins 2008). One theory of sentience suggests that it may confer a selective advantage to organisms whose reproductive success depends on their ability to process a constant influx of complex information from their dynamic environments, and integrate that information into a mental map, comprising the environment and themselves in relation to it (Klein & Barron 2016a, Mallatt & Feinberg 2016, Tebbich et al. 2016, Morand-Ferron et al. 2015). If this is true, then motile organisms and organisms with multiple sensory modalities or high sensory resolution are more likely to be sentient, as the capacity for first-order mental experiences would probably improve their reproductive success (Klein & Barron 2016a, Mallatt & Feinberg 2016).

The principal anatomical evidence against insect sentience is their brain size. Insect brains have substantially fewer neurons than mammals; honey bee and cockroach brains have about 1 million neurons, whereas mouse brains have about 70 million neurons (Giurfa 2007, Herculano-Houzel et al. 2006, Fox 2006). Insect brains are certainly smaller than vertebrates’, but they are also highly efficient, supporting complex behaviors with fewer neurons and less redundancy than vertebrate brain regions with analogous functions (Giurfa 2007, Herculano-Houzel et al. 2006, Fox 2006, Greenspan & van Swinderen 2004). Although it seems likely that there is a lower limit to the number of neurons required to support sentience (Adamo 2016), research into learning capacities of small brains over the past twenty years suggests that the limit may be lower than previously thought. If so, at least some insect species (e.g., fruit flies, dragonflies, ants, and honeybees) meet the basic requirements (Haberkern & Jayaraman 2016, Greenspan & van Swinderen 2004).

Despite their relatively low neuron count, some insects nevertheless show unexpected behavioral sophistication that suggests they are capable of internal representation and affective states, such as associative learning (Haberkern & Jayaraman 2016, Giurfa 2007)^[2]. Long-term change in an individual’s behavior in response to an aversive stimulus could be motivated by a negative affective state. On the other hand, it is possible for at least some instances of behavior change to occur without the individual being motivated by a mental experience. Nociceptor sensitization can have similar behavioral outputs without the associated mental state, and robots (which are presumed not to be sentient) can be programmed to respond to an aversive stimulus as a learning motivator that drives behavior change (Sneddon et al. 2014).

Like vertebrates, an insect’s nervous system is centralized, meaning that information from all over the body is transmitted to a relatively dense concentration of neurons where it is filtered and integrated. Although insects lack a cerebral cortex, which some researchers consider a necessary feature for experiencing affective states, there is emerging evidence that subcortical structures play a role in vertebrate sentience (Koch et al. 2016, Klein & Barron 2016a and 2016b, Mallatt & Feinberg 2016). While the cerebral cortex may be required for secondary (access) consciousness, the midbrain, basal ganglia, hippocampus, and cerebellum seem to be responsible for integrating external stimuli, past experiences, and proprioceptive feedback into a mental simulation of the organism in the environment (Haberkern & Jayaraman 2016, Merker 2006). These subcortical structures are also thought to support emotion and memory in vertebrates, and regions of the insect brain (including the central complex and mushroom bodies) appear functionally analogous and perhaps even homologous to these parts of the vertebrate brain (Barron & Klein 2016, Haberkern & Jayaraman 2016, Pfeiffer & Homberg 2014, Strausfeld & Hirth 2013). If insect brains are, in fact, functionally analogous to these subcortical structures, they may be complex enough to support sentience (Barron & Klein 2016, Mallatt & Feinberg 2016).

How should we apply what we learn about sentience?

The affective states of other individuals cannot be directly observed, and very few species seem capable of reporting their mental experiences. We have yet to find conclusive anatomical, physiological, or behavioral correlates of sentience. Many of the proxies currently used (but especially such measures as encephalization quotient, neuron count, and phylogenetic distance from humans) are vulnerable to taxonomic bias, since our concept of sentience and its indicators are based on humans and our close relatives. These traits may not be generalizable across taxa, or even across life stages of the same species (Adamo 2017). This means the best we can do is to accumulate as much information as possible on relevant measurable traits, and use that evidence to triangulate around the issue of sentience (Sneddon et al. 2014). The current case for insect sentience is not fully compelling, but it is plausible. Further behavioral studies of a wider variety of insect taxa are necessary in order for us to be more certain about their moral status. As it stands, the best strategy is to consider the expected value of our actions: our concern for beings about whose sentience we are uncertain should be weighted by the strength of the evidence and the size of the potential harm. This approach is consistent with the spirit of the arguments made by welfare-concerned philosophers (Birch 2017, Knutsson & Munthe 2017), and the recommendations of the American Veterinary Medical Association (AVMA 2013).

Impact of an Agricultural Insecticides Intervention

Why agricultural insecticides?

If we think it is plausible that many insects are sentient, then their enormous abundance should compel us to try to alleviate the suffering they may experience (Ray 2018, Tomasik 2007). Improving wild insect welfare by altering insecticide practices is a promising intervention for several reasons. While many open questions remain (principally: whether or not insects can experience pain, how painful different insecticidal methods are, and how painful non-insecticide-related deaths are), there are several reasons a welfare intervention for insects impacted by agriculture seems more tractable than one for insects in environments less-influenced by human activity. Crucially, reducing the pain of death without modifying the total number of deaths avoids introducing the new knock-on effects we would expect from interventions that change the total population of a species, making it easier to determine the net effects of the intervention on the system as a whole. Because agricultural lands are intentionally less diverse and less spatially variable than most wild systems, detecting unexpected knock-on effects should also be easier. Interventions on agricultural lands should also be less controversial than interventions in areas which are further from the sphere of direct human influence. Altering an existing practice (i.e. agricultural pest management) is probably more economically and socially tractable than introducing a new practice (e.g. providing food and shelter to insects within the borders of a wildlife reserve). Lastly, insecticide use is already experiencing a decline: the decrease in the amount of insecticidal compounds applied may be partially attributable to increased scrutiny about their environmental and human health impacts (Atwood & Paisley-Jones 2017). If it turns out that alternative insect pest management techniques are expected to be more humane than these insecticidal compounds, advocating for more humane insecticide practices would dovetail with existing environmentalist and public health efforts.

The effectiveness of an intervention to improve wild insect welfare by reducing the painfulness of agricultural insect pest management methods depends on the likelihood that the affected insects are capable of suffering, the number of insects that would be impacted by the intervention, and how much the intervention would improve their welfare. Other considerations are the effects the intervention may have on the welfare of other beings, and the cost of the intervention. The scope of this report is limited to insecticide use on agricultural lands in the United States, and primarily U.S. cropland, which comprises 9% of the world’s croplands (USGS 2017).

How many insects are there?

Most available data about insect abundance only reflect species richness and biomass^[3] as a measure of relative abundance among samples, and not the absolute number of individuals in an area. Existing estimates of the total number of insects alive on Earth at any given time range from 1E+17 to 1E+19 (Ray 2017a, Williams 1960). Despite being widely cited in subsequent research on insect abundance, Williams’ estimate is based on a 1935 sample of soil insects^[4] in the United Kingdom, and was published in 1960 as more of an exercise in population modeling than a rigorous estimate. However, William’s estimate falls within an order of magnitude of the results of a recent model (Bar-On et al. 2018). Bar-On et al. used two parallel approaches to estimate the global biomass of terrestrial arthropods (which includes but is not limited to insects). The first method extrapolates measurements of biomass densities to the global ice-free land area. The second combines Williams’ estimate with the average biomass of an individual arthropod (Box 1). These approaches result in global terrestrial arthropod biomass estimates of 0.4 and 0.1 Gt C, respectively (ibid. p. 40). Considering the average biomass of an individual terrestrial arthropod (0.1 mg C; ibid.), the biomass estimate obtained through the first approach corresponds to 4E+18 terrestrial arthropods globally (Box 1). Williams’ estimate of 1E+18 insects (a taxonomic subset of Bar-On et al.’s terrestrial arthropods) is consistent with this value. Of 7 million terrestrial arthropod species, an estimated 5.5 million are insects (Stork 2018). In the absence of data on the relative abundance of insects versus other arthropod groups, I take Bar-On et al.’s estimate of the number of terrestrial arthropods as being approximately equal to the number of insects. I use 1E+18 insects, a conservative rounding of their estimate, in my calculations below.

Box 1: Estimating global insect abundance

Biomass estimate 1: 2.7E-3 kg/m2 (average measured biomass density of terrestrial arthropods) x 1.3E+14 m2 (global ice-free land surface area) = 4E+11 kg (0.4 Gt C) global biomass of terrestrial arthropods (Bar-On et al. 2018)

Biomass estimate 2: 1E+18 insects (Williams 1960) x 1E-7 kg C (average biomass of individual terrestrial arthropod) = 1E+11 kg (0.1 Gt C) global biomass of terrestrial arthropods (Bar-On et al. 2018)

Abundance estimate 1: 1E+18 insects (Williams 1960, extrapolated from English soil samples)

Abundance estimate 2: 4E+11 kg (biomass estimate 1) (Bar-On et al.2018) ÷ 1E-7 kg C (average biomass of individual terrestrial arthropod) (Bar-On et al. 2018) = 4E18 terrestrial arthropods ≅ 1E18 insects (rounded down)

How many insects are affected by insecticides?

As of the last USDA Agricultural Census in 2017, 3.6E+12 m2 of land was in agricultural use, and 0.45E+12 m2 of that land was treated with insecticides and/or acaricides^[5] (USDA NASS 2019a). The majority of agricultural land was used as pasture (1.6E+12 m2) or for growing crops (1.6E+12 m2). The remainder was woodland (0.29E+12 m2) and other agricultural uses (0.11 E+12 m2)(ibid.). The number of insects per land-use type is a critical knowledge gap: the density of insects and other terrestrial arthropods is likely to be different between a corn field and a potato field, and even more different between cropland and woodland. Setting this uncertainty aside, considering the number of terrestrial arthropods on earth and the amount of US crop land, there are approximately 2.7E+16 individual terrestrial arthropods living on all U.S. cropland, and 0.35E+16 individuals are potentially affected by U.S. agricultural insecticide use in 2017 (Box 2).

Box 2: Estimating insect abundance on U.S. agricultural land

Density: 1E+18 insects globally ÷ 1.3E+14 m2 global ice-free land surface = 7.7E+3 insects/m2

Abundance on US agricultural land: 7.7E+3 insects/m2 × 3.6E+12 m2 agricultural land in the U.S. = 2.7E+16 insects on U.S. agricultural land

Abundance on insecticide-treated US agricultural land: 7.7E+3 insects/m2 × 4.5E+11 m2 U.S. agricultural land treated with insecticides = 0.35E+16 insects on insecticide-treated U.S. agricultural land

This approach to estimating the impact of an insecticides intervention relies on global generalizations about biomass and abundance across different species and landscapes. While it may be useful for grasping the approximate scale of the impact of agricultural insecticides on wild insects, it is too broad to be useful for prioritization between specific insecticide interventions (e.g. whether to focus on soy versus barley cultivation). One approach for generating a more accurate impact estimate would be to incorporate economic injury levels (EILs). An EIL estimate is a tool for determining when pesticide application is cost-effective (Hunt et al. 2009). It is expressed as the number of individual pests of a given species per plant, where the economic injury caused by the pest is equal to the cost of treating the crops with a given insecticide (ibid.). If the number of pest individuals per plant exceeds the EIL, it is cost-effective for the grower to apply insecticides (ibid.). Growers may actually apply insecticides before or after pest populations exceed the EIL (which would result in an over- or under-estimate of the number of insects affected, respectively), but if we assume they are choosing pest management practices that are both economically sound and in line with recommendations for reducing the risk of insecticide resistance, then EILs are an appropriate basis for impact estimates. This method does not incorporate non-target insects that may still be affected by the pest control program, so the estimate generated should be assumed to be the minimum number of affected individuals.

EILs have been published for some major pests of crops that are grown widely in the United States, and are quite specific to the location, pest species, and crop variety. Multiple insect species may be pests of the same crop, and the EIL would be different for each species. Below, I suggest a way in which EIL estimates may be used to approximate the number of insects impacted by a pest control program, based on the kinds of agricultural data that are normally available (Box 3). I have also outlined a table that may be used in conjunction with a review of pest control literature to generate impact estimates for particular agricultural insecticide interventions.

Box 3: Using EILs to estimate pest management impacts

Cultivated area (m2) × expected yield (kg/m2) ÷ average plant mass (kg/plant) = plant count

Plant count × ∑ EIL estimate for pest i (indiv./plant) = number of individuals affected across all targeted species

Insecticide Practices

Developing an effective intervention requires understanding current agricultural insecticide practices The use of agricultural insecticides is of economic, environmental, and human health concern; therefore, information about the amount and type of insecticides used in the United States is made publicly available by the Environmental Protection Agency (EPA) and the U.S. Department of Agriculture (USDA). The USDA produces reports following the U.S. Agricultural Census every five years which include data on the use and sale of pesticides in the United States. The reports are based on data collected by the U.S. Department of Agriculture National Agricultural Statistics Service (USDA NASS) and by private market research companies. The most recent EPA report was published in 2017 and covered market estimates from 2008 to 2012 (Atwood & Paisley-Jones 2017). The most recent USDA Agricultural Census was in 2017, and a report summarizing its findings was released earlier this year (USDA NASS 2019a). The USDA also provides an interactive database of all information from prior agricultural censuses since 1997.

Recent U.S. insecticides market

According to the EPA’s 2017 market report on the pesticides industry, U.S. insecticide use comprised 6-7% by mass and 14% by expenditures of the global insecticides market from 2008 to 2012 (Atwood & Paisley-Jones 2017). Between 2008 and 2012, agricultural applications accounted for 55-57% by mass of the U.S. insecticides market (3-4% of the global market) (ibid.). Home and garden applications were 23-25%, and professional^[6] applications were 17-20% by mass of total U.S. insecticide use (ibid.). The total mass of conventional (i.e. synthetic chemical) insecticides used in the U.S. has decreased by 39% between 2000 and 2012, from 4.5 E+7 kg (99 million lbs) to 2.7E+7 kg (60 million lbs) (Atwood & Paisley-Jones 2017). This excludes biological control techniques that are popular in the integrated pest management approach, such as chemical insecticides derived from microorganisms (e.g. Bt), the genetic modification of crop plants to produce insecticidal or insect-aversive compounds, and the use of “natural enemies” (predators, parasites, pathogens, and competitors) to reduce insect pest populations (U.S. EPA 2016, Landis & Orr 1996). Between 2008 and 2012, agricultural applications accounted for approximately 60% by mass on average of insecticides used (Section 3.2 in Atwood & Paisley-Jones 2017). Across all U.S. market sectors (agricultural, home and garden, and professional), chlorpyrifos and acephate were the most commonly used insecticides by mass of active ingredient used, at 2E+6 to 4E+6 kg (5-8 million lb) each in 2012. 2E+6 to 4E+6 kg (4-8 million lb) of chlorpyrifos and 1E+6 to 3E+6 kg (2-6 million lb) of acephate were used in agriculture in 2012 (Atwood & Paisley-Jones 2017).

Figure 1: Percentage of cropland treated with insecticides per county, 2017 (USDA NASS 2019b).

Table 1: Chemical pesticide use on U.S. agricultural lands, 1997-2017.

Data are from an inquiry of the USDA NASS Quick Stats database. “Other” refers to the domain “CHEMICAL, OTHER (TOTAL)” in the USDA NASS Quick Stats database, defined as chemicals used “to control growth, thin fruit, ripen, or defoliate” in the 2017 Census of Agriculture (USDA 2019a). Census respondents were not asked to specify what type of agricultural land they applied chemical pesticides to (Section 26 in Form 17-A100).

Table 2: Select metrics of U.S. agricultural land usage, 1997-2017.

Farm number and area data reproduced from Table 1: Historical Highlights in the 2017 Census of Agriculture (USDA NASS 2019a).

Table 3: Most popular insecticides in each U.S. market sector, by mass of active ingredient applied (2012).

Use rates and sales are presented as ranges in the USDA report, as the source data is proprietary information from private market research companies (Atwood & Paisley-Jones 2017).

Overview of the major insecticide modes of action

AChE inhibitors

Acetylcholinesterase (AChE) inhibitors are a kind of nerve and muscle agent. They are the most popular insecticide type in the U.S. by mass, both within the agricultural sector and across all market sectors. Organophosphate and carbamate insecticides operate through this mechanism (IRAC 2019a). AChE inhibitors prevent the enzyme AChE from breaking down acetylcholine, a neurotransmitter (Čolović et al. 2013). Acute exposure to organophosphates or carbamates therefore increases the bioavailability of acetylcholine, resulting in over-stimulation of the parasympathetic nervous system until the acetylcholine is depleted. This may manifest as pronounced muscle spasms before paralysis and death (Čolović et al. 2013, Fukuto 1990). Sublethal effects have been observed in some insect species for several organophosphate and carbamate insecticides: carbaryl and parathion affect foraging and long-range pheromonal courtship behavior, diazinon affects foraging and fecundity, aldicarb and malathion affect locomotion, and parathion-methyl induces avoidance of treated host plants (Haynes 1988).

Organophosphates comprise a greater portion of the U.S. insecticides market than any other single chemical class. Of the 3E+7 kg (60 million lb) of all insecticidal active ingredients used in all US markets in 2012, 1E+7 kg (20 million lb) were organophosphates (Atwood & Paisley-Jones 2017). The two most commonly used insecticides by mass between 2008 and 2012, chlorpyrifos and acephate, are both organophosphate insecticides (ibid.). However, organophosphates have decreased in popularity in the past two decades: in 2000, 71% of the mass of all insecticidal active ingredients applied in the United States was organophosphates; in 2012, it was only 33% (ibid.). Other chemical classes, such as pyrethroids and pyrethrins, and neonicotinoids, have seen increased usage in this time period, although insecticide usage by mass has decreased by 39% overall (ibid.).

Because of their past widespread and heavy use, organophosphate insecticides have been a particular focus for EPA review, and this decrease in their popularity is likely attributable to the implementation of stricter regulations for organophosphate use. One example is chlorpyrifos, an organophosphate insecticide that was formerly sold in both the agricultural and home markets. In 2000, most forms of chlorpyrifos were banned from sale for home and garden use, and chlorpyrifos can now only be applied by certified pesticide applicators (Vogel 2016, U.S. EPA 2002). The sale and possession of chlorpyrifos for any use, including agricultural, is currently or will soon be banned in Hawaii, New York, and California (Rogers 2019). Chlorpyrifos is still widely used in agricultural contexts, however, and in 2012 more chlorpyrifos was applied both across all markets and in the agricultural market than any other insecticide (Atwood & Paisley-Jones 2017). Acephate, another organophosphate, ranked second (ibid.).

In addition to increased regulatory oversight, another factor contributing to decreased organophosphate use may be the success of the Boll Weevil Eradication Program, a USDA integrated pest management program which has reduced reliance on the major organophosphate insecticide malathion for boll weevil control (USDA APHIS 2006). In 2005, 5E+6 to 6E+6 kg (10-13 million lb) of malathion were applied across all market sectors—as much as twice the mass of the next most common insecticide that year, chlorpyrifos. By 2012, malathion usage had declined to only 0.5E+6 to 2E+6 kg (1-4 million lb) applied (Atwood & Paisley-Jones 2017).

Table 4: Most commonly used organophosphate insecticides across all market sectors, ranked by range in millions of pounds of active ingredient.

Reproduced from Table 3.8 in Atwood & Paisley-Jones (2017). A dash (-) indicates that the organophosphate pesticide was not one of the 10 most commonly used in the given year.

Sodium-channel modulators

Another group of nerve and muscle agents is the sodium-channel modulators, which include DDT, methoxychlor, and the pyrethroids and pyrethrins. Similarly to AChE inhibitors, these compounds cause nerve hyperexcitation, although the mechanism differs: sodium-channel modulators alter the concentration of sodium ions to affect action potentials (US EPA 2009). As with AChE inhibitors, acute toxicity from sodium-channel modulators results in muscle spasms and paralysis, which eventually leads to death (Toynton et al. 2009, US EPA 2009). The two most commonly-used sodium-channel modulators are permethrin and bifenthrin, both belonging to the class pyrethroids and pyrethrins. In insects, pyrethroids may cause hyperexcitation of the central nervous system and convulsions, or ataxia and incoordination prior to death (Bloomquist 2015, Wolansky & Harrill 2007). Their use is mostly confined to the home and garden markets, where they are used for mosquito, flea, and tick control. Sodium-channel modulators are sometimes used in crop cultivation, but not commonly, due to their toxicity to aquatic organisms (US EPA 2009). Other sodium-channel modulators include oxadiazines and semicarbazones. These compounds cause central nervous system depression resulting in ataxia and paralysis in insects, and convulsions when the unmoving insect is disturbed (Bloomquist 2015). Sublethal effects on insects of pyrethroids include reduced feeding, hyperactivity, and avoidance of pesticide residues (Bradbury & Coats 1989). In particular, sublethal toxicity from permethrin alters foraging behavior, impedes long-range pheromonal courtship behaviors and inhibits the flight reflex (ibid.).

Insect growth regulators

Some insecticides (such as fenoxycarb, pyriproxyfen, and etoxazole) disrupt the growth processes of larval insects by preventing metamorphosis, stopping the biosynthesis of necessary proteins (such as chitin, for forming the exoskeleton) and lipids, or delaying or accelerating moults (IRAC 2019a). Affected insects are usually killed by the resulting malformations before reaching the adult instar (Silva-Aguayo 2002). Insect growth regulators are not among the most commonly used insecticides in any sector (Atwood & Paisley-Jones 2017).

Integrated pest management

Integrated pest management (IPM) refers to a broad approach to controlling pest populations using a combination of biological, cultural, and chemical methods (U.S. EPA 2019). While some IPM methods, such as crop rotation, have been in practice in some form for centuries, it was not until the development of synthetic pesticides after World War II that IPM gained scientific attention (Smith & Smith 1949). A 1987 federal evaluation of the effectiveness of integrated pest management found that IPM practitioners were able to mitigate pest damage to crops and the resulting economic injury while reducing pesticide use (USDA 1989). Since the ratification of the 1972 Federal Environmental Pesticides Control Act, the USDA has undertaken a coordinated IPM initiative, providing recommendations and funding for IPM programs nationwide (Jacobsen 1996, USDA NIFA). The goal of IPM is not to eradicate pest populations, but rather to reduce the population size to a point where the pests do not have a significant detrimental economic or public health impact, while minimizing negative effects of pest control practices on ecosystems or human health. While some non-chemical methods employed in IPM may be more humane alternatives to conventional insecticides, their more complicated population and community effects make it harder to determine the net impact of the intervention on non-target species.

Biological control methods

IPM often includes the use of natural enemies (predators, parasitoids, and pathogens) to suppress pest populations (Shelton n.d.). This can involve the importation of novel enemies or the conservation or augmentation of existing ones (Landis & Orr 1996). Ideally, a natural enemy is specific enough to the target pest or pests that they have minimal impacts on desirable species, and either the pest is extirpated or the predator-prey relationship is stable enough to suppress pest populations long-term (Obrycki et al. 1996). Examples of natural enemies used for biological control include: lady beetles, a specialist predator whose larval and adult stages consume aphid pests (Obrycki et al. 1996); parasitoid wasps, which lay their eggs inside living larvae (and sometimes adults) of crop pests, later killing the host when the wasp larvae emerge (Landis & Orr 1996); and Bacillus thuringiensis (Bt), a bacterium which produces toxins that destroy insect larvae’s midguts, leading to death by starvation or infection (Perez et al. 2015, Shelton n.d.). Parasitoids and pathogens tend to take much longer to kill than predators or chemical insecticides, on the order of weeks compared to hours (Shelton n.d., IRAC 2019b).

Some biological control methods do not involve killing pests directly or intentionally manipulating their natural enemies, but instead alter their reproductive success. One such autocidal technique is the Sterile Insect Release Method (SIRM). Lab-reared individuals of the pest species are made sterile (usually chemically or with radiation) and released into the pest population. Sterile individuals compete with fertile individuals for mating opportunities, which causes the population to decline in subsequent generations (Bartlett & Staten 1996). Related techniques are encouraging hybridization that results in sterile offspring, and introducing mutations that alter sex ratio or change the number of reproductive cycles in an insect’s lifetime (ibid.). Individuals in populations managed exclusively with autocidal methods still die of predation, starvation, exposure, or disease, but after several generations there should be fewer individuals killed than a chemical method which targets adult insects who may already have reproduced.

Cultural control methods

Cultural control encompasses management techniques that operate on the production system rather than the pest. Like biological control methods, cultural control requires an understanding of the pest species’ life cycles, diets, reproductive strategies, trophic relationships, and migration and dispersal habits. Cultural control methods fall into one or more of four functional categories: impeding colonization by pests, making the environment inhospitable to pests, altering the crop to reduce pests’ impacts, and making the environment more hospitable to pests’ natural enemies. For example, physical barriers such as row covers or plastic-lined trenches can prevent colonization (Ferro 1996). Alternating between crops that do not have mutual pests, or shifting planting times so that pests who hatch or emerge from dormancy lack their preferred nutrition and refuge, creates an inhospitable environment for perennial pests (ibid.). Non-crop volunteer plants can be recruited to act as alternate hosts for pests, or to provide supplementary nutrition or refuge for natural enemies. Alternatively, some volunteer plants provide supplementary nutrition and refuge for pests, and eliminating them can reduce pest populations (ibid.). Crops can be genetically modified to be less susceptible to pest damage, either through selective breeding (such as developing cultivars which have a more durable epidermis or produce unpleasant-tasting leaves), or through direct genetic manipulation (such as inserting new genetic material so crops produce plant-incorporated protectants, like Bt toxins) (US EPA 2017, Perez et al. 2015, Teetes 1996). Wild relatives of crop species are a rich source for beneficial mutations to be used in genetic modification of crops; the conservation of genetic diversity is therefore important for developing pest-resistant cultivars (Eigenbrode 1996).

Chemical control methods

IPM encourages practitioners to be more efficient in their insecticide use and incorporate complementary techniques to reduce the total amount of insecticide applied. For instance, IPM practitioners may rotate through insecticides with different modes of action, which reduces the likelihood of pests developing resistance to any one mode of action (IRAC 2019c). To reduce the risk of resistance and minimize harm to non-target organisms, the use of broad-spectrum insecticides and insecticides which are toxic to fish, amphibians, birds, or mammals are discouraged. IPM practitioners prefer chemical control methods that are specific to the target pest, or at least to insects. For example, diatomaceous earth is lethal to arthropods and nematodes but non-toxic to other taxa. The amorphous silicon dioxide in diatomaceous earth abrades the insect digestive tract if ingested, and impedes respiration by blocking the spiracles and trachea. The primary insecticidal mechanism of diatomaceous earth is dessication: silicon dioxide destroys the protective cuticle on an insects’ exoskeleton that prevents water loss (Faulde et al. 2006). Apart from mild irritation from the abrasive particles in diatomaceous earth, silicon dioxide seems to be harmless to non-arthropods (Bunch et al. 2013).

Non-insecticidal chemical control methods are antifeedants, repellents, and confusants. These compounds are usually botanical in origin. Certain plant volatiles deter pest insects at a distance because they are noxious to pests or attractive to their natural enemies; pyrethroids (including pyrethrin and permethrin), which are neurotoxic insecticides derived from chrysanthemums, act as repellents at lower concentrations. Other volatiles may attract pests, and can be applied to nearby volunteer plants to confuse herbivorous insects that locate food by chemosensation (Silva-Aguayo 2002). Synthetic insect pheromones can be used to disrupt mating by inundating the chemical landscape with false signals and overwhelming genuine ones, or to bait lethal traps (the method of killing in pheromone-baited traps is usually contact with an insecticide, or sometimes drowning in standing water) (Flint & Doane 1996). Some plants produce noxious-tasting or toxic compounds which deter pests from feeding on that plant. Some of these antifeedants are insecticidal, such as terpenes (isolated from conifer resin), which cause the insect to cease feeding altogether and starve to death (Silva-Aguayo 2002).

Considerations and Conclusions

There are several factors to consider when comparing the expected welfare impacts of different insect pest management techniques. Broadly, they are: the intensity of the negative experiences caused by a particular method, the duration of those experiences, and how many individuals are affected. A more humane insect pest management method would be less painful, faster, or affect fewer individuals than the original method. It is important to bear in mind that the individuals under consideration will die by some means, whether or not their deaths are the direct result of a pest management program. Insects on agricultural land may die of starvation, desiccation, freezing, drowning, disease, parasitism, predation, insecticide exposure, or trauma from harvesting and transport processes. It is unknown how common, how painful, or how quick each of these causes of death are. My comparisons between the insect pest management methods covered in this report are therefore speculatory and I would readily change them with new information.

Comparing insect pest management methods

In theory, most agricultural pest management programs aim to eradicate the pest population from the managed area, as doing so would eliminate economic injury to the crop from pest activity. In practice, environmental and economic constraints mean that extirpation is rarely achieved. Pest insect populations may only be suppressed and later rebound, or if the population is entirely eliminated then the managed area may be recolonized. Managing pest insect populations in this way might result in many more insects being killed than if they were successfully extirpated from managed areas and prevented from recolonizing, as ongoing management would be required to reduce economic injury by successive generations of pest insects.

The disvalue of killing more insects in total over multiple generations must also be balanced with any negative repercussions that could arise from eradicating the population entirely from the managed area. If the harm caused by eradication would outweigh the harm done to future generations of insects, then a more humane insect pest management program would avoid manipulating the number of insects killed directly by a given method, and instead try to shorten or lessen the painful experience it causes. This approach may help skirt the issue of introducing new ripple effects in the managed area and its ecological context, as it minimizes changes to insect population size and existing trophic dynamics.

If a death by other causes (e.g. predation) is expected to be better than a death by exposure to an insecticide, then it would be more humane to use non-lethal pest insect management methods (e.g. applying sublethal concentrations of insecticidal compounds as repellants, installing physical barriers, providing alternate host plants, or genetically modifying crops to be unpalatable to pests). If, however, death by other causes (e.g. parasitism or predation, which may or may not be part of a pest management program) is expected to be worse than death by insecticide, then the most humane management strategy would be to apply the fastest-acting and least-painful insecticide. Among the major groups of insecticides reviewed in this report, I expect the nerve and muscle agents (AChE inhibitors and sodium-channel modulators) to be preferable to insect growth regulators, or insecticides acting on the midgut (namely, Bt toxins). In absence of information about how painful these respective methods might be, it would be better to use the faster-acting nerve and muscle agents (which kill over minutes to hours) than the slower-acting growth regulators or midgut disruptors (which kill over days to weeks) (Shelton n.d., IRAC 2019b).

If eradicating a pest insect population is unlikely to have outsized negative consequences for other individuals, then the more humane pest management option may be to prevent subsequent generations of that population from coming into existence. While the duration and intensity of the negative experiences involved in a particular cause of death are still relevant to this approach, the primary focus is on reducing the number of individuals affected over the long term. The most humane method used to prevent future generations of a pest insect population will depend on the expected painfulness of death caused directly by pest management versus the death expected otherwise (including indirect effects of pest management, as well as causes independent of management). Extirpation of a pest insect could be achieved by combining physical barriers (to prevent the population being augmented through immigration) and either non-lethal methods for reducing reproductive rate, or lethal methods targeting pre-reproductive individuals. Techniques for non-lethal population control include inhibiting mating signals or releasing sterile individuals into the population (SIRM). Some insecticides or natural enemies primarily affect juvenile insects (such as insect growth regulators, or certain parasitoid wasps), killing them before they reach reproductive maturity. It may be that larval insects are less likely to be sentient than adults, in which case it would be better to reduce reproductive rates by targeting juvenile insects than by killing adults who have not mated yet.

In summary, the relevant considerations for determining the welfare impact of a particular pest insect management program are: the ultimate cause of death, its duration and painfulness, the number of individuals affected immediately by the program, the number of individuals expected to be subject to this program in the future, the life stage affected and the expectation of sentience at that stage, and non-target effects on other individuals in the agroecosystem and beyond. Using nonlethal cultural, physical, or chemical control or a faster-acting lethal method (such as nerve and muscle agents, or potentially predation) is probably a higher-welfare option than using slower-acting insecticides (e.g. chitin-synthesis disruptors or Bt toxins) or some biological control methods (pathogens and parasitoids).

Recommendations for future research

There are many questions that I was unable to answer when researching the content of this report. The following stood out as especially important for any further progress on a pest management intervention to improve insect welfare:

1. What symptoms do insects show when exposed to lethal and nonlethal doses of particular insecticides? How long does it take an insect to die after being exposed to a given insecticide?
2. How common are each of the causes of death experienced by insects in agroecosystems not under any kind of pest management program? How painful and how long are those experiences expected to be?
3. What trophic relationships do pest insects have in agroecosystems? Would eradicating a particular population have negative ramifications for other individuals?
4. What evidence do we have for sentience in the kinds of insects affected by agricultural pest management, which are different species than the insects that are normally studied?
5. At what developmental stage do features indicative of sentience appear in the relevant insect species?
6. How do different pest management methods affect the reproductive rates of pest insect populations? Does preventing colonization of an agroecosystem by a pest insect affect the growth of that insect’s population outside the management area?
7. What is the average density of insects and other terrestrial arthropods on different types of agricultural land? On what types of agricultural land are particular insecticides used?
8. What amount of the less-common insecticides not covered in the EPA report are applied annually?
9. What are the economic injury levels (EILs) for particular crop-pest pairs? Can this information be used to generate estimates of number of insects affected by pest management?
10. On which insect species and crops are particular insecticides or pest management methods used?

Questions 1 and 2 seem important to address first, as they play a significant role in determining the valence of pest management methods. I encountered very little research examining the symptoms of lethal toxicity or time until death in insects exposed to different insecticides. A more intensive and focused literature review may turn up answers, suggest experts to contact for anecdotal evidence of symptoms, or determine that there is a gap in the literature that should be addressed in future studies. A crucial consideration for a pest management intervention is the painfulness and duration of the average death of agricultural pest insects in the absence of any management. Knowing how likely and how bad a death by a cause such as disease, dehydration, predation, or starvation is could make the difference between advocating for a lethal or non-lethal management program.

Questions 3 through 6 are important for determining the best life stage to target with a pest management program, both in terms of the welfare of the insects killed, and in terms of how many individuals are affected over the entire management period. I feel there is enough evidence of sentience in some insects to make it worthwhile to consider the potential welfare impacts of insecticides and other pest insect management techniques; however, research into question 4 would be useful as a means of continually evaluating the value of insect welfare interventions, in agroecosystems and in general.

Questions 7 through 10 are intended to help prioritize the regions, crops, and insecticidal compounds or other pest management methods for intervention.

Conclusion

1. The evidence supporting insect sentience is sufficient to argue that, in combination with the large number of insects, we should afford some consideration to insect welfare.
2. Agricultural pest insect management practices may be a particularly tractable avenue for improving the expected welfare of a large number of insects.
3. Chemical insecticides remain a major part of many agricultural pest management programs. Their effects vary depending on the insecticide’s mode of action, and I did not find much information about the symptoms of toxicity in insects from particular compounds.
4. Non-insecticidal methods of pest control may be gaining popularity, and should be considered when developing more humane pest insect management practices. The net welfare impact of non-lethal methods is dependent on ecological factors and the specific organisms involved.
5. Key unknowns make it difficult to recommend a particular insecticide or non-insecticidal pest control method as more humane. However, nerve and muscle agents (such as organophosphates and carbamates, or pyrethroids and pyrethrins) are faster-acting than insecticides with other modes of action (such as insect growth regulators, or Bt toxins).

Acknowledgments

I thank Cameron Meyer Shorb, Michelle Graham, and Abraham Rowe for their guidance and support throughout this research project. I also thank Jane Capozzelli, Simon Eckerström Liedholm, and Luke Hecht for their insightful advice and valuable feedback during the research and writing processes.

References

Adamo, S. A. (2017). The “Precautionary Principle”: A work in progress. Animal Sentience. 16(4). https://animalstudiesrepository.org/cgi/viewcontent.cgi?article=1222&context=animsent.

Adamo, S. A. (2016). Subjective experience in insects: Definitions and other difficulties. Animal Sentience. 9(15):1-4. https://animalstudiesrepository.org/cgi/viewcontent.cgi?article=1153&context=animsent.

Allen, C. and Trestman, M. (2017) Animal Consciousness. In: E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2017 Edition). https://stanford.library.sydney.edu.au/index.html. Retrieved 10 October 2019.

https://plato.stanford.edu/archives/win2017/entries/consciousness-animal/.

Atwood, D. & Paisley-Jones, C. (2017). Pesticides industry sales and usage: 2008-2012 market estimates. U.S. EPA. https://www.epa.gov/sites/production/files/2017-01/documents/pesticides-industry-sales-usage-2016_0.pdf.

AVMA. (2013). AVMA guidelines for the euthanasia of animals: 2013 edition. https://www.avma.org/KB/Policies/Documents/euthanasia.pdf.

Bar-On, Y. M., Phillips, R., & Milo, R. (2018). The biomass distribution on Earth. Supplementary information appendix. PNAS. 115(25):6506-6511. https://www.pnas.org/content/pnas/suppl/2018/07/13/1711842115.DC1/1711842115.sapp.pdf.

Barron, A. B. & Klein, C. (2016). What insects can tell us about the origins of consciousness. PNAS. 113(18):4900-4908. https://www.pnas.org/content/pnas/113/18/4900.full.pdf.

Bartlett, A. C. & Staten, R. T. (1996). The Sterile Insect Release Method and other genetic control strategies. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 5 August 2019. https://ipmworld.umn.edu/bartlett.

Birch, J. (2017). Animal sentience and the precautionary principle. Animal Sentience. 16(1). https://animalstudiesrepository.org/cgi/viewcontent.cgi?referer=http://animalstudiesrepository.org/&httpsredir=1&article=1200&context=animsent.

Bloomquist, J. R. (2015). Insecticides: Chemistries and characteristics. 2nd edition. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 31 October 2019. https://ipmworld.umn.edu/bloomquist-insecticides.

Bradbury, S. P. & Coats, J. R. (1989). Comparative toxicology of the pyrethroid insecticides. Reviews of Environmental Contamination and Toxicology. 108:133-177. https://www.researchgate.net/profile/Joel_Coats/publication/20508882_Comparative_Toxicology_of_the_Pyrethroid_Insecticides/links/55a9190a08aea3d086802d28.pdf.

Broom, D. M. (2007). Cognitive ability and sentience: Which aquatic animals should be protected? Diseases of Aquatic Organisms. 75:99-108. https://www.int-res.com/articles/dao_oa/d075p099.pdf.

Bunch, T. R., Bond, C., Buhl, K. & Stone, D. (2013). Diatomaceous earth general fact sheet. National Pesticide Information Center, Oregon State University Extension Services. Retrieved 15 August 2019. http://www.npic.orst.edu/factsheets/degen.html.

Čolović, M. B., Krstić, D. Z., Lazarević-Pašti, T. D., Bondžić, A. M., & Vasić, V. M. (2013). Acetylcholinesterase inhibitors: Pharmacology and toxicology. Current Neuropharmacology. 11(3):315-335. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3648782/pdf/CN-11-315.pdf.

Dawkins, M. S. (2008). The science of animal suffering. Ethology. 114(10):937-945. https://doi.org/10.1111/j.1439-0310.2008.01557.x.

Eigenbrode S. D. (1996). Host plant resistance and conservation of genetic diversity. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 15 August 2019. https://ipmworld.umn.edu/eigenbrode.

Elwood, R. W., Barr, S., & Patterson, L. (2009). Pain and stress in crustaceans? Applied Animal Behaviour Science. 118(3):128-136. https://www.sciencedirect.com/science/article/pii/S0168159109000409#aep-bibliography-id22.

Faulde, M. K., Tisch, M. & Scharninghausen, J. J. (2006). Efficacy of modified diatomaceous earth on different cockroach species (Orthoptera, Blatellidae) and silverfish (Thysanura, Lepismatidae). Journal of Pesticide Science. 79:155-161. https://link.springer.com/content/pdf/10.1007%2Fs10340-006-0127-8.pdf.

Ferro, D. N. (1996). Cultural control. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 7 August 2019. https://ipmworld.umn.edu/ferro.

Flint, H. M. & Doane, C. C. (1996). Understanding semiochemicals with emphasis on insect sex pheromones in Integrated Pest Management programs. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 15 August 2019. https://ipmworld.umn.edu/flint.

Fox, D. (2007). Consciousness in a cockroach: The insect nervous system provides clues to attention, consciousness, and the origin of the brain. Discover Magazine. January 2007. Retrieved 4 September 2019. http://discovermagazine.com/2007/jan/cockroach-consciousness-neuron-similarity.

Fukuto, T. R. (1990). Mechanism of action of organophosphorus and carbamate insecticides. Environmental Health Perspectives. 87:245-254. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1567830/pdf/envhper00420-0231.pdf.

Giurfa, M. (2007). Behavioral and neural analysis of associative learning in the honeybee: a taste from the magic well. Journal of Comparative Physiology. 193(8):801-824. https://link.springer.com/article/10.1007/s00359-007-0235-9.

Greenspan, R. J. & van Swinderen, B. (2004). Cognitive consonance: Complex brain function in the fruit fly and its relatives. TRENDS in Neurosciences. 27(12):707-711. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.304.3883&rep=rep1&type=pdf.

Haberkern, H. & Jayaraman, V. (2016). Studying small brains to understand the building blocks of cognition. Current Opinion in Neurobiology. 37:59-65. https://www.sciencedirect.com/science/article/pii/S0959438816000088#sec0010.

Haynes, K. F. (1988). Sublethal effects of neurotoxic insecticides on insect behavior. Annual Review of Entomology. 33:149-168. https://www.researchgate.net/profile/Kenneth_Haynes/publication/19877837_Sublethal_Effects_Of_Neurotoxic_Insecticides_On_Insect_Behavior/links/0046351e92dbce78c0000000.pdf.

Herculano-Houzel, S., Mota, B., & Lent, R. (2006). Cellular scaling rules for rodent brains. PNAS. 103(32):12138-12143. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1567708/.

Hunt, T. E., Wright, R. J., & Hein, G. L. (2009). Economic thresholds for today’s commodity values. Proceedings of the UNL Crop Production Clinics (2009). 93-96. https://cropwatch.unl.edu/documents/Economic%20Thresholds.pdf.

IASP. (2017). IASP Terminology. Retrieved 9 December 2019. https://www.iasp-pain.org/Education/Content.aspx?ItemNumber=1698#Nociception.

IRAC. (2019a). IRAC Mode of Action Classification Scheme. Version 9.3. IRAC International MoA Working Group. Retrieved 17 June 2019. https://www.irac-online.org/modes-of-action/.

IRAC. (2019b). Test methods. IRAC Methods Team. Retrieved 7 August 2019. https://www.irac-online.org/methods/.

IRAC. (2019c). Resistance management. IRAC Outreach Team. Retrieved 15 August 2019. https://www.irac-online.org/about/resistance/management/.

Jacobsen, B. (1996). USDA integrated pest management initiative. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 1 July 2019. https://ipmworld.umn.edu/jacobsen-usda.

Klein, C. & Barron, A. B. (2016a). Insects have the capacity for subjective experience. Animal Sentience. 9(1). https://animalstudiesrepository.org/cgi/viewcontent.cgi?referer=http://animalstudiesrepository.org/&httpsredir=1&article=1113&context=animsent.

Klein, C. & Barron, A. B. (2016b). Reply to Adamo, Key et al., and Schilling and Cruse: Crawling around the hard problem of consciousness. PNAS. 113(27):E3814-E3815. https://www.pnas.org/content/113/27/E3814.

Knutsson, S. & Munthe, C. (2017). A virtue of precaution regarding the moral status of animals with uncertain sentience. Journal of Agricultural and Environmental Ethics. 30(2):213-224. https://www.researchgate.net/profile/Christian_Munthe/publication/315836483_A_Virtue_of_Precaution_Regarding_the_Moral_Status_of_Animals_with_Uncertain_Sentience/links/59e76ae84585152d5f04edd2/A-Virtue-of-Precaution-Regarding-the-Moral-Status-of-Animals-with-Uncertain-Sentience.pdf.

Koch, C., Massimini, M., Boly, M., & Tononi, G. (2016). Neural correlates of consciousness: progress and problems. Nature Reviews Neuroscience. 17:307-321. http://www.crm.cat/en/Activities/Curs_2017-2018/Documents/Consciousness1.pdf.

Kollmann, M., Huetteroth, M., Schachtner, J. (2011). Brain organization in Collembola (springtails). Arthropod Structure and Development. 40(4):304-316. http://kops.uni-konstanz.de/bitstream/handle/123456789/28911/Kollmann_289118.pdf?sequence=2&isAllowed=y.

Landis, D. A. & Orr, D. B. (1996). Biological control: approaches and application. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 1 July 2019. https://ipmworld.umn.edu/landis.

Mallatt, J. & Feinberg, T. E. (2016). Insect consciousness: Fine-tuning the hypothesis. Animal Sentience. 9(10). https://animalstudiesrepository.org/cgi/viewcontent.cgi?article=1141&context=animsent.

Mellor, D. J. (2016). Updating animal welfare thinking: moving beyond the “five freedoms” towards “a life worth living.” Animals. 6(3):21. https://doi.org/10.3390/ani6030021.

Merker, B. (2006). Consciousness without a cerebral cortex: A challenge for neuroscience and medicine. Behavioral and Brain Sciences. 30(1):63-81. https://pdfs.semanticscholar.org/3db9/c2f8ca0bc9d73532f2e950a1ecfe7063064d.pdf.

Morand-Ferron, J., Cole, E. F., & Quinn, J. L. Studying the evolutionary ecology of cognition in the wild: a review of practical and conceptual challenges. Biological Reviews. 2015:1-25. https://www.researchgate.net/profile/John_Quinn13/publication/271538545_Studying_the_evolutionary_ecology_of_cognition_in_the_wild_A_review_of_practical_and_conceptual_challenges/links/5b72ca87299bf14c6da1efdc/Studying-the-evolutionary-ecology-of-cognition-in-the-wild-A-review-of-practical-and-conceptual-challenges.pdf.

Obrycki, J. J., Tauber, M. J., Tauber, C. A., & Ruberson, J. R. (1996). Prey specialization in insect predators. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 2 July 2019. https://ipmworld.umn.edu/obrycki-tauber-ruberson.

Perez, J., Bond, C., Buhl, K. & Stone, D. (2015). Bacillus thuringiensis (Bt) General Fact Sheet. National Pesticide Information Center, Oregon State University Extension Services. Retrieved 7 August 2019. http://npic.orst.edu/factsheets/BTgen.pdf.

Pfeiffer, K. & Homberg, U. (2014). Organization and functional roles of the central complex in the insect brain. Annual Review of Entomology. 59:165-184. https://www.annualreviews.org/doi/full/10.1146/annurev-ento-011613-162031?casa_token=Zk4_WBrBRgIAAAAA%3AWnI9dNv8oMWmvHDR3gX4u6jTCpTYutEWIPw8219K1e76sslW17cWaiaK4XfCx6w4BT_Pa9LTryp8#_i20.

Ray, G. (2017a). How many wild animals are there? Wild Animal Suffering Research Archives. Retrieved 6 May 2019. https://was-research.org/writing-by-others/many-wild-animals/.

Ray, G. (2017b). Which invertebrate species feel pain? Wild Animal Suffering Research Archives. Retrieved 6 May 2019. https://was-research.org/writing-by-others/invertebrate-species-feel-pain/#Insects.

Ray, G. (2018). Invertebrate sentience: Urgent but understudied. Wild Animal Suffering Research Archives. Retrieved 6 May 2019. https://was-research.org/paper/invertebrate-sentience-urgent-understudied/.

Rogers, J. (2019). California bans pesticide linked to brain damage in children. Associated Press. 9 October 2019. www.apnews.com. Retrieved 31 October 2019. https://apnews.com/2b433b664d834f3faa4e39ff257cabb6.

Schukraft, J. (2019). Invertebrate sentience: A useful empirical resource. Effective Altruism Forum. https://forum.effectivealtruism.org. Retrieved 22 August 2019. https://forum.effectivealtruism.org/posts/AMFuSWrsutBFraZtE/invertebrate-sentience-a-useful-empirical-resource.

Shelton, A. Biological Control: A guide to natural enemies in North America. Cornell University Department of Entomology. Retrieved 7 August 2019. https://biocontrol.entomology.cornell.edu/index.php.

Silva-Aguayo, G. (2002). Botanical insecticides. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 15 August 2019. https://ipmworld.umn.edu/silva-aguayo-botanical.

Smith, R. F. & Smith, G. L. (1949). Supervised control of insects utilizes parasites and predators and makes chemical control more efficient. California Agriculture. 3(5):3-12. http://calag.ucanr.edu/archive/?type=pdf&article=ca.v003n05p3.

Sneddon, L. U., Elwood, R. W., Adamo, S. A., Leach, M. C. (2014). Defining and assessing animal pain. Animal Behaviour. 97:201-212. https://animalstudiesrepository.org/cgi/viewcontent.cgi?referer=http://animalstudiesrepository.org/&httpsredir=1&article=1068&context=acwp_arte.

Strausfeld, N. J. & Hirth, F. (2013). Deep homology of arthropod central complex and vertebrate basal ganglia. Science. 340(6129):157-161. https://science.sciencemag.org/content/340/6129/157.abstract.

Stork, N. E. (2018). How many species of insects and other terrestrial arthropods are there on Earth? Annual Review of Entomology. 63:31-45. https://doi.org/10.1146/annurev-ento-020117-043348

http://ecologylabs.org/wp-content/uploads/2018/03/Estimate-of-insect-species.pdf.

Teetes, G. L. (1996). Plant resistance to insects: A fundamental component of IPM. In: E. B. Radcliffe, W. D. Hutchison, & R. E. Cancelado [eds.], Radcliffe's IPM World Textbook. https://ipmworld.umn.edu. Retrieved 15 August 2019. https://ipmworld.umn.edu/teetes.

Tebbich, S., Griffin, A. S., Peschl, M. F., & Sterelny, K. (2016). From mechanisms to function: an integrated framework of animal innovation. Philosophical Transactions of the Royal Society B. 371(1690):1-11. https://royalsocietypublishing.org/doi/10.1098/rstb.2015.0195.

Tomasik, B. (2017). Which stimuli are painful to invertebrates? Essays on Reducing Suffering. https://reducing-suffering.org. Updated 30 July 2018. Retrieved 22 August 2019. https://reducing-suffering.org/which-stimuli-are-painful-to-invertebrates/.

Tomasik, B. (2009). Do bugs feel pain? Essays on Reducing Suffering. https://reducing-suffering.org. Updated 28 July 2017. Retrieved 22 August 2019. https://reducing-suffering.org/do-bugs-feel-pain/.

Tomasik, B. (2007). Humane Insecticides. Essays on Reducing Suffering. https://reducing-suffering.org. Updated 12 November 2018. Retrieved 26 November 2019. https://reducing-suffering.org/humane-insecticides/.

Toynton, K., Luukinen, B., Buhl, K., & Stone, D. (2009). Permethrin General Fact Sheet. National Pesticide Information Center, Oregon State University Extension Services. Retrieved 25 July 2019. http://npic.orst.edu/factsheets/PermGen.pdf.

USDA APHIS. (2006). Boll Weevil Eradication Factsheet. Plant Protection and Quarantine. Retrieved 24 July 2019. https://web.archive.org/web/20070204143907/http://www.aphis.usda.gov/publications/plant_health/content/printable_version/fsbollweevil.pdf.

USDA NASS. (2019a). United States: Summary and state data. 2017 Census of Agriculture. Volume 1: Geographic Area Series; part 51. https://www.nass.usda.gov/Publications/AgCensus/2017/Full_Report/Volume_1,_Chapter_1_US/usv1.pdf.

USDA NASS. (2019b). Ag Census Web Maps. 2017 Census of Agriculture. Retrieved 29 October 2019. https://www.nass.usda.gov/Publications/AgCensus/2017/Online_Resources/Ag_Census_Web_Maps/index.php.

USDA NIFA. Integrated pest management program (IPM). Retrieved 2 July 2019. https://nifa.usda.gov/program/integrated-pest-management-program-ipm.

U.S. EPA. (2019). Integrated pest management (IPM) principles. Retrieved 2 July 2019. https://www.epa.gov/safepestcontrol/integrated-pest-management-ipm-principles.

U.S. EPA. (2017). Insect resistance management for Bt plant-incorporated protectants. Retrieved 15 August 2019. https://www.epa.gov/regulation-biotechnology-under-tsca-and-fifra/insect-resistance-management-bt-plant-incorporated.

U.S. EPA. (2009). Permethrin facts: Reregistration eligibility decision (RED) fact sheet. Retrieved 25 July 2019. https://www3.epa.gov/pesticides/chem_search/reg_actions/reregistration/fs_PC-109701_1-Aug-09.pdf.

U.S. EPA. (2002). Chlorpyrifos facts. Retrieved 10 May 2019. https://www3.epa.gov/pesticides/chem_search/reg_actions/reregistration/fs_PC-059101_1-Feb-02.pdf.

U.S. EPA. (2016). What are biopesticides? Retrieved 1 July 2019. https://www.epa.gov/ingredients-used-pesticide-products/what-are-biopesticides.

USGS. (2017). New map of worldwide croplands supports food and water security. Retrieved 9 December 2019. https://www.usgs.gov/news/new-map-worldwide-croplands-supports-food-and-water-security.

Vogel, D. (2016). Chlorpyrifos status update. EPA Pesticide Program Dialogue Committee. Retrieved 10 May 2019. https://www.epa.gov/sites/production/files/2016-05/documents/session-2-chlorpyrifos-human-health-risk-assessment-approach.pdf.

Waldhorn, D. (2019). Invertebrate sentience: Summary of findings, part 2. Effective Altruism Forum. https://forum.effectivealtruism.org. Retrieved 22 August 2019. https://forum.effectivealtruism.org/posts/JtmiKxwCL7CQ4iL4p/invertebrate-sentience-summary-of-findings-part-2-1.

Williams, C. B. (1960). The range and pattern of insect abundance. The American Naturalist. 94(875):137-151. https://www.jstor.org/stable/pdf/2458376.pdf?casa_token=vMq0wVTqe10AAAAA:YbMo1x6UOypH9Qu_6FhGlNw4ndtK_WEY0oA4bJvrNJ_RYrJqbV24Wyi7qKFfaIz0aeTnv4wT9PLHQi_C99TOyAHobAKwPVzg4DuBo2n9cuGThLE7o1A.

Wolansky, M. J. & Harrill, J. A. (2007). Neurobehavioral toxicology of pyrethroid insecticides in adult animals: a critical review. Neurotoxicology and Tetralogy. 30(2):55-78. http://www.latomeq.qb.fcen.uba.ar/imagenes/Review_Pyrethroids_Corrected_Proof.pdf.

Notes

Last year, the year before, the year before that, and the year before that, we published a set of suggestions for individual donors looking for organizations to support. This year, we are repeating the practice and publishing updated suggestions from Open Philanthropy program staff who chose to provide them.

The same caveats as in previous years apply:

• These are reasonably strong options in causes of interest, and shouldn’t be taken as outright recommendations (i.e., it isn’t necessarily the case that the person making the suggestion thinks they’re the best option available across all causes).
• In many cases, we find a funding gap we’d like to fill, and then we recommend filling the entire funding gap with a single grant. That doesn’t leave much scope for making a suggestion for individuals. The cases listed below, then, are the cases where, for one reason or another, we haven’t decided to recommend filling an organization’s full funding gap, and we believe it could make use of fairly arbitrary amounts of donations from individuals.
• Our explanations for why these are strong giving opportunities are very brief and informal, and we don’t expect individuals to be persuaded by them unless they put a lot of weight on the judgment of the person making the suggestion.

In addition, we’d add that these recommendations are made by the individual program officers or teams cited, and do not necessarily represent my (Holden’s) personal or Open Phil’s institutional “all things considered” view. Also, I just want to note that per our policy we’re no longer publishing all potentially relevant relationships.

Suggestions are alphabetical by cause (with some assorted and “meta” suggestions last).

Aaron's notes:

• To save some time on formatting, I'm not copying over the full list of suggestions; read them at the original post.
• The suggestion that most surprised me was Engineers Without Borders, a large and well-known charity that doesn't seem like a "typical" Open Phil suggestion (though it could be very cost-effective as far as I know).
• The suggestion that may be most surprising to someone who hasn't seen past Open Phil grants is California YIMBY ("yes in my backyard"), which supports changing California state policy to allow more housing construction. 
  • An excerpt of Alexander Berger's case for YIMBY (emphasis mine):
    • "California is large enough that state policy reforms could make a meaningful dent in the national problem, but state policy reforms are not necessarily vastly harder here than in other places. Between reducing rents and allowing more people to be able to move to or remain in high-wage areas, we roughly estimate the social value of each new home in coastal California to be in the low hundreds of thousands of dollars, which means that even a very small improvement in the state’s housing policies could deliver a high return."
       

I'm having a hard time finding any EA-focused information about banks. This seems to me like it should be very important for EA, perhaps even more so than charities. I'm not very tech-savvy, so I may just not be looking in the right place. Thanks.

Hello folks, I'm interested in giving over the long term and I've found some of the research (1) around patient philanthropy convincing. I'm interested in investing some money (say 10% of income) and putting this in something like the things suggested here (2). I've looked a bit at DAFs but the fees look quite high and I wonder if I could assemble something better myself. My donations would be to longtermist causes. Does anyone know of a good guide on setting up a simple long term fund that I could use specifically for donations in the future please? The ideal would be something as simple as EA funds where I can clearly track my donations and keep them separate from everything else. Thanks!

(1) https://philiptrammell.com/research

(2) https://80000hours.org/2015/10/common-investing-mistakes-in-the-effective-altruism-community/

Originally posted on the EA subreddit.

First, I will present a rough sketch for why genetic enhancement could be a plausible cause X. Then I will list some specific proposals for genetic interventions. I will conclude by responding to objections. If there is interest, I may write more posts on this topic.

Basic argument

There are two main ways to think about genetic enhancement as a potential EA cause area. One perspective is focused on improving short-term human welfare. While reducing “defects” (disabilities, depression, etc.) would be a major focus, this perspective could additionally encompass increasing the frequency of beneficial traits, such as longevity-promoting alleles of the FOXO3 gene. The key idea is that enhancement is performed to increase individual well-being.

The other way of thinking about genetic enhancement, and the one I prefer, takes a long-term view. Changes are made with the far-future of our society in mind. For instance, by drastically increasing IQs, we could put our civilization in a better position to solve complex challenges that exist today or will arise down the line. Other interventions such as increasing empathy and decreasing Dark Triad personality traits could be used to influence the values of our descendents and avert future moral catastrophes.

The short-term case for genetic enhancement

If we want to improve short-term human well-being, we can group most possible interventions into two broad categories:

1. Improve the quality and duration of life for presently-existing humans.
2. Change which (and how many) humans will be born in the first place.

The first method is far less philosophically controversial. Everyone agrees that once you’re born, it’s better to live a happy life than a miserable one. On the other hand, when it comes to changing the number and identity of people in existence, we enter the muddy waters of population ethics, fraught with paradoxes and impossibility theorems. For the most basic version of my argument to work, it suffices to assume that when selecting among a fixed number of “potential children” who could be born, we should choose the ones with the highest expected well-being.

I have not yet said anything about genes specifically! Of course, both environment and genetics could be factors when evaluating the expected well-being of potential future humans. However, living environment in general is already subject to more optimization (see next paragraph). It also appears that many traits that contribute to a happy, successful life are highly heritable, including psychological traits (see appendix 1). Further, taking a slightly longer-term view, genetic information is directly passed on for multiple successive generations, while environment is ephemeral. I think a focus on genes is justified.

Going back to the dichotomy of altruistic interventions, the first option is the tack that most effective altruists interested in short-term human welfare have taken. It is also the strategy most popular among do-gooders in general. In comparison, little philanthropic effort is devoted to efforts to (e.g.) decrease the number of children born with severe congenital diseases or to increase the frequency of welfare-promoting alleles in the population. Aside from philanthropy, people in general seek out the best living environment for themselves and their children. On the other hand, although sexual selection does exist, people generally do not explicitly optimize the genetics of their progeny. So it’s clear that improving population genetics is comparatively neglected.

Further, I would argue that genetic enhancement is quite important. A meta-analysis of twin studies found that genetic factors explain 36% of variation in subjective well-being (appendix 1). While environmental conditions obviously play a huge role in well-being as well, the role of genes cannot be understated. If all EA optimization to date has only been targeting 64% of the problem, there might be a lot of low-hanging fruit in that remaining 36%.

It is harder to tell how tractable genetic enhancement is, which depends on the specific type of intervention. I may write a follow-up post exploring this issue in depth if there is sufficient interest.

The long-term case for genetic enhancement

The long-term trajectory of our civilization depends largely on how we navigate the minefield of emerging technologies that will be developed over the next couple of centuries. It likewise depends on the values that our descendents hold.

It seems that increasing IQ and other traits such as affective empathy (see appendix 1) could help humanity to reach a prosperous future. Imagine a world full of people as bright as John von Neumann and as ethical as Gandhi. Wouldn’t such a world be in a vastly better position to avoid existential risks and risks of future suffering than our present Earth?

Bostrom, and others, are fond of saying that we are the “stupidest possible biological species capable of starting a technological civilization.” In many ways, genetic enhancement would change this outlook. If the brightest humans on Earth one day become much wiser than we currently are, then I believe it is highly likely that we could improve the chances of successfully mitigating existential risk.

Consider the case of aligning an artificial intelligence. One model of AI development is that it is a constant race between capabilities gained by factors which provide virtually no safety guarantees, such as increasing hardware capacity, and capabilities gained by a principled understanding of AI design, such as the presentation of causality provided by Judea Pearl. MIRI believes that if we have a principled understanding of advanced AI design before we can build it via stupid means, like simulating evolution, then our chances of aligning such AI are increased dramatically.

In order to gain the upper hand in this race for technological capabilities, we could leverage genetic enhancement to maintain a favorable balance in differential technological development. For instance, progress in hardware is arguably bottlenecked by economic demand, and would not be significantly accelerated by the advent of a hundred John von Neumann level scientists. However, deep insights into the nature of intelligence are the type of thing we should expect if we have a highly competent core group of humans working on the problem.

Ensuring that this highly skilled group of people comes into existence would be much easier to accomplish than the widespread adoption of genetic enhancement implied by the short-term view I presented above. Therefore, it is likely more tractable. Still, selecting embryos for intelligence is highly controversial, and would not be something that most ethics panels would currently approve of. 

In the long term, I believe selecting embryos for favorable traits will happen anyway, regardless of ethical qualms, because once the technology has been demonstrated, countries unwilling to adopt it will risk falling far behind. EAs can therefore do research to track attitudes, and find the right time to begin implementing the strategy outlined above.

Interventions

I have explained why genetic enhancement is an area worth considering, but what can effective altruists actually do to advance the cause? I will now list a few promising, non-coercive forms of genetic enhancement. For each of the following items, EA work could focus on advocacy, research, or technical implementation.

I would highly recommend Gwern’s article on “Embryo selection for intelligence” for a detailed comparison of the feasibility and effectiveness of several different genetic interventions. (Although the title refers to intelligence, the analysis applies equally well to other genetic traits.)

Incentive-based genetic enhancement: We might consider programs that pay for people with desirable traits to reproduce. For example, we could provide high pay to quality sperm and egg donors based on their genetic profiles. We can also leverage recent research in genetics that predicts success based on one’s genetic profile, and pursue further research along these lines. This will help us make the case for incentive-based genetic enhancement, and will provide an effective means to discover high-quality donors.

Embryo selection: First performed in 1990, preimplantation genetic diagnosis (PGD) is a process by which a handful of eggs fertilized via IVF are screened for genetic diseases prior to being implanted in utero. The screening has traditionally been done using FISH (to detect chromosomal abnormalities like Down’s syndrome) or PCR (to diagnose monogenic disorders like sickle-cell anemia). Of the screened embryos, the healthiest one is implanted. There are several different potential variations of the procedure. As would be expected:

• The more embryos produced, the better (but diminishing marginal benefits)
• The more traits screened for, the better.

Germline gene engineering: There is a lot of hype surrounding the idea of using CRISPR to modify embryos. As many of you know, it’s already been done. While CRISPR does have a great potential to e.g. treat monogenic diseases, Gwern doesn’t think it will produce enormous effects like future versions of embryo selection could. This is partially because genetic studies are not good at isolating which specific single nucleotide polymorphisms are causally responsible for a trait. If we modify a gene which is merely correlated with the trait we are trying to augment, it may be ineffective or possibly even backfire. Moreover, if we want to edit polygenic traits, many individual edits would be required, but currently CRISPR can only be used to make ~5 edits reliably. On the bright side, CRISPR would allow us to increase the frequency of rare beneficial alleles and even create novel “mutations” that we hypothesize to be beneficial (the latter, though, is very risky and wouldn’t pass an ethics board).

Iterated embryo selection: This is a hypothetical technology that could be used to exert a great degree of control over the genome. It involves collecting stem cells from different donors, differentiating these cells into sperm and eggs, and then allowing the gametes to fertilize each other. The zygotes with the most desirable genomes would be differentiated back into sperm and eggs, and the rest discarded. The process can be repeated for several iterations, “compressing multiple generations of selection into a few years or less.” Gwern is quite optimistic about the potential of this technology, expecting it to increase IQ by multiple standard deviations in one generation.

Genome synthesis: This refers to creating a completely new genome from scratch. This procedure would allow the greatest degree of control, and Gwern is quite optimistic about it, but there are several technical challenges that would need to be overcome for implementation.

In general, I think there is high value in having EAs enter government and work on shaping relevant regulations in a more positive direction. It’s additionally possible that conducting more rigorous genetic studies would be useful, but it’s not clear how EA can have a large counterfactual impact there, because academia and the genetic profiling industry are already working on the issue.

Objections and responses

What if increasing these supposedly positive traits results in negative consequences?

It’s important to avoid status quo bias. To quote Bostrom and Ord:

Reversal Test: When a proposal to change a certain parameter is thought to have bad overall consequences, consider a change to the same parameter in the opposite direction. If this is also thought to have bad overall consequences, then the onus is on those who reach these conclusions to explain why our position cannot be improved through changes to this parameter. If they are unable to do so, then we have reason to suspect that they suffer from status quo bias.

But we can give an explanation: evolution! Doesn’t natural selection already work in favor of desirable traits? Isn’t it hubris to think we know better than nature?

Bostrom and Ord give four reasons why this argument is dubious:

1. The environment of our evolutionary ancestors is different in many ways from our modern world. It’s possible that what is beneficial today was an evolutionary disadvantage for our ancestors, or vice versa.
2. There may have been trade-offs in the past that are no longer relevant. For instance, we no longer have to worry so much about large brains imposing high metabolic costs, because food is widely available.
3. Evolution is a blind process, and it’s possible it just never happened to stumble onto the correct combination of genes.
4. What we care about is not the same as evolutionary fitness. Evolution doesn’t optimize for happiness. The ability to rape and plunder might increase genetic fitness, but we don’t consider them good. Likewise, there could be traits which humans value but that hurt fitness.

Genetic enhancement technology would only be available to the rich. It would greatly increase inequality.

I would like to point out that genetic enhancement is not necessarily a zero-sum game. Sure, there are some genetic traits that are almost exclusively positional goods, i.e. they benefit one person by increasing their status over others. Examples might include physical attractiveness or height. On the other hand, many other characteristics such as health, well-being, and intelligence are considered good in and of themselves. We should focus on the latter.

The enhancement of economically advantageous traits such as intelligence would grow the overall pie of the economy, which we could then redistribute more equitably. Then we’re back to standard political debates about how to set the marginal tax rate.

While it is quite probable that the wealthy would have earlier access to reproductive technology, the price would eventually drop to the point where it could be made available to anyone. If necessary, governments could provide social security benefits to subsidize access for the poor.

One way of alleviating the harm due to inequality is by advocating a tax on innate, unearned qualities, such as favorable genetics and inheritance. I believe that these policies will be popular once the technology comes up on the horizon, and will likely play a large role in mitigating the worst risks of inequality.

Genetic enhancement is too taboo for advocacy to make headway.

I see a few reasons for optimism. First of all, surveys indicate significant support for genetic enhancement and similar ideas. A Pew Research poll found that a majority of Americans would support using gene editing on embryos to treat diseases, although only 20% supported using the same technique to increase IQ. Furthermore, when push comes to shove, 90% of fetuses diagnosed with Down syndrome are aborted.

The bottom line is that genetic enhancement isn’t too far out of the Overton window for advocacy to be futile. At the same time, it’s not so universally accepted that we can assume highly effective enhancement will happen before artificial intelligence arrives. It’s in the perfect zone where a concerted effort on the part of EA advocates could make a difference, either by shifting the development timelines or by influencing policies and norms surrounding its use.

AI timelines are too short for genetic enhancement to have any impact

AI timelines are quite uncertain. According to a survey of AI researchers, the median estimate for the arrival date of “high-level machine intelligence,” defined as AI capable of exceeding humans at all tasks, is 40–50 years away, which would allow for one or two intermediate generations. On the other hand, the 75th percentile date is more than 100 years away. Personally, I’m skeptical of the reliability of these estimates, and I recommend to take them with a grain of salt.

In any case, Bostrom and Shulman have argued that even a single generation of iterated embryo selection for intelligence, limited to a small proportion of the population, could have a massive impact on society. The same goes for genome synthesis and advanced forms of “one-shot” embryo selection. There is likely substantial room for impact in the median scenario.

Genetic enhancement would greatly decrease genetic diversity, leaving us more vulnerable to pandemics or other unforeseen disasters.

While this is technically true, I’m not too worried about it personally. The marginal risk increase seems small enough that the benefits of genetic enhancement dwarf it. It should be noted that this objection is strongest in scenarios where enhancement is highly widespread and where technology enables large jumps in one or two generations.

Selecting for high IQ would make the world more vulnerable to agential risks (e.g., lone wolf extremists building nukes in their backyard)

(Relevant background reading: Torres, Bostrom)
While a society composed of high IQ individuals would indeed be more likely to contain individuals capable of building WMDs in their backyard, such a society would also have smarter control and surveillance methods for preventing acts of terror. Moreover, intelligent people tend to commit less crime. But even if we assume that this one specific risk would increase, it still seems likely to me that total existential risk would decrease, for the reasons I mentioned in the “long-term” argument.

We risk creating a race of enhanced humans who won’t care about (or will subjugate) the rest of us.

First of all, this concern is usually based on science fiction like Gattaca. I would warn that generalizing from fictional evidence is not a reliable way to arrive at true beliefs.

Beyond that, research shows that intelligent people are more altruistic and less discriminatory rather than the opposite.

Even assuming the above research is false, from a principled perspective, there seems to be no compelling moral reason to keep our gene pool the way it is. It is typically assumed that creating a class of people who are highly cognitively capable would be unfair to the rest of us.

However, disparities in cognitive ability already exist via natural means. To the extent that these natural disparities are acceptable, then inducing further changes doesn’t fundamentally “change the rules of the game.” In fact, these technologies could actually level the playing field, if we allowed broad distributed access to them.

Historically, increases in the average intelligence of populations has been overwhelmingly considered a positive thing. The long running Flynn effect has likely contributed to lifting nations out of poverty, achieving the exact opposite effect of the dystopian worries we are often reminded of.

Conclusion

Genetic enhancement is a plausible candidate for a cost-effective cause area. There are both short-term and long-term arguments for the desirability of genetic enhancement, and several different approaches that could be used to improve the gene pool. The field is highly neglected and important, but tractability remains uncertain.

The Centre for Effective Altruism has stated that they believe EA needs to incorporate a variety of different approaches to addressing existential risk. I believe genetic enhancement is a cause area that belongs within the effective altruist portfolio. At the very least, it deserves much more attention at the level of cause prioritization than it has received until now.

Appendix 1: Selected traits with heritability estimates

You can see more at SNPedia.

Neuropsychological disorders

• Anxiety: 30-67%
• Major depression: 35-50%
• Epilepsy: 70-88%
• Schizophrenia: 81%
• Alzheimer’s: 58-79%

Physical disorders

• Type-1 diabetes: 88%
• Heart disease: 34-53%
• Breast cancer: 25-56%
• Stroke: 32%

Important psychological constructs

• Affective empathy: 52-57%
• Intelligence: high (see Heritability of IQ for background; the topic is somewhat contentious, but most estimates are >50%, with recent ones >80%)
• Well-being: 36%
• Dark Triad traits: 31-72%
• Big Five personality traits: 41-61%

Also worth pointing out: George Church has a list of single gene variants that have already been shown to have a large effect on human welfare.

Appendix 2: Genetic enhancement for animal welfare

We already have evidence that artificial selection pressure applied by humans can powerfully shape evolution. Look no further than the wide variety of domestic animals and cultivated plants that owe their unique features to the whims of their breeders. Unfortunately, animals have generally been bred for their utility to humans even when this comes at the expense of their own welfare. Over the past century, livestock breeders have substantially increased productivity through genetic modifications. The same modifications have introduced a plethora of horrific afflictions that affect billions of farm animals every year.

While animal breeding may seem like a cause for despair, it also offers us a potential solution to reduce farm animal suffering. Adam Shriver has argued that we should genetically engineer livestock that have a diminished (ideally eliminated) capacity for suffering. This is a research area with huge potential for impact, although it does have some drawbacks as well. Specifically, we would need to be sure that we were actually reducing the feeling of suffering, and not just the behavioral reaction to the feeling. Furthermore, some ethical theories suggest that it is wrong to exploit and kill animals full stop, even if the animals are not in pain. Finally, reducing pain may hamper efforts to completely end the practice of animal farming, as animal rights “abolitionists” argue. Despite the concerns, I would nonetheless be excited to see more research on this issue.

In recent years, there has been growing moral concern for the suffering that wild animals experience due to natural processes such as disease, parasitism, and Malthusian scarcity. Geneticist Kevin Esvelt, one of the pioneers of CRISPR/Cas9 gene drive technology, has suggested that we may have a moral obligation to use our technological powers for the benefit of wild-animal welfare. For decades, transhumanist philosopher David Pearce has been arguing along similar lines. While these ideas are still extremely speculative, further research on the topic could be valuable.

Since it has been more than a year since we last posted an update on our activities, we thought it would be time to post another one. This year, we actually prepared a short annual report that you can also read on our website.

We share this report below in full, together with information on how to access GPI's research as well as some current opportunities to engage with GPI.

GPI's research

• All of GPI's working papers are now available to read on our website. The vast majority of these has been finalised this year and a selection of the highlights can be found at the end of the annual report below.
• We have also started to publish talk recording on our website. At the moment this includes six talks by GPI researchers about their recent research, as well as this year's two memorial lectures.
• GPI's research agenda (which was updated in February 2019) can be found here.

Current opportunities

• Early Careers Conference Programme (ECCP). The ECCP is intended for PhD students and early career researchers in economics, philosophy, and other relevant fields. Participants will come to Oxford for a month to work on a particular research project related to global priorities research and then attend a final conference to discuss their research. More info here. The Forethought Foundation also offers its Global Priorities Fellowship again this summer. Applications for both programmes close on 10 January 2020.
• 4th Oxford Workshop on Global Priorities Research. The 4th Oxford Workshop on Global Priorities Research will take place on 16-17 March 2020. We have still some places available for academics interested in global priorities research to join the workshop. Applications close on 20 January 2020.
• Predoctoral Research Fellows in Economics. We are currently looking for Predoctoral Research Fellows in Economics to join our team in Oxford. Applications close on 24 January 2020 (noon UK time). More info on the predoc programme can also be found in this post from the last round.

Global Priorities Institute Annual Report 2018-19

Hilary Greaves, Director

The Global Priorities Institute (GPI) exists to develop and promote rigorous academic research into issues that arise in response to the question “What should an actor do with a given amount of resources, insofar as her aim is to do the most good?”. The investigation of these issues constitutes the enterprise that we call global priorities research. It naturally draws upon central themes in (in particular) the fields of economics and philosophy; the Institute is interdisciplinary between these two academic fields. This report provides a brief summary of GPI’s key activities and progress over the past academic year, and our key aims for the next year.

Introduction

The academic year 2018-19 has been an exciting one for GPI. It was our first full academic year since GPI was formally established in early 2018, and correspondingly a period of intense innovation, growth and progress within the organisation. We started the year with two new postdoc hires, significantly increasing the research capacity of our team. We settled on a main research focus of “longtermism” – roughly, the idea that the most cost-effective opportunities to do good lie in the domain of beneficially influencing the course of the very far future, whether via mitigation of risks of premature human extinction or otherwise. We developed and self-published 13 working papers, most of them centrally related to this theme. We published a significantly updated version of our research agenda, and devoted significant time and energy to building a network of academic researchers based elsewhere who have similar research interests and motivations. We made significant progress on developing internal structures and processes optimised for GPI’s aims, opportunities, constraints and activities, building the core of a healthy and well-functioning organisation that should stand us in good stead for effectiveness and growth over the coming years.

Going forward, our key near-term ambitions are:

• to grow our research team further, with a particular focus on economics;
• to broaden and deepen our network of aligned external researchers, including providing excellent experiences for visiting early-career researchers who are interested in moving into global priorities research; and
• to expand our own research output, continuing to focus mostly on longtermism for the next year.

I’d like to take this opportunity to wholeheartedly thank the many people and organisations who have helped GPI to get off to such a healthy and rapid start. This includes, but is not limited to: our funders (including the Open Philanthropy Project and several private donors), our advisors (both our formal Board of Advisors and the many people who advise us ad-hoc in an informal capacity), the many external researchers who have enthusiastically engaged with our research themes and events, and the Faculty of Philosophy at the University of Oxford.

1. Research

As mentioned above, our main research focus over the past year has been on “longtermism”. We have implemented a fairly tight focus around this theme, with every researcher pursuing one or more projects that are centrally related to it. This is a genuinely collaborative group enterprise, and a high proportion of our papers have more than one author (not necessarily the norm in the academic disciplines we are working in). 

In September 2019, we started to publish working papers on the GPI website. We now have 13 papers published. Those will serve as the starting point for a working paper series on global priorities research. Additionally, in Trinity term we made video recordings of four GPI research talks, and one additional talk was recorded at an external event. These five talks were published on the GPI website in August 2019. We are planning to continue recording talks by researchers about their recent work to publish and disseminate.

Since our post-PhD research team is currently comprised entirely of philosophers rather than economists, this is reflected in our research output to date, although we hope to make significantly more progress on the economics side shortly. A selection of our key working papers over the past year is summarised below as an Appendix to this report.

2. Academic outreach

A key part of GPI’s mission is to increase the production of high-quality global priorities research in academia more broadly (as opposed to only conducting such research ourselves). For this reason, we have put a lot of effort into identifying and engaging with external academics with nearby research interests who are potentially interested in moving into the areas we seek to promote. (Those areas are articulated in GPI’s research agenda, which we substantially updated in February 2019.)

One of the key vehicles for this “outreach” activity has been a series of two-day workshops at which both GPI and external researchers present work in progress. For this purpose, we construe “work in progress” very broadly – participants are free to give very short (10 minute or even 3 minute) outlines of very early-stage thoughts for discussion, in addition to more conventional academic presentations of work that is already relatively well developed. The workshops have also served as a very effective vehicle for informal discussion, allowing all parties to identify like-minded potential collaborators. We have so far hosted the first two workshops in our series of “Oxford Workshops on Global Priorities Research”, as well as a special themed workshop on the economics of catastrophe.

During the summer we ran, for the second year, our Early Career Conference Programme (ECCP). This programme hosted 14 early career researchers for periods of up to about eight weeks, working closely with a GPI researcher on a project closely related to GPI’s research agenda, and often unconnected to their own previous work. The culmination of the programme was the titular conference at which all participants presented their projects, with a prize issued for the best project write-up. We also hosted ten more senior academic visitors for varying periods of time (most of whom visited during the summer, to overlap with the ECCP).

3. Current team and growth ambitions

GPI is keen to expand significantly over the coming few years. We have made good progress in generating research output with the limited capacity we currently have, but there is a vast amount more to be done. To this end, we are keen to hire additional post-PhD researchers (at all levels of seniority), as fast as the combination of available talent and organisational capacity permit. We are actively building our hiring pipeline via events like our workshops and ECCP, and are excited by the level of enthusiasm expressed in our enterprise by (especially) younger researchers in both economics and philosophy. We are particularly keen to build our economics team.

In GPI’s short history to date, we have been funded by grants totalling £2.5m from the Open Philanthropy Project and donations totalling £3m from other private donors. We are fundraising on an ongoing process to facilitate our growth plans.

Appendix: Selected working papers published by GPI during the academic year 2018-19

Hilary Greaves, William MacAskill: The case for strong longtermism

Let strong longtermism be the thesis that in a wide class of decision situations, the option that is ex ante best is contained in a fairly small subset of options whose ex ante effects on the very long-run future are best. If this thesis is correct, it suggests that for decision purposes, we can often simply ignore shorter-run effects: the primary determinant of how good an option is (ex ante) is how good its effects on the very long run are. This paper sets out an argument for strong longtermism. We argue that the case for this thesis is quite robust to plausible variations in various normative assumptions, including relating to population ethics, interpersonal aggregation and decision theory. We also suggest that while strong longtermism as defined above is a purely axiological thesis, a corresponding deontic thesis plausibly follows, even by non-consequentialist lights.

Christian Tarsney: The epistemic challenge to longtermism

Longtermism holds that what we ought to do is mainly determined by effects on the far future. A natural objection is that these effects may be nearly impossible to predict—perhaps so close to impossible that, despite the astronomical importance of the far future, the expected value of our present options is mainly determined by short-term considerations. This paper precisifies and evaluates a version of this epistemic objection.

Andreas Mogensen: Staking our future: deontic long-termism and the non-identity problem

Greaves and MacAskill argue for ​axiological longtermism​, according to which, in a wide class of decision contexts, the option that is ​ex ante best is the option that corresponds to the best lottery over histories from ​t onwards, where ​t ​is some date far in the future. They suggest that a ​stakes-sensitivity argument may be used to derive ​deontic longtermism from axiological longtermism, where deontic longtermism holds that in a wide class of decision contexts, the option one ought to choose is the option that corresponds to the best lottery over histories from ​t onwards, where ​t is some date far in the future. This paper argues that there are strong grounds on which to reject the ​crucial bridge principle (concerning stakes-sensitivity) that is involved in Greaves and MacAskill’s argument.

Andreas Mogensen: ‘The only ethical argument for positive 𝛿 ’?

This paper considers whether a positive rate of pure intergenerational time preference is justifiable in terms of agent-relative moral reasons relating to partiality between generations, an idea I call ​discounting for kinship​. It responds to Parfit's objections to discounting for kinship, but then highlights a number of apparent limitations of this approach. The paper argues that these limitations largely fall away when we reflect on social discounting in the context of decisions that concern the global community as a whole.

Owen Cotton-Barratt, Hilary Greaves: A bargaining-theoretic approach to moral uncertainty

This paper explores a new approach to the problem of decision under relevant moral uncertainty. It treats the case of an agent making decisions in the face of moral uncertainty on the model of bargaining theory. The resulting approach contrasts interestingly with the extant “maximise expected choiceworthiness” and “my favourite theory” approaches, in several key respects. In particular, it seems somewhat less prone than the MEC approach to ‘fanaticism’: allowing decisions to be dictated by a theory in which the agent has extremely low credence, if the relative stakes are high enough.