Encouraging epistemic modesty, scout mindset and generating new ideas, should be considered among the most effective things people can do for world peace.
What are the most effective things people can for world peace? Below you find my line of reasoning and I believe it warrants the conclusion mentioned in the title.
Assumption: world peace is a worthwhile cause area
The goal of this forum post isn’t to argue that world peace is a worthwhile cause area; I lack information to properly judge the importance (especially the probability of escalation). So instead we just assume it is important and work from there. People interested in exploring this assumption might want to start here: [1][2][3]
The right time: now
There is this human tendency to work on problems just after they’ve occurred. For example in 1953, there was a big flood in the Netherlands, and started improving our dikes afterwards [1]. Other examples include the response to the subprime mortgage crisis [1] and the Chernobyl disaster: in both cases increased safety procedures were established after they were needed.
I fully expect there to be lots of research on international conflicts in the decades to come. We already see this happening: there is a slow increase in calls for research, increase in funds and increased public attention. But we don’t need slow, decisions are being made right now, so sooner is better.
The right next action: consciously decide if this is worth your time.
Most of us have never experienced war. War is something that happens on television. And on an emotional level we might have gotten used to the rule “things that happen on tv have no power in the real world.” Or they might have power in the real world, but not as much as your boss yelling at you because you missed a deadline. The current situation might not feel urgent, and perhaps not even feel fully real. The danger of this is:, apathy: not acting until it is too late.
Again, I’m not saying that acting is the right course of action, but apathy definitely isn’t. Fortunately, it is easy enough to avoid: make a conscious decision if you’re going to spend time on this. Apathy might tell you to choose between “now” or “later”, but as we’ve seen in the first section, there seem to be only two optimal choices for most people: “now” or “never”.
The right scope: how to avoid nuclear war in the near future.
If you decide you want to do something to help with the war, the next question is: what is the most effective thing to do? Since this is a rather broad question, let us first look if we can narrow down our options.
First, we can choose between focusing on the current situation, or focusing on world peace in the long run. This is actually a bit of a false choice, because we can have it both: we can focus on the near term right now and focus on the long run in a couple of years. Assuming that world peace is a worthy cause area, we should therefore choose to do both, but start with the near-term right now.
This still leaves us with a huge number of actions we can take: we can shelter refugees as Poland is doing, we can send household equipment to Ukraine, we can encourage Russians to desert and we can boycott the Russian economy. The choice might not seem easy, until we take a step back and look at the scope/importance of things. Economic impacts seem (by far) to have the lowest scope, and all arguments relating to “yeah, but that will increase the gas prices for people in my country” seem to be suffering from scope insensitivity. On the other extreme we have a nuclear war, which will affect millions/billions of people for many years to come [1]. This makes a nuclear war so much bigger in scope that it is very likely that your most effective action can be found in mitigating that risk.
The right method: research
How to help mitigating the risk of nuclear war in the near future? To answer this question we need to ask ourself another question: what is holding progress in this area back at the moment? MacAskell argues that high-profile disasters usually do not lack money [1]. This seems as high-profile as it gets and we therefore should expect donating money to be the top priority. Similarly we should expect that, given the high-profile nature, plenty of people would be willing to help. So the current the situation seems neither money-constrained nor time-constrained.
Instead, this situation is knowledge constrained. We need to have a better understanding of what is going on, and what is the best course of action. I’ll support this claim with three arguments.
First, let us consider what “not knowledge-constrained” looks like. In that case I’d expect there to be a “world peace plan”, a publicly available comprehensive document with a title such as: “37 theories of war compared, how to judge which theory suits your current war, and how to determine the best course of action.” This document would provide answers for every type of war, and for everyone: politicians, researchers, individuals… everyone. One might ask: isn’t this putting the bar a little too high? And I don’t think so. This very serious problem, and the bar should be through the roof. No compromises. This is not a time to skimp, in any way, on quality. So yeah, having a fully comprehensive document about every war for everyone is putting the bar high, as we should. I could not find such a document the 80.000 hours website [1]. If such a publicly available document would exist, I would expect it to find it on the 80.000 hours website. It is not there, therefore I conclude it does not exist, and conclude that we are (in some ways) knowledge-constrained.
Second, we seem not only knowledge-constrained, but severely so. Looking at the current state of the internet, it is abuzz with podcasts, blogposts, YouTube clips that all offer different viewpoints, different assumptions, different conclusions, different everything. We seem to be living in the age of pundits. Hardly anyone is comparing viewpoints or relating their viewpoints to a bigger picture. They dive right in. A good example of this is the recent 80k podcast [1], which manages to look at a specific viewpoint within the first 5 minutes while looking at only one alternative theory, spending only two sentences on the alternative theory, without proper nuance (they say “not 100% clear cut” which is a severe understatement) and to top it off they don’t offer any insight in the assumptions behind their own viewpoint before analyzing the situation from this viewpoint. This leads people to think that it is okay to have arguments like: “Yeah, this theory seems all right, it is certainly better than some other theories I’ve heard.” It is not. That kind of thinking gets people killed, literally. You don’t support a theory by discounting a single other theory, you don’t provide vague certainty estimates like “not 100% clear cut” and you always do mention the biggest assumptions before moving to conclusions. This single-mindedness can be observed almost everywhere. This indicates: i. that people believe we are knowledge constrained (why else offer new viewpoints?) ii. that knowledge might exist somewhere, but it isn’t easily available. Both cases support the idea that we are knowledge-constrained
Finally, in his EA Global talk Brain Tse argues that we’re not just be a little knowledge constrained, but might even be in the very first phase of understanding: finding the right paradigm, finding the right questions, and defining the right structure. If this is the case than we might not even be knowledgeable enough to realize how much knowledge we lack, which should be a terrifying idea.
People might say: “there probably already is a lot of research! There are people who have dedicated their life to this question, why do we need more/new research?” There are two problems with line of reasoning. First “probably” doesn’t meet the bar. Either you know, and please let me know in the comments, or you don’t. This logic seems to assume topics are well studied unless proven otherwise. This seems unscientific. We should assume there is more to learn, unless proven otherwise. The second problem is the statement “people have dedicated their life to this topic therefore no more research is needed.” The current replication crisis might be the clearest counterexample.
So I conclude that we are knowledge constrained. Research is still fragmented or not readily available. We are still in the early stages of understanding conflicts. This indicates that contributing to research might be the most effective action in this cause area.
The right mindset for research: epistemic modesty & scout mindset
So, what is the best way to contribute to research on avoiding nuclear war in the near future? First of all we need to recognize the phase this research is in, and then act accordingly.
In the previous section we argued that this research is in its early phase. Some people call this the disentanglement phase and define it as: research that involves disentangling ideas and questions in a “pre-paradigmatic” area where the core concepts, questions, and methodologies are under-defined. This phase might be better understood by an example. Consider the question “what is the best thing to do to help the world?” Before effective altruism entered the scene you might have encountered many different articles that answer this question from a single viewpoint. You might have an article which said “climate change is really important!” or an article about animal suffering. There were lots of different viewpoints that might make you believe that they answered the main question: “what is the best thing you can do to help the world?” Only when effective altruism entered the scene did research in this question really mature. Now we have lists of possible answers (cause areas), and an ITN framework to compare them. Although plenty of questions remain, frameworks can still be improved, this is a different – much more well-defined – type of research.
So, how does one approach entanglement research? What do you do when you are still in a pre-paradigmatic phase? The most important thing is the attitude: epistemic modesty and a scout mindset. To quote: I call it scout mindset. It’s what allows you to recognize when you were wrong, to seek out your blind spots, to test your assumptions and change course. It’s what prompts you to honestly ask yourself questions like “Was I at fault in that argument?” or “Is this risk really worth it?” As the physicist Richard Feynman said: “The first rule is that you must not fool yourself – and you are the easiest person to fool.” [1] The most important thing for any research is to have the right paradigm; without the right paradigm, you might confidently believe and act completely wrong. The right time to find the right paradigm is right at the start. So we should out with ensuring we don’t have any blind spots.
The best way to avoid blind spots is to cast a wide net. As both Karnofsky and Tetlock argue, the best way to come to right conclusions is to look at a problem from many (very) different angles. Karnofsky explains this as follows: “If I’m deciding whom to trust about baseball predictions, I’d prefer someone who voraciously studies advanced baseball statistics and watches a huge number of baseball games, rather than someone who relies on one type of knowledge or the other.” I think this is indeed the best way forward. We need to distinguish three phases in research:
Create a list of theories/ideas,
Compare these theories/ideas,
Draw conclusion.
We need to realize that these have to be done consecutively: we cannot draw conclusions without having compared different ideas, and we cannot properly compare theories if we might be missing theories.
So the first step is to have a complete list of frameworks through which to analyze the war. Many frameworks already exist and merely need to found, but also many frameworks will still need to be created. It is like a brainstorm session in a meeting: initially, it is about quantity and diversity and ideas should not (yet) be judged or discarded. In appendix 1 I included 14 different perspectives, which seem important to have included before the next phase. This list contains perspectives to which most people can meaningfully contribute. The end result of this phase would be a long-list of ideas/theories through which to analyze the war. This seems doable in one month, and only afterwards should we proceed to step 2.
In a way, nothing of this is really new. Epistemic modesty and scout mindset are established principles in effective altruism. Furthermore, most if this just boils down to: “do proper research.” However, in practice these principles are absent more often than not. Most articles mix the three phases up: they draw conclusion without properly comparing competing theories, or they compare theories without checking if theories are still missing. To me, the most worrying part about this is what it signals. It seems to say: “it is okay to have an opinion without checking all competing theories” and “it is okay to not worry if your theory is correct.” It lowers the bar for research, and this is not a topic with the luxury of a low bar.
Some people might argue that there is not time for nit-picky detailed research. I might agree that we need preliminary conclusions pretty soon, but I don’t think that properly hedging research, or listing assumptions take much time. I’m not against taking shortcuts, I am against taking shortcuts without letting your audience know.
And this is the second thing you can do to help. Not actually doing research itself, but help maintain the values that are essential to proper research: epistemic modesty and scout mindset. Especially in these early phases of the research. So next time you encounter a podcast/blog/video, ask yourself: is everything mentioned properly supported but most of all: are shortcomings/assumptions/gaps in knowledge explicitly mentioned.
Conclusion
What is the most effective thing people can do concerning the current Ukrainian-Russian conflict? Although impossible to say for certain (I don’t know your other opportunities), I would say that for most people the following two things should be near the top of their list:
Help maintain the values that are essential to proper research: i. encourage people to consciously make a decision on whether or not the current conflict is worth their time, ii. encourage people to focus on the near-term nuclear threat (instead of, for example, economic issues in their own country) and iii. be very critical of what you read and try to foster the values of epistemic modesty and scout mindset.
Contribute to research by exploring the current conflict from a specific perspective, without jumping to conclusions or judging other perspectives. See appendix 1 for example perspectives.
Appendix 1: possible perspectives
Below are listed 14 perspectives, each of which might yield a piece of the puzzle to answer the question: how to avoid nuclear war in the near-term. This list is far from exhaustive and only meant as a starting point. Included in this list are first steps that can be taken and a time estimate. Neither should be considered definite, but are only meant to clarify the perspective, and perhaps make it easier for people to start. Not every perspective is suitable for everyone, but many perspectives allow for a wide variety of people to contribute.
Theoretical perspective: Homo economicus.
Time estimate: 40 hours.
Relevant because: This isn’t the most neglected perspective and, as such, perhaps not the most important to put your time into. On the other hand, having a clear overview of this perspective with careful and explicit assumptions will always be a benefit.
First steps: What would everyone involved in this conflict do if they acted fully rational and fully selfish?
Theoretical perspective: Behavioral economics.
Time estimate: 200 hours
Relevant because: People don’t always act rational
First steps: Behavioral economics is a field lacking a unifying theory. This makes it harder to apply. A possible first step would be to create a unifying theory. In an article you’ll find here I look at 210 different types of irrational behavior and show that 110 of those can be unified under 5 rules of irrationality. These 5 rules follow directly from Trivers’ theory of self-deception as also described by Hanson and Simler. The article is still very much under construction, and although I’m probably biased, I think that it offers a good starting point for analyzing the current situation from the perspective of behavioral economics. I’d be interested to hear what you think of these rules, and how you would apply them to the current situation.
Outlier behavior
Time estimate: 20 hours
Relevant because: Both theoretical perspectives above assume there is some structure to the way people behave. However some people seem to be wired a little differently and might not fit the box. Although I don’t expect this is the case for Putin, but if is the case, it is essential to know. This perspective seems to have been excellently explored here.
First steps: Find someone with experience as a psychiatrist, present them with the fact found by the intel-gathering/fact checking perspective and ask their opinion. Or get in touch with the author of the article above and see how you can apply these ideas to the current situation.
Literature research: the history of the start of wars
Time estimate: 400 hours
Relevant because: i. It might help to avoid mistakes made in the past, ii. there probably is a wide body of literature on this topic.
Possible first steps:
Make a long-list of wars last couple of centuries,
Sort this list based on: a. how much do we know about this war? b. how similar is this war to the current situation?
For every war (starting at the top), review the literature on what factors seemed to have contributed to the start of the war.
Literature research: the history of peace
Time estimate: 40 hours
Possible first steps: 1. Find long periods of peace. 2. What do they have in common?
Relevant because: if one only studies war, one might conclude “these 6 factors were present in the most all wars!”, while ignoring the fact that these factors were also present in times of peace. One cannot study war without studying peace.
Empirical research: watching/reading first-hand accounts of wars
Time estimate: 60 hours
Relevant because: Doing scientific literature research only might lead to group think. It would be nice to have some people who start forming their ideas/opinions from scratch. This will not have the same quality as decades of dedicated research, but again, it might offer something unique which the other perspectives miss. An example of an analysis from this perspective can be found here.
Possible first steps: Watch/read first-hand account of wars, while avoiding secondary or interpreted sources. The following two books and six documentaries are examples which fit these criteria.
Occupational Hazards – Rory Stewart (book), more detailed than the movie so I prefer this.
The Fog of War: Eleven Lessons from the Life of Robert S. McNamara
Empirical research: First-hand experience
Time estimate: 40 hours
Relevant because: Another unique perspective is the perspective of people who have actually experienced a war. Some things might be astonishingly obvious for them, while being completely hidden for others.
Possible first steps: Find people have experienced war. Ask them their opinion. For maximum diversity it would be nice to have both elder people who have experience WW2, and younger people who have lived in a more contemporary warzone.
Intel-gathering / fact checking
Time estimate: 300 hours
Relevant because: Only with good data can we come to good decisions.
Possible first steps: One approach would be to have a decentralized Wikipedia-like approach where everyone can contribute. This leverages the huge amount of people who are willing to contribute to the avoiding a war. Since most of the data which is relevant isn’t secret, it really can be as simple as someone calling their aunt who lives in Russia.
A decentralized approach needs a platform for people to meet. Perhaps the best way is to simply join https://twitter.com/bellingcat. I haven’t looked into this properly. There might be better organizations available. From what I know this would seem the best first steps:
Make a list of possible organizations that could jumpstart a decentralized intel-gather approach.
Approach them and pitch the idea.
In case of success, proceed to step 5
In case of failure, look for other ideas, most likely someone is willing and able to code a platform like this
Avoid unstructured heaps of facts, and perhaps sort things along the following lines
Facts about Putin. What is he like? What is his current situation? How much power does he hold himself?
Facts about the Russian political situation. Who are the people who make/influence the decision? Are there neutral parties whom Russia trusts?
Facts about the economic situation.
Facts about the population in Russia. There are roughly 150 million people living in Russia. It seems likely that at some point they will play a role. What are their opinions? What do they know about the war? For the sake of diversity it would seem realistic to further subdivide this category in: people living in the urban areas, people living in rural areas, and soldiers in the army.
Controversial perspective: What would Donald Trump do?
Time estimate: 3 hours
Relevant because: Love him or hate him, you cannot deny that Donald Trump has a very different style. Furthermore this style gives results; at least sometimes. The controversy of his style might indicate that understanding this perspective might actually fill in some blind spots the other perspectives missed.
Possible first steps: 1. Invite some friends, 2. Grab a beer, 3. Ask yourself, what would Donald Trump have done if he was still president? 4. Once you’re sober, turn this into a single page of understandable text.
Controversial perspective: Seduction community
Time estimate: 40 hours
Relevant because: The seduction community has a wealth of information concerning interpersonal relations. Their information is well documented (books [1][2][3], YouTube clips [1][2][3], blogposts[1][2], etc.) and most often based on experience. Most importantly however, their controversial nature indicates that they have some assumptions which are not shared by some other perspectives.
Possible first steps: 1. Turn the current situation into metaphors, for example: i. If you were to seduce Putin, what would you do? ii. If Putin were a tough looking guy and he was trying to steal your girlfriend. How would you handle this? 2. What does your answer imply for the real world?
Ideally this perspective is further subdivided according to the different branches of the seduction community.
Teachers | Police officers | Friends and Family | Relationship Counselors
Time estimate: 10 hours each
Relevant because: These four perspectives have one thing in common: actual experience in conflicts. Although their types of conflict resolution might not closely match the current situation, I think that people who have lots of experience with the current situation are non-existent. These perspectives might help generate more possible solutions.
First steps: There are at least two strategies that can be used in each perspective: data-analysis and hypothetical.
The first, data-analysis
With a group of people, identify conflicts you’ve been in
Divide them into a list with good outcomes and a list with a bad outcome,
Can you find commonalities within the list and differences between the lists?
How might this relate to the real world? Are there certain things that should or shouldn’t be done?
The second, hypothetical, is simply to think of a hypothetical situation which metaphorically resembles the current situation. For example, as a teacher you might consider a case where “a student repeatedly violates the boundaries of fellow students by stealing their stuff”. What would you do? Is punishing the best idea? What other options are there? What do these options look like in the real world? For example, if you think that calling the kids parents is the best idea, what would this look like in the real world? Perhaps this indicates that getting third parties involved is a good idea. Can we get a third party involved?
This post left a bad taste in my mouth, and I wanted to briefly touch on why:
1. You say that the right time to act is now, but this is extremely ambiguous.
What should people do now? Maybe you’re referring to some of the actions mentioned later in the post like “consciously deciding if this is worth your time” and “doing research”.
This reminds me of a scene from Friends where one of the characters says that he has a plan. And his plan is that someone should come up with a new plan.
And there seems to be an inconsistency in your approach. You say there is a “slow increase in calls for research, increase in funds and increased public attention.” Instead, we should… call for more research??
2. You say that the right method is research, but your support for this is not strong.
You say you looked for different theories of war on the 80,000 hours website. I think this pays too much deference to the 80k team. They’re smart, but not all-knowing. There are other people with views on war. I searched Google Books for “theories of war” and got 4.9 million results. If someone writes up the 5 millionth book on war, is that a tractable way of reducing war?
More generally, the existence of research and the existence of easily comprehensible and actionable plans are two different things.
And are different theories of war even the right topic of research to focus on in the first place? There are other things that it could be helpful to understand (e.g. effective activism tactics, moral circle expansion, technical capabilities of weapons systems, the history of military/political leaders).
And is research the right course of action? How about political lobbying? How about relief for refugees? How about grassroots activism?
It seems like the sum of your reasoning on this is that the scope of nuclear war is really large. So figuring out how to stop nuclear war would be really good. So we should figure out how to stop nuclear war. Much of the rest of your post felt like applause lights.
Hello Nathan! Thank you for your reply! I appreciate the honesty and your comments are very clear. Allow me to elaborate:
1. Yes! The Friends scene where a character has a plan to make a plan (together) is exactly what I mean! I feel we are doing too little planning. Why wouldn’t this be a valid argument? It is like when you’re on a holiday, you’ve just arrived and everyone just starts doing something. One person starts building a tent, someone else is going for the dishes. You notice that everyone seems to forget to go to the camping owner to see if you’re even allowed to set up your tent. It also seems that we might need to go shopping before we start cooking. I think the right thing to do in those circumstances would be to call everyone together and make a plan together.
2a. Once again, your comment is very clear. If we already have 4.9 million books, why write another one? My point is this: even if there 4.9 million books, their content does not seem to have reached the EA community. What I’d like to see is to have a dedicated (EA) team find (and summarize) the best books and figure out the implications for EA, as I (tried to) describe in appendix 1.
2b. You suggest many different topics of research. That’s great. I agree that all of these are very much worth studying. It was never my intention to limit the scope of research (on the contrary!) and I think you mention worthwhile avenues. You also mention ideas other than research. I’m not a fan of those. I feel the comparative advantage of EA is “think first, act afterwards.” Also I feel that relief for refugees is lower in scope (and neglectedness).
Your summary does capture the essence of my article. However, I feel it doesn’t do it justice. I still feel that noticing that EA seems to be having a “act first, think later” mindset (instead of our comparative advantage “think first, act later”) is extremely important. And that appendix 1 offers both an indication of where our thinking is lacking, and indication of how to start improving our thinking.