Education Matters for Altruism

First written: 23 Sept. 2013; last update: 31 Oct. 2014


Learning is an extremely important activity for altruists, and by this I mean not just discipline-specific study of issues in activism and cost-effectiveness analysis but also big-picture insights about the world at large, drawn from many academic and non-academic perspectives. Learning can seem suboptimal and not helpful in the short run, but used properly, it can pay off more than most financial or single-domain-focused investments. It's important for young activists not to neglect learning in order to just "do more to help now." That said, the reverse is also true: They shouldn't only learn, because doing is an important part of learning, and staying connected with some shorter-term projects can prevent you from drifting away permanently from the reality of the suffering whose prevention ultimately makes the learning worthwhile.


"Want to save the world? LEARN!!!!
Then learn MORE!!!!
Then, probably have kids [or not] and KEEP LEARNING!!!!"
--Michael Vassar

There's a stereotype that says younger people are naïve and idealistic, while older people are experienced and cynical. Insofar as this is true, there are probably many contributors, but one suggestion is that young people "think they know everything" and are assured that the changes they want to make will be beneficial. The older people align more with Edmund Burke in suggesting that "society is complicated, and changing it without messing things up is harder than it looks." There are obviously other differences as well -- e.g., that young people have different values than older ones, such as being less opposed to gay marriage or more supportive of animal welfare, and these are not due to amount of experience but merely differences in culture.

I often find that those who think a policy or strategic issue is "just obvious" tend to be those who know least about it. The more you probe, the more you see that different sides have reasons for thinking as they do, and it's less obvious how to weigh up all the competing factors. This isn't to say there aren't places where most reasonable people agree that policies are sorely misguided and need changing. But if you haven't found at least one major drawback to whatever proposal you think should be adopted, you might want to dig deeper into the complexities at play.

Returns to wisdom

There's an admirable tendency among some activists to "do something now." Given how full the world is of horrors, this heart-felt urgency is not only understandable but actually an important feeling to have. If we lose our sensitivity to the urgency of the issues at stake, we risk not giving them the weight they deserve, thereby drifting into either apathy or another moral stance that takes a dispassionate approach to suffering and doesn't attend to its overwhelming importance.

Yet, when I look back at the history of my altruistic efforts since the year 2000, what I see is that I was often mistaken, not only about which causes would be best to support but even what their signs were, i.e., whether they made the world better or worse. While financial markets can return 5-10% per year, and internal growth rates on movement-building may return more, the "rate of return on wisdom" can be extraordinary. This wisdom is not just about new data or new arguments but more broadly, new ways of understanding how the world works, looking at epistemology, and deciding what you care about.

The value of education

When I look back at my history since 2000, I also see that many of the most valuable tools I acquired came from education. By this I mean not just learning about effective altruism, rationality, ethical issues, strategic considerations, career choice, and so on, but also learning about the intellectual foundations of various academic disciplines. A non-exhaustive sample:

  • Mathematics: probability, real analysis, abstract algebra, and general "mathematical sophistication"
  • Statistics: Data-analysis tests, Bayesian reasoning, properties of estimators, machine-learning models, Occam's razor
  • Computer science: Programming, algorithm design, computational complexity, Turing machines
  • Physics: Relativity, quantum mechanics, calculus of variations, cosmology
  • Biology: Evolution, competition and cooperation, diversity and convergence, ecology, biological-machine architectures
  • Economics: Marginal thinking, Pareto and Kaldor-Hicks efficiency, von Neumann-Morgenstern utility, macroeconomic models, game theory
  • Psychology: Cognitive biases, reinforcement learning, theories of emotion, persuasion, disorders, evolutionary psychology
  • Cognitive science: Brain algorithms, learning and memory, cellular mechanisms, philosophy of consciousness
  • History and political science: Government systems, power structures, evolution of nation states, the roles of religion and culture, how to lobby elected officials
  • International relations: Realism vs. liberalism, causes of war, correlates of cooperation
  • Sociology: Social movements, how memes spread and change, persuasion models, human diversity and universals
  • Literature and humanities: Alternate ways of seeing the world, exploring other people's ways of thinking, reflecting on what you care about most, understanding social relations
  • Philosophy: Epistemology, rationality, ethics, approaches to looking at reality
  • ...and on and on.

You'll notice that my list contains basically every academic subject! And there's a reason for this: Reflecting on what we value and how best to change the world requires familiarity with a wide array of ways of thinking and intellectual considerations that we should heed. If you're trying to advance the state of the art in one particular domain, it suffices to dig very deep into that field and probe it as much as possible. But if you're trying to do the most good in a global sense, you need a global perspective, which requires learning a little bit of everything.

When people ask me what major to choose in college, my first answer is: "Introductory studies." This is a joke, though I wish it weren't. What I mean is, take as many "intro" courses as you can, because these allow you to level up by exploring new perspectives on the universe that different domains have to offer. The returns then diminish as you keep taking more and more courses in the same field. Of course, some areas are more important than others, so you'll ultimately want to focus on them, but every subject has something you can take away.

From Eliezer Yudkowsky's "Twelve Virtues of Rationality":

Study many sciences and absorb their power as your own. Each field that you consume makes you larger. If you swallow enough sciences the gaps between them will diminish and your knowledge will become a unified whole. If you are gluttonous you will become vaster than mountains.

As a substitute to an Introductory Studies major, you might consider listening to at least one or two lectures from every academic department on a YouTube channel for a university. Get a taste of what kinds of issues a field deals with and what tools it uses to approach its problems.

Robin Hanson:

I was a physics student and then a physics grad student. In that process, I think I assimilated what was the standard worldview of physicists, at least as projected on the students. That worldview was that physicists were great, of course, and physicists could, if they chose to, go out to all those other fields, that all those other people keep mucking up and not making progress on, and they could make a lot faster progress, if progress was possible, but they don't really want to, because that stuff isn't nearly as interesting as physics is, so they are staying in physics and making progress there.

For many subjects, they don't think it's just possible to learn anything, to know anything. For physicists, the usual attitude towards social science was basically there's no such thing as social science; there can't be such a thing as social science. [...]

It's just way too easy to have learned a set of methods, see some hard problem, try it for an hour, or even a day or a week, not get very far, and decide it's impossible, especially if you can make it clear that your methods definitely won't work there.

You don't, often, know that there are any other methods to do anything with because you've learned only certain methods. [...]

I can tell you there are a lot [of methods] out there. Furthermore, I'll stick my neck out and say most fields know a lot. Almost all academic fields where there's lots of articles and stuff published, they know a lot.

Altruism can motivate learning

Philomaths like myself love to learn about all subjects, but it's interesting that I became much more of a philomath after inclining toward altruism. Before I cared about making a difference in the world, I played video games and often disliked doing homework. Once I realized that I had enormous potential to reduce suffering by others, I became a better student, in part because I knew good grades would help me be more successful in attaining a lucrative or influential career, but also because when you want to make the world better, you really have to know how it works. What before had been dry facts and unmotivated theories were now valuable pieces of information -- gold nuggets of data and insight waiting to be harvested.

This is how I see learning to this day: Each new datum or idea is a little boost to my overall ability to approach activism and life in general with more wisdom. It's like accumulating experience points, which is what makes RPGs so addictive.

Taking the big picture: Textbooks, review articles, and Wikipedia

The world is so vast that it's easy to get lost in a single domain. Even within a single intellectual community focused on a single problem, there will be new discussions every day, new ideas to explore, and new posts to comment about. Our friends on Facebook, online forums, in-person groups, mailing lists, news websites, etc. can keep us focused on a particular outlook on the world.

Social networks and communities are extremely valuable in many ways, for emotional support, intellectual development, and sustaining altruistic motivation. That said, it's important also to take a bigger-picture view and think about what other thousands of intellectual communities you haven't explored yet. What other topics and world views should you learn about?

Social communities can also lead us to spend most of our time reading material that our friends have written, even though our friends' writings may not contain as much deep scholarship as well renowned writings by more mature authors. (Yes, that statement has something to say about this essay. :P) There can also be a tendency to focus on things that are new and shiny over those that are older but more solid. Robin Hanson said it well:

as a blog author, while I realize that blog posts can be part of a balanced intellectual diet, I worry that I tempt readers to fill their intellectual diet with too much of the fashionably new, relative to the old and intellectually nutritious. Until you reach the state of the art, and are ready to be at the very forefront of advancing human knowledge, most of what you should read to get to that forefront isn't today's news, or even today's blogger musings. Read classic books and articles, textbooks, review articles. Then maybe read focused publications (including perhaps some blog posts) on your chosen focus topic(s).

I think textbooks are often under-read, outside of course requirements, relative to the density of insight they contain. Close runners up are Wikipedia, review articles, and popularization books.

I can't stress the importance of Wikipedia enough. I try to read Wikipedia as much as possible, and if I'm reading anything not on Wikipedia, I ask myself, "Is this really better than reading a Wikipedia article on this topic or on some other topic?" Wikipedia allows you to see a broad range of perspectives that you'd miss reading a few individual opinions on the issue, and it often gives more context and perspective.

News vs. the long term

There's a similar phenomenon with current events: It seems we sometimes pay too much attention to them relative to their overall importance. Obviously politicians, policy analysts, journalists, and lobby groups need to be completely up to speed on the news. Voters too should have some general understanding of current issues (though maybe at the level of reading Wikipedia rather than CNN). It's important for scandals to be reported widely in order to serve a deterrence function. However, in terms of its broad, big-picture significance, news is not really different from history. It's good to be somewhat familiar, to update your world models for how society functions, what kinds of actions people take, and in general how social trends evolve. But it's not crucial that you read today's news. You could just as well read last year's news, or, for that matter, news from 1925. The general principles of psychology, politics, economics, and society that news illustrates are relatively similar then as now. (See also "Appendix: Why can you only promote new media?" and "Appendix: News values are not optimal for learning")

A good history course will not just teach from a textbook but will assign primary-source readings as well. These allow you to see historical developments from a different level of abstraction. Getting up close with details can give you a better intuitive picture of what kinds of things are being described by the high-level narrative. This is similar to studying some particular examples within a statistical sample in addition to merely reporting aggregate metrics. News is like a primary-source document in a history class: It's important for building intuition and highlighting more general principles, but it also shouldn't be overdone relative to bigger-picture insights.

Robin Hanson has a criticism of news reading on the grounds of its superficiality and evanescence. I'm more positive about news than he is and consume some of it myself, because I think it's an important source of data on how the world works in a number of domains. That said, news stories often overemphasize extreme anomalies rather than systemic trends, and in any event, we may not want to read it excessively at the expense of already-digested insights by many other smart minds.

Learning more in younger years

Academics are an important source of insight, but there are many more: Friendships and social relations, experiences in the workplace, feeling new emotions, and doing new things. These all expand your repertoire for what life is like in various corners of the globe and various avenues of thought. The number of unique experiences, while finite, is astoundingly large.

Many algorithms for approaching the explore-exploit tradeoff and function optimization introduce randomness to prevent search from getting stuck with just what seems the best so far. In general, it makes most sense to front-load the exploration early on. For example, in simulated annealing, the search begins at a high temperature, where we often switch randomly to seemingly suboptimal places, and as time goes on, we settle down into the best optimum we've found. An epsilon-decreasing strategy in the multi-armed bandit problem embodies a similar idea.

The same concepts apply to altruism: At the beginning, a large part of our focus should be on learning, observing, and expanding our world models, without getting mired into a local optimum too early on. With time, we should begin doing more hands-on work -- relatively more talking and relatively less listening.

Think of another perspective: In science, there's a tension between funding work with immediately tangible benefits versus basic research that may pay off in 50 years or may accomplish nothing. Generally, basic research is under-funded because it's less profitable in the near term, but I think general wisdom is that basic research has higher expected returns per dollar. It seems to me as though many altruists also incline toward work with immediately tangible benefits -- understandably so, because the world contains so much suffering, and we want to do something about it now. Yet in terms of expected value, the "basic research" approach to altruism may have higher payoff. This is not an unqualified statement; there are many cases where basic research stays in its own bubble and never has much outside impact -- both in science and altruism. It's good to keep one foot in each of the practical and academic worlds to make sure that you don't drift away from focusing on relevant topics and so that you have a better sense of how the research might actually be used.

Why we shouldn't front-load learning to an extreme degree

It might be tempting to suggest that young people should spend all their time learning, saving the "hands dirty" work for later, once they've better figured out how to best make a difference in the world. While it definitely makes sense to do more learning early, I think it's also valuable to do some concrete projects alongside the more abstract work. Here are two reasons:

  1. Risk of losing your ideals. Motivations are fickle creatures, and yesterday's overridingly important ideal may be tomorrow's childish fad. If we spend all our time on intellectual examination, we may forget about the hard, raw suffering that motivated us to explore intellectual topics in the first place. If we keep putting off "getting our hands dirty," we may do so indefinitely and not realize the full potential of the insights we're accumulating. Of course, there is something to be said for specialization: Maybe some people should spend most of their lives on more intellectual pursuits to better advise other on-the-ground activists. But the intellectuals at least need to be sure they're taking the proper steps to transmit this information in an effective way rather than letting it go to waste.
  2. Learning from concrete doing. If you're trying to learn more about the best ways to do X, it helps to actually try doing X in various ways, rather than just theorizing about it from your armchair. There's a lot to be learned, even about higher-level principles, from the "mundane" work of planning, organizing, administration, leading a team, coordinating projects, and so on. These are at least as much something to explore and use for updating your world model as are the smooth manifolds of general relativity. Knowing more about how things work on the ground can better inform your pursuits in the ivory tower.

There's a quote attributed to Abraham Lincoln (perhaps incorrectly): "Give me six hours to chop down a tree, and I will spend four hours sharpening the axe." This nicely illustrates the idea of front-loading learning, but I would modify the recommendation a little. First try taking a few whacks, to see how sharp the axe is. Get experience with chopping. Identify which parts of the process will be a bottleneck (axe sharpness, your stamina, etc.). Then do some axe sharpening or resting or whatever, come back, and try some more. Keep repeating the process, identifying along which variables you need most improvement, and then refine those. This agile approach avoids waiting to the last minute, only to discover that you've overlooked the most important limiting factor. Finally, one important difference between tree chopping and altruism is that in altruism, the goals themselves are much less clearly defined, and figuring them out is part of the metaphorical "axe sharpening" process.

Experts, age, and challenging yourself

One of my teachers in 8th grade gave really hard tests. He explained the rationale with the following analogy. Suppose you want to improve at ping pong. Are you going to get better by playing against your little sister? Not unless she's good. Instead, you'll get better by playing against someone who's very talented -- someone who can challenge you. Similarly, challenging tests help students significantly improve their thinking skills.

The effective-altruism movement is mostly composed of young people. This is great, but it can also lead to lots of discussions in which young people talk exclusively to other young people. Often I feel there's not enough external input, from experts and older thinkers who can provide wisdom, experience, and difference of perspective.

I find it extremely useful to constantly seek outside perspectives, especially from senior policy analysts, leading scientists, and other objectively impressive public figures. Reading the big names in a field seems to me a better way to challenge my views and absorb crucial insights than having yet another conversation with 20-something-year-olds. There's no question that young people can be brilliant and have important new contributions, but expertise and wisdom from age often beat intelligent theorizing, except in a few narrow domains where theory is all we have to go on.

It's easy to get trapped in your own intellectual bubble and think that your group of friends has the answers. Make sure to keep stepping outside that bubble. Keep getting restless in search of new ideas and views different from your own. That's where you can challenge yourself.

Things you never knew you never knew

Sometimes effective altruists seem obsessed with "quantified efficiency": Metrics indicating performance, whether in terms of QALYs per dollar, time management, survey results, or diet/exercise optimization. Quantification is helpful and can often highlight major gaps that your qualitative mind may have missed, or conclusions that weren't obvious before crunching the data. Numbers allow for doing many more computations than our brains could execute on their own.

At the same time, I fear that some effective altruists get carried away and miss the forest for the trees. Remember "garbage in, garbage out": Metrics are only as good as the reasoning that tells you they're important to optimize, and they can also miss crucial considerations. Hyper-optimizing on narrow or even fairly broad metrics may mean sacrificing important pieces of unmeasured value. Metric optimization can give an illusion of progress, when in fact a broader view of the situation would suggest that things are a lot fuzzier than you thought. Sometimes the qualitative, big-picture approach of the human brain is more adept at solving these macro-level problems than other tools currently at our disposal. Holden Karnofsky discusses this and related issues in "Passive vs. rational vs. quantified."

One big-picture domain where metric optimization has a hard time is with discovering, in the words of Disney's Pocahontas, "things you never knew you never knew." It's not clear how to build metrics that capture dimensions of a problem you haven't yet thought to explore. If we hyper-optimize too much on our current goals, we may neglect the importance of taking a step back to think about the big picture. Narrowed focus is one of several concerns raised by the authors of "Goals Gone Wild: The Systematic Side Effects of Over-Prescribing Goal Setting."


From a short-term perspective, it feels like studying abstruse topics in mathematics or political theory isn't helping anyone. It doesn't seem morally urgent. Yet, when I look back on my life with a bird's eye view, I can see that if I hadn't studied big ideas from a far-sighted perspective, I would be spending time on things significantly less important. In the long run, the seemingly unhelpful details of how sodium-potassium pumps work actually matter insofar as they allow you to better understand what life and consciousness are at a deep level, which then has major implications for altruism at large.

Michael Vassar encourages people to "Be More Curious!" Curiosity is Eliezer's first virtue of rationality. While it's important to constrain your curiosity so as to focus on the more altruistically important topics, don't confine it so much that you avoid making serendipitous leaps to unanticipated regions of state-space during your search for the best ways to improve the world.

By a similar token, boredom can be a virtue as well if it keeps you exploring many domains of life rather than focusing too much on a narrow cause that would appear less important than other causes if you took a broader perspective.

One of my friends used to observe what a tragedy it was that the world contained far more books than any person could ever read. I sympathize with bookworm Henry Bemis from "Time Enough at Last." When I encounter interesting new reading material, I feel almost sexual excitement, and sometimes I avoid exposure to articles about fascinating topics in order to not arouse myself when I'm busy. While it is superficially sad that we can't learn everything or even more than a tiny sampling of the world's knowledge, on reflection it's not so troubling for a few reasons:

  • Sometimes when you're really hungry, you want to eat everything at the (vegetarian) buffet. Then as you eat, you enjoy what you have, and you feel you don't need as much. Likewise, we can enjoy learning what we can learn without needing everything. Happiness is in our minds, not in external facts about the world. Of course, one difference between food and learning is that from an altruistic perspective, you want to curb overeating to improve your life expectancy so that you can do good work longer, while it's altruistically beneficial to try to overconsume knowledge as long as doing so is an effective use of time.
  • Other people and institutions can compensate for the incompleteness of what we personally know. We can develop relationships with others who complement our own expertise and who can fill us in where our own insight is lacking. Taking a more expansive view of ourselves to include others, or even to include the universe as a whole, can allay the seeming sadness of the fact that "we" can't know everything.
  • Knowledge is ultimately not valuable in its own right but only as a means toward helping others. It's instrumentally useful to crave knowledge just as it's instrumentally useful to crave food or sleep or warmth. Food and sleep are not intrinsic values that we should create more of for their own sakes; they're ultimately just tools that help us move toward what actually matters, even if we selfishly find them pretty enjoyable in the moment.

Trying new things takes effort

Recently I browsed through the top 100 most popular podcasts on iTunes. My instinctive reaction was to navigate to the ones that matched my interests and therefore seemed most alluring. Other podcasts on unfamiliar topics or by authors I had never heard of seemed less exciting. "Familiarity breeds liking," as the psychologists say, and the familiar topics have more fuzzies associated with them from positive past experiences.

Yet, I realized, my emotional response here is not fully adequate. While topics that interest me may be more relevant, there's also diminishing marginal value to additional exploration in the same fields. In contrast, listening to a completely new topic or author would expand my horizons, and rather than seeing it as uninteresting because I haven't tried it, I should encourage myself to explore new things more than I might do naturally. In the explore-exploit tradeoff, my brain seemed to be favoring exploitation relatively more than seemed optimal from a learning standpoint.

It can actually take some work to explore completely new topics or authors. If I'm looking to relax or comfort myself, I'm more likely to go with a familiar topic and writer that I know are fun. In contrast, trying something new requires effort before it pays off with new rewards from discovery of new insights.

Get the gist

When I was in the middle of college, I had a troubling realization: Many of the things I had learned in high school, or even just a few semesters ago, were now vacant from my brain. Am I losing my memory already? I wondered. And if knowledge fades this quickly, is there a point in learning it in the first place? Of course, knowledge learned once is easier to re-learn. But it seems an uphill battle to try to keep re-learning information that will just fade once again.

I'm no longer particularly troubled by the evanescence of memory for a few reasons. For one thing, if I have a useful insight or learn an important fact, I often write it down, either in an essay or on Wikipedia. This creates a sort of extended memory for myself. Moreover, the most important discoveries should be salient enough that I remember them when other memories have faded.

In addition, I think much of the value of learning comes from the way knowledge shapes your mind in general. I can't recall many of the methods of solving differential equations that I once knew, but I do remember the gist of what those methods involved. Eliezer Yudkowsky has discussed the idea of absorbing "the reason and the rhythm behind ethics". I like this phrase. When I read about something, I aim to get a sense for the "rhythm" of the field -- whether physics, philosophy, or politics. I try to absorb an intuitive understanding of the kinds of things that the field involves, both at a high level and on a daily basis. I don't need to know most of the detailed facts, but studying some detailed facts with an eye toward honing my instinctive sense of the topic is important.

The "gist" of a subject is something that lingers long after the specific facts have faded. And in any case, getting the gist is often the most action-relevant endpoint anyway. Brains evolved to forget unhelpful facts (although energy and storage savings were presumably another motivation for humans' limited memories). Our intuitive understandings are beautifully distilled summaries of the essence of large amounts of data. When I read about a topic, I typically hunt for the gist and don't worry about remembering the details. They're at my fingertips via a web search anyway.

Many school tests, as well as trivia shows and cocktail-party banter, emphasize the wrong thing: They value memorization of words and facts. This makes testing easier, because measurement is more objective, and it does at least provide some signal of understanding, because you can't pass a history test without having read the textbook or listened to the lecture. But it emphasizes exactly the wrong end of knowledge: knowing trivia rather than looking for an overall gist. To their credit, many educators aim to shape tests in ways that emphasize concepts more, and of course, essay writing is much more conducive to understanding over regurgitation.

Seeking the gist applies in the realm of quantitative sciences too. I've heard several people say things like "You can't really understand physics without math", but I think this isn't accurate. Equations in physics express ideas that can be described by words and analogies. Of course it would be necessary to introduce many new concepts to explain physics in detail without math, but there's nothing fundamentally privileged about mathematical symbols. Mathematicians often highlight the "main idea" behind a proof before delving into the gory details, and this main idea is typically the most important point to take away from the proof. If you remember the main idea, you can probably reproduce the detailed manipulations. Sometimes I read a proof, understand each step, but still don't grok what's really going on. In such cases, I need to take a step back and ask myself what the overall strategy is. When I figure that out, I can finally feel satisfied with the proof.

Luke Muehlhauser described his impression that fellow employee Carl Shulman gave answers using facts rather than overall impressions. In contrast, Luke explained:

I’ve read a lot of facts, but do I remember most of them? Hell no. If I forced myself to respond to questions only by stating facts, I’d be worried that I have fewer facts available to me than I’d like to admit. I often have to tell people: “I can’t remember the details in that paper but I remember thinking his evidence was weak.”

I agree with Luke, and I think absorbing the gist of a situation is probably the best strategy. Often I read by asking myself, "How do I feel about this?" Then the feeling is what I remember. Of course, when emotion is involved, memories are retained better, so it may not be a bad strategy for remembering details as well.

In general, whenever I read something, I look for the emotionally salient features of the content. How does this relate to reducing suffering? How does this expand my understanding of the way the world works? What implications does this have for my opinions on consciousness, interpersonal comparisons of utility, or other instances of moral uncertainty? A main reason many students dislike school is that they can't see the relevance of what they learn to something they care about. When you're aiming to improve the world, almost everything you learn is somehow relevant to things you care about, and hence the process of reading as if you were at an art gallery ("how do I feel about this?") can be applied almost universally.

Appendix: Why can you only promote new media?

It seems that a lot of movies, TV shows, music, etc. that are produced now are no better than those that have already been made. Indeed, many of them are just repeats of the same ideas over and over. Due to natural variation, some of the older productions are better than the newer ones, and maybe the older writers put more time and thought into their creations.

So why is it that most people focus on "new" movies, shows, etc.? Why not optimize for the best ones of all time, maybe with some randomness to make sure you do enough sampling of under-rated ones? Similarly, rather than making a new movie that's basically the same as 10 other previous movies, why not just rebrand the old movies and make them "the new hot thing"? It seems like we need something to be "new" to be interested, but the best things have mostly already been done. There's so much material out there that no one could possibly watch it all anyway. So why do we need more?

Part of the reason for focusing on new media is that movies, TV shows, etc. are to some degree about providing conversation topics and connection with other people and not just the content itself. It helps to have a way to decide what the current fashion is. But fashions could be set by marketers without needing to have the content itself be new, just as old-fashioned clothes sometimes come back into style.

Another reason to need new media is the suspense factor -- if the product is new, you can keep people waiting for it. This is somewhat puzzling. Why would I have suspense waiting for a new product when I can watch a thousand old things that people also had suspense waiting for? It's not like the new thing is that much better. Maybe we just like to be teased by media producers but can't be teased with media that are already accessible.

I find a similar and unfortunate pattern with blog posts: People love to read a blog post just when it comes out, but they less often explore the archives of a blog years back. Maybe this is because blogs typically have date organization that makes older articles less visible. But I think there's also some sense that the older posts are "out of date." I explicitly organize my writings in a non-temporal fashion so that high-quality older articles are not lost. Thanks to Toby Ord for first bringing up this trend with blog posts vs. static articles in conversation.

Appendix: News values are not optimal for learning

Summary: News is a peculiar institution. It purports to cover important issues of the day, yet it systematically misses some of the world's most important problems that occur every day. This appendix discusses "news values" -- the criteria that determine whether a story is newsworthy. I suggest some pros and cons of news relative to other ways of learning. I think news has a place in society but not such a prominent place as it seems to hold in intellectual life.

When I was in 10th grade, I wrote a research paper for my English class evaluating the economic and environmental effects of the 2002 Farm Bill. My teacher thought the topic was so important and my paper so well written that I should submit it to a newspaper for publication. I curated a shortened form of the paper into an editorial column consistent with the style of the local newspaper, which I read daily and with whose tone I was familiar. I submitted it for review in early 2003, and eventually, I received a call back from the newspaper.


"Hi, is this Brian?"


"I'm calling from the local newspaper about the article you submitted. I thought it was well written and a very strong piece, but the problem is, it's not ... current. The Farm Bill is something that happened last year. So I'm afraid we can't publish it."

"I see. Well, thanks for looking at it."

I also tried submitting the piece to a number of other current-affairs publications without success.

What are the biggest stories that news reporters aren't covering? The stories that don't qualify as news: The systemic problems and long-term suffering that aren't flashy enough to become headlines.

Take a look at Wikipedia's list of "News values": the characteristics of a story that tend to qualify it for news coverage. Some examples:

  • Frequency: Events that occur suddenly [...]. Long-term trends are not likely to receive much coverage. [...]
  • Unambiguity: Events whose implications are clear make for better copy than those that are open to more than one interpretation, or where any understanding of the implications depends on first understanding the complex background in which the events take place.
  • Personalization: Events that can be portrayed as the actions of individuals will be more attractive than one in which there is no such "human interest."
  • Meaningfulness: This relates to the sense of identification the audience has with the topic. "Cultural proximity" is a factor here -- stories concerned with people who speak the same language, look the same, and share the same preoccupations as the audience receive more coverage than those concerned with people who speak different languages, look different and have different preoccupations. [...]
  • Consonance: [...] the media's readiness to report an item. [...]
  • Time constraints: [...] strict deadlines and a short production cycle, which selects for items that can be researched and covered quickly.

These are hardly criteria for optimally selecting topics to learn about! Rather, as one might expect, these values are more optimized for entertainment by the audiences, combined with feasibility to produce a story quickly by the media outlets.

But why would I read a hastily prepared story about what happened this morning instead of reading a more careful, accurate, and complete historical account of what happened five years ago in a Wikipedia article? What exactly is the urgency of finding out what happened today? This obsession with the latest-breaking stories and forgetting about stories from a few months ago is peculiar. Perhaps there's some psychological basis for it; even I feel some sense that current news can be more "exciting" than old news. Maybe it has to do with being kept in suspense about the outcome? But in that case, why wouldn't more novelists release their books one chapter at a time to force audiences to wait to hear the ending? Or maybe it has to do with the fact that when you can act on news, you really do need the latest information. Last week's stock price isn't that informative, nor is the fact that your tribe leader caught a saber-toothed cat last fall. In the ancestral environment or local communities where news is directly actionable, getting the latest updates matters a lot. However, most news that people consume these days is not immediately actionable but rather serves to update their long-term views of the world.

If you want a silent, systemic issue to be covered in the news, you have to make it into a short-term story -- by carrying out a protest, or publishing a new book, or at least making some tangential connection to a topic that's already in viewers' minds. Among other things, this incentivizes sensationalism and publicity stunts in order to get air time. The actual importance of the story to the world is not necessarily a dominant consideration. Of course, this is probably not a conspiracy by the corporate media to hide information from the masses; most likely it's a reflection of the evolved interests of tribal primates. Within your 150-person band of hunter-gatherers, gossip about who's having sex with whom and who got in a fight with whom are among the most important things for you to learn about from the standpoint of passing on your genes.

To its credit, some news criteria make sense. There is reason behind focusing more on unusual stories because those do more to update your world models about what kinds of things can possibly happen. Of course, it's important not to confuse news for being a representative sample of occurrences in the world, or else one's picture will be rather skewed. You might come to believe that every politician is corrupt and every celebrity has a drug problem, and that shark attacks and plane crashes are significant risks.

News also has the virtues of concision and diversity: It doesn't dwell on a single story for too long, with some exceptions for ongoing trials or scandals, and it does report about a wide variety of events, especially when you include business news, science news, etc. Concision and diversity are both important for preventing oneself from getting stuck in a single, suboptimal topic for too long. Yet these same virtues can be fulfilled by reading a broad sampling of Wikipedia articles, with the added benefits of greater accuracy, completeness, and historical perspective than one gets from news stories.

I don't think news, especially in balance with other sources, is dramatically lower-quality than Wikipedia. Probably reading a good news article is almost equally useful as reading a good Wikipedia article. There's high variation in quality of news, with some articles providing among the best information you can consume at a given moment, and others providing little more edifying than cartoons (or maybe less). But I do think it's a shame the way society privileges news over foundational learning that's not current but is ultimately more important.

We conduct foundational research on how to best reduce suffering. Our goal is to identify cooperative and effective strategies to reduce risks of dystopian futures.

Get involved