
Robin Hanson - The Long View and The Elephant in the Brain
What this covers
Dwarkesh Patel interviews Robin Hanson, an economist at George Mason University, on why no existing unit of organization—individual, firm, or nation—reliably pursues long-term interests, and why the motivations driving most human activity remain hidden from consciousness. Hanson grounds his argument in evolutionary biology: individual organisms discount the future sharply because they share only half their genes with descendants, creating a built-in 50% discount per generation that matches observed human behavior. Organizations fail to accumulate resources over centuries because the humans who run them seize capital for immediate use, and law will not enforce perpetual saving contracts. The conversation ranges across how these constraints play out across institutions and social life, from academia to paternalism to electoral politics.
The second thread concerns hidden motives. Hanson proposes that the conscious mind functions not as a decision-maker but as a "press secretary"—retrofitting explanations for behavior that is actually driven by status-seeking and loyalty-signaling. Conversation, medicine, paternalism, and academic publishing all operate largely as vehicles for demonstrating competence, affiliation, or dominance, while participants publicly rationalize these actions as information-sharing, care, or truth-seeking. He distinguishes dominance (coercion through threat) from prestige (admiration earned through achievement) and shows how modern institutions rhetorically adopt prestige while structurally operating through dominance. The interview moves through specific domains—why academics optimize for impressiveness over usefulness, how political ambiguity serves coalition-building, why remote work may unlock global productivity gains, and what meditators and effective altruists claim versus demonstrate—before closing with advice on skill-building and institutional reform.
Hanson argues that no existing unit of selection—neither individuals, organizations, nor nation-states—reliably takes the long view because each is ultimately controlled by short-horizon biological agents, but evolutionary pressure will eventually produce units that collect resources and promote themselves over the long run; meanwhile most human behavior is driven by hidden, status-and-loyalty-signaling motives that our conscious minds rationalize after the fact.
- Human discount rates (~factor of 2 per generation) trace to genes only being half-shared by descendants, so no biological agent truly values the distant future
- Organizations cannot save resources long-term because the humans running them grab the resources, and law refuses to enforce perpetual saving contracts
- The conscious mind functions as a 'press secretary' justifying actions, not a 'president' directing them, so most human activity (conversation, medicine, paternalism, academia) serves hidden signaling motives
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Dominance is the most common organizing principle across all social animals (e.g., the chicken pecking order), and social groups evolve to make status-striving serve the group: babbler birds gain status by watching for predators (risky to themselves, useful to the group) and high-status birds force-feed resisting low-status birds, so the status hierarchy becomes the mechanism by which the group shares food and organizes cooperation in each participant's self-interest.
“social animals all through history of biology have organized themselves within the social group via dominance... it's say the chicken's pecking order”
Young people should invest in general, long-lasting skills (writing, talking, basic analysis, understanding how social groups work, coordinating with people, social skills like convincing associates you're with them) rather than short-term faddish investments like the latest programming language or toolkit, because skills collected when young are used across an entire varied and changing life; also, life is long, so a current screw-up is rarely permanent—Hanson started his PhD at 34 with young children.
“early in life you want generic skills... that's a pretty short term investment you wanna in early in life you wanna be making the long-term investments sort of learn about learn how to write learn how to talk”
Humans have a separate, socially acceptable kind of status called prestige (being admired for being the best at something, while self-effacing) versus dominance (making people obey through threats), and much human hypocrisy consists of pretending dominance is prestige—e.g., bosses, who in modern workplaces must give orders and control pay (double dominance), avoid direct orders and instead project prestige (impressive degrees, height, articulateness, stamina) so subordinates can obey while telling themselves they are merely deferring to prestige.
“in humans we have the separate kind of status called prestige and that's the one it's okay to have... whereas we're not supposed to be acknowledging or seeking or accommodating dominance”
Innovation depends fundamentally on change happening at all—not on the motives behind it: the key process is that things get tried (for whatever reason), and occasionally one works well and people keep it; deliberate institutional reform by people seeking improvement is a smaller, less important process than the simple fact that things sometimes change, and many institutions (the US Constitution, standard business practices) fail to improve mainly because they don't change.
“the key to innovation is just that sometimes things change when you don't change you can't innovate”
Even when biased (e.g., after reading The Elephant in the Brain you'd be skeptical of your own importance-justifications), actually trying to evaluate a question beats not doing it, because—like the bear-in-the-woods parable (you don't need to outrun the bear, just the other person)—success in many areas only requires being better than how others are doing it, not being objectively great.
“even when you're broken and biased like actually trying to look at a question beats not doing it”
Paternalism (treating other adults, or one's children, as parents treat children, ostensibly for their own good) is often less about helping and more about being dominant: studies show kids who learn an instrument get no average academic, cognitive, or discipline benefit, so when parents push instruments or bedtimes 'for the child's good,' the real drivers are often the parent's status (bragging, prestige-by-association, child's class advantage) or the parent's own convenience (alone time, avoiding a cranky morning that affects the parent), which parents are reluctant to admit.
“this study that said well if you look for academic or cognitive benefits like they're better at school or more disciplined... that's just not there on average kids who play an instrument are not getting any of those benefits out of it”
If you only care that the future be big regardless of who is in it, then economic growth is the right thing to promote because growth literally measures how much stuff is happening; but Hanson does not predict that kind of preference will win, since wanting a big future indiscriminately is not stable to competition—biological organisms take a long view to promote themselves and create descendants, not to maximize total biomass.
“if you only just care that everybody's happy in the long run... then you just want the future to be big right and economic growth is the future being big”
The mere fact that a group (Christians, effective altruists, meditators) claims to be more truth-oriented or less deluded than others is not evidence that they actually are, because everyone likes to identify as the truth-seekers; concrete evidence is required, and the thing people are usually happy to say they were 'deluded' about is simply not yet having joined the new group they now belong to.
“just everybody likes to think of themselves as the people who are into truth and who are less diluted than the other people so i just can't take people at their word”
Dominance is not one party imposing while everyone resists; humans constantly negotiate relative status in interactions and usually reach an unequal but mutually agreed arrangement (awkward interactions occur when status isn't yet agreed), and the submissive party actually likes being submissive in many ways they won't admit—evidenced by BDSM, where submissives also enjoy it.
“humans are constantly negotiating a relative status in their conversations... when you see people awkwardly interacting that's often when they have not agreed on their relative status”
Human discount rates of roughly a factor of two per generation can be explained evolutionarily: genes can only save resources via the whole organism, and descendants share only half one's genes, so passing resources to the next generation carries a built-in 50% discount, producing a per-generation discount factor of about two that matches observed human discount rates.
“those descendants only have half as many genes and so that means that uh your genes have a you know choice... that produces a discount factor of once of generation a factor of two and that's in the ballpark of what actual human discount rates are”
Because rates of return on assets have historically exceeded the economy's growth rate, any organization that chooses to grow assets will grow as a fraction of the economy until it dominates it, at which point interest rates would fall to the growth rate; this opportunity has long existed but no one has successfully built an organization that persistently collects resources and grows faster than the economy.
“interest rates have been higher than gross rates... any organization that chooses to grow assets will grow as a fraction of the economy until it comes to dominate the whole economy and then at that point interest rates would fall”
We are unusually good at sensing hypocrisy and false motives in rivals (and even over-attribute delusions to them because we are eager to trash rivals), but we have a much harder time noticing hidden motives that we all share—e.g., if everyone is deluded about medicine being helpful, that won't surface in rival-accusations, which only work by appealing to commonly shared values.
“when we have rivals we are unusually able to sense when they might be hypocritical... what we have a harder time doing... is noticing our own things and things in common”
The conscious mind is not the king or president directing behavior but a 'press secretary' whose job is to watch what is going on and produce explanations that make you look good—especially defending against accusations of norm violations—so your conscious job is not to know why you actually do things but to have good reasons to explain them.
“your conscious mind is more a press secretary its job is to watch what's going on and come up with good explanations that make you look good and in particular that protect you against accusations of norm violations”
Much signaling is untargeted (muscles, voice, wealth cost nothing extra to display to additional observers), but loyalty signals are inherently targeted/narrowcast at particular people; for example medicine is largely a loyalty signal—spending money to help someone get medical care signals you care about them—and such signals retain weight even among people who know each other well, especially in high-stakes moments like serious injury.
“loyalty signaling matters about as much as ability signaling... medicine is largely a loyalty signal that is you spend money to help somebody else get medical care and that's a signal to them that you care about them”
Signaling became more important, not less, as communities grew: in a 30-person forager band people knew each other for life so any single act only slightly adjusted a solid impression, but in larger farming villages and the modern world people must judge near-strangers from a small number of crude clues, so individuals have a lifelong incentive to make those clues (muscles, voice, wealth, credentials) look good for the moment they are judged.
“once you moved say to a village-sized world for the farming world of a thousand people now you didn't know everybody that well and now any one thing somebody did is much more of a potent signal about them”
Casual conversation is not primarily about sharing information, despite that being the stated motive; the puzzles (we don't track information debts, we're more eager to talk than listen, topics bounce randomly and we must always have something relevant to say) reveal that conversation is a game of showing off the contents of one's 'mental backpack'—testing how useful, full, and impressive each other's store of relevant tools is, as a proxy for future helpfulness as an ally.
“if we were exchanging information first of all we'd be keeping track of some debts... we'd be less eager to talk and more eager to listen”
We talk more than we listen in conversation because we are usually more interested in improving others' impression of ourselves than in evaluating them: we typically already have a good-enough gauge of others, judgments often involve a threshold (are they above/below 'good enough') that is rarely near the border, but for ourselves there is no threshold—we always want to be rated higher because others may talk about us to third parties.
“for each person there's a threshold of whether they're good enough for you... but for yourself though there's no threshold effect you just always want to be better”
Being clearly articulate can harm a politician because political success requires compromise positions that sit in the middle of large coalitions; clear, simple, absolutist positions are easier to express and inspire with but alienate voters who want the compromise, so politicians are purposely ambiguous and speak differently to different audiences, with 'articulate' meaning an emotionally evocative but deliberately imprecise style.
“politicians uh if they were very clear about their positions that would be a big turn off to a lot of people for whom the clarity is not the position they want so if they can be vague the vague words can encompass a lot of possibilities”
Academia primarily optimizes for impressiveness, not originality or usefulness: journal referees evaluate papers mainly by difficulty (hard methods, hard-to-obtain data, polish, absence of flaws) rather than importance or originality, because students, reporters, and funders all want impressive professors—so it is unsurprising that one can be far more original or useful than typical academics, but doing so earns little reward.
“the referees who are looking at it do not ask how important this is or how original this is or how useful this is they ask overwhelmingly how impressive is this does it use difficult methods”
Firms are not pure profit maximizers: Hanson lists 20-odd things firms do that don't maximize profit, concentrated where politics gets in the way, so a better first approximation treats a firm as a battlefield in which political coalitions fight, and the rules/procedures adopted favor each coalition's conflict position rather than the firm as a whole—e.g., a coalition may discourage its members from telecommuting because absent members lose influence in meetings.
“the first approximation of firms is as a foreign battlefield in which different political coalitions fight and the rules and procedures that will be adopted will be the ones that favor each coalition”
Humans are unique among animals in having norms—rules of what people should/shouldn't do enforced collectively, where seeing a violation obligates you to appeal to others and discuss collective punishment, and failing to do so is itself a violation; norms preceded law (until ~10,000 years ago they handled all interpersonal problems) and remain a human 'superpower,' with law added for tasks norms can't handle.
“humans are unique compared to other animals and having norms that is rules of what people should or shouldn't do that we enforce via collective groups”
Trump's 2016 win was a credible loyalty signal: a group felt ignored by cultural elites, and Trump deliberately chose cultural styles unlike those elites (going out of his way not to project high-class signals) and provoked elites to yell at him, because making the out-group denounce you is the best way to credibly signal 'I'm one of you, not one of them.'
“trump went out of his way to not project high class... signals his winning in 2016 was basically saying there was a group of people out there who felt that they were being ignored and he basically credibly signaled i'm one of you”
Remote work will become a very big deal over the next ~30 years (not 5–10) because the under-appreciated draw is planet-wide agglomeration: big cities are more productive because more interaction partners enable more specialization, and remote work could make the entire planet act as one big city, allowing currently-local jobs (hairdressers, plumbers) to be specialized globally like Apple/Google engineers, an enormous productivity gain that will be worth paying the costs of overcoming the political obstacles (lobbying, alliance-forming, reading loyalty) that have so far blocked it.
“the huge draw is agglomeration cities bigger cities are much more economically productive than smaller cities... remote work allows sort of planet-wide specialization... it can be as if the entire planet was one big city”
People deviate from optimal behavior when accounting for the long term: a true long-view agent would adopt a very low discount rate, and since resources are generically useful regardless of which future obtains, anyone who discounts the future and ignores resource-collection reveals they are not actually interested in the long-term future.
“if you have the long view you would take a very low discount rate that is you you would think the future was just as important”
Nation states are not long-view units because the people who run them spend the resources: in democracies voters elect politicians who spend accumulated resources on the present, and even dictators set things up for themselves or their children rather than for indefinite organizational resource-accumulation.
“in a democratic nation state for example if the country collects a bunch of resources then the voters go tasty and they can vote in a politician who will vote to spend that”
A meme cannot promote itself alone, just as a single gene cannot; genes promote themselves only via organisms (large collections of genes that can plan, store resources, and make strategies), and similarly memes promote themselves only via organizations, so predicting that a unit will successfully plan to promote itself requires a unit capable of planning, not a bare meme.
“genes really don't promote themselves by themselves genes promote themselves via organisms... similarly for organizations you know they have to be have to have a lot of parts that all work together”
Analyzing motives is more robust when focused on distal (fundamental) causes rather than proximate ones: proximate causes (the conscious thoughts just before a choice) are complicated and vary across time, place, and culture, but distal forces are simpler and more stable, so finding the common element behind a behavior across history should take priority over explaining cultural variation.
“in our book... we're focused more on distal motives... those are complicated and context dependent and in many ways i think it's more robust and even easier to focus on distal causes”
In politics—and in organizational promotion generally—loyalty signaling takes second place to nothing and often outranks ability: a manager or CEO selecting subordinates for promotion commonly values demonstrated loyalty over competence, so one must show loyalty before showing ability.
“honestly ability takes a second place to loyalty in politics and in many other areas of of leadership even for a manager or a ceo often when they are looking for subordinates for promote loyalty matters more to them than ability”
Successful startup founders are 'overdetermined' selections from multiple tails: they are extreme optimists (most startups fail, so you need a weird expectation you'll succeed where others failed), contrarians (a startup is a contrarian bet on a different business concept), generalists (you must do everything, not specialize), and systems thinkers (you need an abstract systemic theory of the industry)—and Asperger's correlates with several of these (systems thinking especially), so the founder-Asperger's link is real but not necessarily from any single dominant cause.
“the most fundamental thing about a startup person is they're just on the tail of a distribution of optimism... it's got to select for contrarians... a startup is just you have to do everything you can't just be a specialist”
The concern that remote work increases unemployment because the best 10% of plumbers will do all plumbing while the rest are jobless is exactly the desired outcome of a specialized economy: we want the best people doing each task and the others to go do something else, which is the nature of specialization and what we should want more of.
“that second thing is exactly what we want and exactly what we get in other areas yes we want the best people to be doing it and the other is to go do something else”
The vast majority of innovations in any domain (medicine, etc.) are bad, but net progress only occurs if you try things and discard most; so while the conventional praise for making the US Constitution hard to change is partly valid (more ways for things to go wrong than right), you still need enough trial—ideally with good incentives (venture-capital-style selection beats a government agency) to pick better ones—rather than too little or random change.
“the vast majority of innovations are bad... nevertheless we're only going to have a net change if we try things out so we both have to try a lot of things and expect most of them will be bad”
Even when you try hard not to be random, you are in fact fairly random because there is so much you don't know and so much context you fail to account for, so your best guess is still pretty random—which means it's easy to be overconfident; you still want the genuinely best experts unless their overconfidence is so severe it destroys their advantage, in which case picking random people in their role can be better.
“even when you're trying not to be random you are in fact being pretty random... there's a lot of things you don't know there's a lot of context you're not taking into account”
Concerns that remote work kills the serendipitous innovation of in-person city interactions are answerable: in the past most social interactions (office chats, after-work drinks, birthday parties, talks) happened in person and the minority of online interactions couldn't fill all social roles, but as more interaction types move online—birthday parties, drinks, business meetings and their follow-up chats—those online contexts can themselves become the basis for serendipity, new conversations, and innovation; this just requires work to make online interactions natural and satisfying, which is why it takes ~30 years.
“as we do some more things online then those things can happen online instead if you have the birthday party online then you'd be chatting with people at the birthday party online”
We are still early in the multi-century process of learning to optimize large organizations: two centuries ago there were essentially no large organizations (even nations were run by small ones), and figuring out how to set up organizations that actually maximize profit (rather than letting coalitions pursue their own interests while claiming profit-maximization) could take several more centuries at the current rate.
“two centuries ago there were basically not large organizations they didn't exist almost everything was done by small organizations... we are still a long way from optimizing organizations”
Nerds may generate more original social-science insights because socially-skilled people glide intuitively through their social world and never notice contradictions between the social theories they espouse and their actual (e.g., dominance-seeking) behavior, whereas nerds, lacking that intuitive smoothness, build explicit theories about social life that come into conflict with one another and with observed behavior, giving them more chances to notice when a theory fails.
“most people who are socially skilled... will talk about how you know we're all egalitarian here at this company and then they will be dominant in their activities... and they won't notice the competition contradiction”
Most governance and most innovation should happen at the local level (church, homeowners' association, firm), not at glamorized national/international bodies (UN, World Bank), because national/international governance is heavily lobbied (so any individual has little influence) and is a poor venue for innovation; instead, innovate in small organizations and have larger ones copy the practices that work.
“most governance happens at a local level your church your home owners association... and most innovation should happen at that level too... then the larger organizations will copy the things that work at smaller homes”
Organizations have not become long-view agents that accumulate resources because they lack separate existence: they are tools that genes and individuals use, and whenever an organization (church, firm, club) collects resources, the humans running it eventually grab those resources for themselves, so individuals and genes 'rebel' to bring the organization back in line.
“if a church collects resources or a firm or club well later on some humans are in charge of that organization then they just take the resources they just grab them”
Hanson predicts that eventually some unit of selection will emerge that takes the long view with respect to itself, collects resources, and promotes itself; once enough such units exist they will drive interest rates down to no longer exceed growth rates, and the world will be full of creatures who take the long view, so the long term will no longer be neglected.
“i predict that eventually there will be some sort of unit of selection that will take the long view that will be able to collect resources... and then long term will not be neglected”
If you take a long view but do not act to promote yourself, then whatever in you that produces the long view is dying out—you are not taking the long view about the long view itself; a long view that lasts requires focusing on preserving and promoting the thing that embodies the long view.
“if you take a long view but you don't promote yourself well whatever it is in yourself that makes you take the long view that's dying out you aren't taking the long view on the long view”
When forecasting, the easiest things to predict are the expected ones—items on the long list people have been predicting for decades—where the task is to determine which will finally actually happen; it is easy to confuse a clear view with a near one (you can see a distant mountain clearly without it being close), so for trends like space, fusion, or telecommuting the question is not whether they'll happen eventually but when, by examining the concrete details of what enables them.
“it's easy to confuse a clear view with a long distance you know you're standing on a mountain you see across the other mountain and you see it clearly but that doesn't mean it's close”
Hanson's method for finding important work is iterative: focus on the most interesting topic, learn enough that it changes what now seems most interesting, switch, and repeat until reaching a topic nothing beats; there is no real shortcut because you only discover which topics are most important and high-potential by diving into them—but the first practical step is easy, since most people can't even explain why their field matters, so simply pushing into 'why is medicine/advertising/transportation important' and building a simple model beats not doing it.
“i repeatedly did that until i got to a place where nothing else beat that as a more interesting topic which to me meant it seemed to be very important”
Social media has not increased the total amount of signaling—roughly 80% of what we do involves signaling and has for a long time; social media merely moves signaling around to different people in different ways, just as ancestors consciously signaled by buying houses, wearing clothes, painting fences, and being polite.
“i would say you know 80 of what we do with signaling in some sense and that's been true for a long time and social media just moves it around”
Long-term resource accumulation by organizations could in principle work via constitutional contracts that mandate saving and are enforced by law, but history shows this fails because associated parties petition courts to refuse enforcement, so law specifically prevents arbitrary organizational contracts that allow indefinite resource saving into the future.
“when organizations tried this people associated went to the courts and said no no we want those resources and they push the courts to interpret law so that these con these contracts are not enforced”
Hanson is skeptical that meditation reduces self-deception because meditators seem obsessed with achieving a prestigious state of mind that functions as their pecking order, with little evidence it helps the world or even helps them much, and the opportunity cost (e.g., taking a decade off) is enormous; he would be convinced only by a book that goes area-by-area listing the specific false beliefs (the 'a's' people believe vs the 'b's' that are really true) the way The Elephant in the Brain does.
“whatever the state of mind they seem to get into i'm not seeing much evidence that it's like this thing that helps the world a lot... what i see is that it's very prestigious in their world”
More capable people tend to be both better at realizing their base motives and better at justifying them, because abilities of all sorts (IQ, wealth, good outcomes) tend to be roughly correlated across people.
“people who are more capable a tend to be more capable of b for almost all a and b that's just a general thing for people”
Having a large lay/amateur audience is essentially useless for an academic career: it doesn't get you a job, a journal editorship, or an NSF grant, because that is not what academics compete for; and best-sellers require hundreds of thousands to millions of sales, not the thousands that an academic book or modest following produces.
“having amateurs like you out there is pretty useless for an academic... it doesn't get you a job in academia it doesn't get you a journal editorship it doesn't get you a grant from the nsf”
Robin Hanson does not consider himself a famous economist; he describes himself as on the margin, lucky to have survived in academia and gotten tenure, and more famous among amateurs than among academics.
“no no i'm not a famous economist i'm on the margin i'm lucky to have survived in academia”