Helpless Individuals

Previously in seriesRationality: Common Interest of Many Causes

When you consider that our grouping instincts are optimized for 50-person hunter-gatherer bands where everyone knows everyone else, it begins to seem miraculous that modern-day large institutions survive at all.

Well—there are governments with specialized militaries and police, which can extract taxes.  That's a non-ancestral idiom which dates back to the invention of sedentary agriculture and extractible surpluses; humanity is still struggling to deal with it.

There are corporations in which the flow of money is controlled by centralized management, a non-ancestral idiom dating back to the invention of large-scale trade and professional specialization.

And in a world with large populations and close contact, memes evolve far more virulent than the average case of the ancestral environment; memes that wield threats of damnation, promises of heaven, and professional priest classes to transmit them.

But by and large, the answer to the question "How do large institutions survive?" is "They don't!"  The vast majority of large modern-day institutions—some of them extremely vital to the functioning of our complex civilization—simply fail to exist in the first place.

I first realized this as a result of grasping how Science gets funded: namely, not by individual donations.

Science traditionally gets funded by governments, corporations, and large foundations.  I've had the opportunity to discover firsthand that it's amazingly difficult to raise money for Science from individuals.  Not unless it's science about a disease with gruesome victims, and maybe not even then.

Why?  People are, in fact, prosocial; they give money to, say, puppy pounds.  Science is one of the great social interests, and people are even widely aware of this—why not Science, then?

Any particular science project—say, studying the genetics of trypanotolerance in cattle—is not a good emotional fit for individual charity.  Science has a long time horizon that requires continual support.  The interim or even final press releases may not sound all that emotionally arousing.  You can't volunteer; it's a job for specialists.  Being shown a picture of the scientist you're supporting at or somewhat below the market price for their salary, lacks the impact of being shown the wide-eyed puppy that you helped usher to a new home.  You don't get the immediate feedback and the sense of immediate accomplishment that's required to keep an individual spending their own money.

Ironically, I finally realized this, not from my own work, but from thinking "Why don't Seth Roberts's readers come together to support experimental tests of Roberts's hypothesis about obesity?  Why aren't individual philanthropists paying to test Bussard's polywell fusor?"  These are examples of obviously ridiculously underfunded science, with applications (if true) that would be relevant to many, many individuals.  That was when it occurred to me that, in full generality, Science is not a good emotional fit for people spending their own money.

In fact very few things are, with the individuals we have now.  It seems to me that this is key to understanding how the world works the way it does—why so many individual interests are poorly protected—why 200 million adult Americans have such tremendous trouble supervising the 535 members of Congress, for example.

So how does Science actually get funded?  By governments that think they ought to spend some amount of money on Science, with legislatures or executives deciding to do so—it's not quite their own money they're spending.  Sufficiently large corporations decide to throw some amount of money at blue-sky R&D.  Large grassroots organizations built around affective death spirals may look at science that suits their ideals.  Large private foundations, based on money block-allocated by wealthy individuals to their reputations, spend money on Science which promises to sound very charitable, sort of like allocating money to orchestras or modern art.  And then the individual scientists (or individual scientific task-forces) fight it out for control of that pre-allocated money supply, given into the hands of grant committee members who seem like the sort of people who ought to be judging scientists.

You rarely see a scientific project making a direct bid for some portion of society's resource flow; rather, it first gets allocated to Science, and then scientists fight over who actually gets it.  Even the exceptions to this rule are more likely to be driven by politicians (moonshot) or military purposes (Manhattan project) than by the appeal of scientists to the public.

Now I'm sure that if the general public were in the habit of funding particular science by individual donations, a whole lotta money would be wasted on e.g. quantum gibberish—assuming that the general public somehow acquired the habit of funding science without changing any other facts about the people or the society.

But it's still an interesting point that Science manages to survive not because it is in our collective individual interest to see Science get done, but rather, because Science has fastened itself as a parasite onto the few forms of large organization that can exist in our world.  There are plenty of other projects that simply fail to exist in the first place.

It seems to me that modern humanity manages to put forth very little in the way of coordinated effort to serve collective individual interests.  It's just too non-ancestral a problem when you scale to more than 50 people.  There are only big taxers, big traders, supermemes, occasional individuals of great power; and a few other organizations, like Science, that can fasten parasitically onto them.

 

Part of the sequence The Craft and the Community

Next post: "Money: The Unit of Caring"

Previous post: "Rationality: Common Interest of Many Causes"

235 comments, sorted by
magical algorithm
Highlighting new comments since Today at 6:25 AM
Select new highlight date
Moderation Guidelines: Reign of Terror - I delete anything I judge to be annoying or counterproductiveexpand_more

I'd like to see some examples of types of large institutions that you believe should exist, but don't due to lack of coordination.

I'm not sure I'm following the logic here. The failure of science to raise money via voluntary means is evidence that it is too much of a non-ancestral problem?

Well, I'll agree that if we somehow had science as it exists now for a few hundred generations, we'd probably be better at funding it. But thats true of anything. Standard economics predicts that funding large-scale public goods is difficult via voluntary means, and public choice explains why its difficult for governments too. If you believe Coase this difficulty is a feature, not a bug, because it takes transaction (i.e., organizational) costs into account.

Of course, it shouldn't come as any surprise to anyone that a scientist is complaining that people don't fund enough science ;) To be honest, I don't know where I could donate money to science to make a difference. Its very hard for non-scientists to judge the feasibility of scientific projects. So much of science seems to be a complete and utter waste of smart people and resources.

One thing we can do is promote the use of prizes over normal funding.

In conservation biology, flagship species play the role of cute puppies:

These species are chosen for their vulnerability, attractiveness or distinctiveness in order to best engender support and acknowledgment from the public at large. Thus, the concept of a flagship species holds that by giving publicity to a few key species, the support given to those species will successfully leverage conservation of entire ecosystems and all species contained therein.

This is fighting a bias with a bias: people do not care as much as they should about conservation while they care too much of cute puppies. Science in general could adapt this technique: use "popular" subjects to attract funds to "good but unpopular" subjects.

If this is too much of the "dark side" for you, umbrella species might be more appropriate.

This is fighting a bias with a bias: people do not care as much as they should about conservation while they care too much of cute puppies.

Surely this is fighting your preference with a bias. How robust do you think your argument is, not that conservation is important, but that it's more important that the other things people invest in?

As has been already mentioned, science of old was largely funded by individuals; scientists (or 'natural philosophers') were largely patrons of wealthy individuals. Presumably, this was because supporting an individual provides a good emotional fit (and thus serves as a good signal of status) in a way that funding pure research does not.

"But it's still an interesting point that Science manages to survive not because it is in our collective individual interest to see Science get done, but rather, because Science has fastened itself as a parasite onto the few forms of large organization that can exist in our world."

Can't it be both? It's not as if the government's support of science is completely unrelated to its being in our collective interest. It's awfully cynical simply to presume, without comment, that governments don't ultimately work for our collective interest (even if they can at times be very misguided).

More generally, this post seems full of lots of lines, like the above, that sort of seem true at first glance, but upon closer inspection are either quite banal, vacuous, or only questionably correct. How about some supporting evidence or arguments?

But individuals do fund science (at least in the US): individuals give pretty substantial amounts to universities, even if it's less overall than what's provided by governments.

As you sort of allude to later, the issue may be less that individuals don't fund science, as that they don't fund particular science. I would speculate that this is in no small part because we generally realize that we wouldn't be very good at picking particular science to fund, so we give general, and let other people decide exactly what projects to pursue. This opens the process up to lots of problems, but it's not obvious that it's worse than the feasible alternatives.

(In the same way, lots of people choose to give to generalist charities like Oxfam, rather than trying to evaluate specific projects for themselves, though (a) I suspect it's easier to tug people's heartstrings for charitable projects; and (b) people probably overestimate their knowledge of what works in charity more than in science.)

I recall reading somewhere that one of the reasons Harvard has so much money is that the majority of donations they receive are earmarked for a narrow range of projects and they receive more money than they can spend in those areas (while other areas remain underfunded). I can't find the article but maybe someone else remembers it. Regardless I wouldn't be so quick to assume that many people are funding universities without restrictions on how the funds must be used.

I'm not sure I buy the Harvard story (even if it had some truth to it, there are lots of ways to get around this sort of thing, and if it were ever a serious constraint I'd imagine Harvard would be pretty good at finding those ways by now). But your broader point is valid; it would be good to have actual data about this.

The Harvard story is actually true, and yes, they have found the ways around it.

Basically, one main strategy is that you invest the original donation in such a way that the dividends and capital gains go into the general endowment. Within a generation or two the problem is solved. You can do that when you're immortal.

"But by and large the answer to the question "How do people maintain the functioning of their bodies" is "They don't!" The vast majority of people - some of them extremely skilled in medicine and physiology - simply fail to exist in the first place."

Managing to misunderstand a question you yourself ask is a pretty impressive feat.

I appreciated the post, but can't deny Annoyance has a point here...

This kind of thing is often considered one of the main roles of government: funding important projects on a constant basis over a long period of time. It's hard to fund those with charity; charity funding tends to be inconsistent over time, and people who do give to charity are likely to give to whatever cause is "popular" that year. I wouldn't want to try to fund "maintenance of one specific bridge every year over the next 50 years" with just charitable contributions from people who use that bridge and benefit from it, because some years they might donate more then enough to maintain the bridge, but some years they might not. Taxation and a constant revenue stream are just a more efficient and consistent solution to the problem. That's likely true in funding long-term science as well.

In fact, this is a historical theory for how the first large governments formed. In both ancient Egypt and ancient China, the large central governments were able to build massive canal systems that dramatically improved farming and therefore the quality of life of everyone. It would have been pretty unlikely to create those through just the voluntary contribution (either in money or in labor) of people; too many people would have defected to make that a practical solution, at least over the long term.

That being said, if you're relying on governments for that, there's always the risk that the government will decide to take your tax money and spend it on something useless (in the case of Egypt, that would be the pyramids), so some kind of auditing of the government's spending and feedback to improve it is fairly important.

As I understand it, this is supposed to be one of the things government is for - to coordinate spending when trying to do these things privately would suffer from a free rider problem. I try not to talk about politics on here - mind-killer and all that - but one way to address this would be to try to improve the rationality of voters. I increasingly believe that preparing people to be participants in a democracy should be the primary function of schools.

Couldn't you greatly increase voter rationality much faster and cheaper by limiting the franchise or encouraging less rational people not to vote?

Does a concept of "less rational people" even have any sense

And most likely it would be highly damaging, I don't have any evidence with me but I'm willing to bet real money on prediction market for that, that in cases where some class of people are excluded from voting (vote rarely like young people, or don't matter much due to the way votes are counted), their interests are seriously underrepresented in government decisions, relative to interests of classes of people who are voting a lot, in a way that matters (old people, regional interests in first past the post systems).

If you found a way to identify less rational people that didn't exclude any groups that were identifiable enough for government policy to discriminate against them, it could work, but that sounds hard.

Limiting the franchise sounds tough but there are some obvious ways to reduce the number of less rational people who are voluntarily choosing to vote: reduce the number of get-out-the-vote campaigns, specifically those that mindlessly repeat slogans about voting, and focus more on campaigns with a rational explanations of why people ought to vote Of course... perhaps get-out-the-vote campaigns tend to have more of an impact in demographics that vote less on average, like young people, so they might have other beneficial effects in that way (and given that probably the portion of potential voters who aren't swayed by mindlessly spouting slogans about voting is vanishingly small, this would probably dramatically outweigh any effects of them on voter rationality). Still, it's a thought. Get-out-the-vote campaigns just seem INCREDIBLY weird to me.

Get out the vote campaigns are often just excuses to get out the vote for groups that are likely to vote according to the organizer, if for no other reason than that a particular neighborhood might be inclined to vote one way.

There aren't enough of them. Obvious problems would result.

Just the weekend before last I was hearing a scientist say "we have 30 projects we know, know, will give good results. That's not even counting the worthy speculative work. We can fund: maybe five."

Science as an institution is absolutely poverty-bound. Find a way of getting it a bigger bite of GDP, and the speedup could be immense.

Find a way to give it a bigger bite of GDP, and new bureaucrats will arise to seize the money for worthless projects, generating more noise in the journals? We could be in a situation analogous to the situation in some countries where giving money to beggars just reallocates more of the economy to begging without increasing the income of the average beggar.

No doubt some of the marginal money would be wasted, but that's always true and is true now. Science is and would be worth it even if the haircut was immense, and I don't see a reason that the additional spending would be that much more wasted.

Also, the begging scenario you describe isn't particuarly scary. If giving more money to scientists meant there were more scientists each with the same funding levels we have now, that seems like a perfectly fine outcome. If it meant there were more fundraisers seeking money for science and each raised the same quantity of funds, that also seems like a fine outcome.

I'm confused a bit. You see existing science funding as being mostly/entirely wasted on worthless projects?

Could you elaborate here?

Are you saying you think the impact of a higher portion of GDP spent on "science" would be negative, more or less neutral, or positive (but with diminishing returns)?

Do you think that a marginal dollar that is allocated to research on, for example, the synthesis of conductive plastics increases or decreases the signal:noise ratio? What fields do you think would have a marginal output of less than $1 of value generated per dollar you invest?

Mostly fields which are producing lots of "statistically significant" results and no universal generalizations (where "This happens to everyone with gene X" is a universal generalization even if not everyone has gene X). Conductive plastics doesn't sound too bad because you can tell whether or not a plastic conducts pretty clearly.

You're only looking at funds for "Science" that get used for something useful. How much of the funds are used suboptimally or completely wasted? Pretty much every funding (science, charities, government etc.) except for capitalist free market (and even that only in case where there's little potential for abuse) is extremely inefficiently spent, as there's no optimization mechanism based on results, so it's optimized for some very indirect proxies (number of publications, emotional appeal, political interest). Basic research almost by definition doesn't have anything to directly optimize on. So no - just giving "Science" more money wouldn't necessarily improve everyone's well-being.

I cannot think of any plausible mechanism how basic research can be funded in a self-optimizing manner. Prediction markets on its long term impact? That's the best I can think of, but considering how unproven real world prediction markets are even in far easier cases I wouldn't really have high hopes for that.

There are optimization pressures from peer review and academic jostling for position. My point above was that things which passed all the quality checks were being stopped for no reason except lack of money. If you expect any good from the existing research, you should expect those unfunded efforts would have been at least as good, giving a linear speedup per added money.

The market optimizes for the aggregate desires of the buying public and for the ability to produce something valued more by the public than its raw materials - if it achieves scientific usefulness, that's a byproduct.

And I suspect a great deal of scientific research is done at companies which then keep it secret. That work isn't wasted, but it isn't doing the kind of good that shared information can.

In addition to the emotional issues you raise, there's the question of thresholds and scalability. If the puppy program already exists, giving $10 will help more puppies. But, for many scientific research projects, there's no point in even starting with less than $100K in hand. That could be $10 each from 10,000 people. An easy decision, perhaps, for the 9999th person, but who wants to give the first $10?

Elsewhere I've suggested "Social Escrow" as a solution. You pledge a certain amount, contingent on enough other people doing so and perhaps on other objective criteria. "Send us two checks. We'll tear up both if not enough other people send checks. We'll tear up the second if the research doesn't meet kilometerstone X by date Y."

Kickstarter has some of these features, but doesn't seem to fund science.