Funding science is actually a badass thing to do

OR: The Three Sirens of Entropy

photo cred: my dad

It’s a grim time for science in America. The National Science Foundation might be forced to fire half its staff, grants are being frozen and reviewed for ideological purity, and universities may see their cut of those grants reduced by 40%.

We were bound to end up here sooner or later. Science funding has been riding an atomic shockwave since 1945, buoyed by the bomb and the Cold War and the conviction that we could vanquish our enemies if we just kept cutting checks to the geeks. Now the specter of Soviet submarines isn’t so scary anymore, and our most feared and hated enemies happen to share the same country with us—we’re out of our Red Dawn era and into our Civil War era. And so people are wondering: why are we spending billions on basic research, again?

There’s a good answer to that question, but nobody’s giving it. We all assumed that science was self-evidently worthwhile, thus allowing our arguments to atrophy and leaving us with two half-assed defenses. On the one hand, we have romantics who think science is important because “something something the beauty of the universe triumph of the human spirit look at this picture of a black hole!” And on the other hand we’ve got people who think science is like having a Geek Squad that you can call upon to solve your problems (“My tummy hurts! Scientists, fix it!!”). These points aren’t completely wrong, but they’re achingly incomplete—why should we pay for people to stand agog at the wonder of the universe, and why would we let them do anything that isn’t immediately related to some pressing problem?

And then there’s an alarmingly large contingent who thinks there isn’t any argument to make. In their heart of hearts, they think the NSF and the NIH are, in fact, charitable organizations. In this view, science funding is just welfare for eggheads, and scientists are a bunch of Dickensian beggars going, “Please sir, can you spare a few pence so I can run my computational models?” Witness, for instance, the Johns Hopkins cardiologist who thinks that all NIH-funded scientists owe an annual thank-you note to the American public.¹

Even the folks who have a soft spot for science often think of it as a nice-to-have—you know, let’s first make sure we build enough battleships, mail enough checks to the right people, etc., and then if there’s a little left over we can toss a few bones to the nerds. If we’re all really good, we can have some science, as a treat.

If this is all we got, it’s no wonder that people wanna smash the science piggybank and distribute the pennies to their pet projects instead. The case for science should be a slam dunk; we’ve turned it into a self-own. So lemme take a crack at it.

KNOW WAY

There is only one way that we improve our lot over time: we get more knowledge. That’s it. Everything that has ever made our lives better has come from collecting, cultivating, and capitalizing on information.

When I say “knowledge,” I mean everything we’ve figured out, from the piddling (“cotton is more comfortable than burlap”) to the profound (“energy cannot be created or destroyed”). I mean the things you have to learn in school (“kidneys filter your blood”) and the things that now go without saying (“it’s better if the king isn’t allowed to kill whoever he wants”). Some of this knowledge looks science-y, like when engineers use the theory of relativity to make GPS work, and some of it looks folksy, like when a lady on Long Island invents a little plastic table that prevents pizza from sticking to the top of the box.

KNOWLEDGE

No matter where it comes from, it’s all part of the same great quest to de-stupify ourselves. If you’re doing anything remotely good—basically, unless you are shaking down local businesses for protection money or blowing up UNESCO World Heritage sites—you are either enabling, conducting, or cashing in on the search for knowledge, and therefore you’re part of the project.

We don’t talk about the history of our species this way. In school, I learned things like “the Egyptians made hieroglyphics” and “the Romans did aqueducts” and “the Pilgrims wore hats,” but nobody mentioned that if I had been born an ancient Egyptian, a Roman, or a Pilgrim, there’s a 50/50 chance I would have died before I turned 15. I might have starved or frozen, or maybe I would have been executed for believing in the wrong god, or maybe I’d be done in by microscopic invaders that I didn’t even know existed. (Making it to your quinceañera was once a much greater reason for celebration, not that anyone but the king could afford streamers and cake.) And if I wasn’t dead, I would be working—farming, shepherding, child soldiering, etc.—not sitting in social studies making dioramas.

Nor did anyone explain why I got to have a different life than my ancestors did: I was born into a society where we know more. We know how to grow enough food for everybody, how to keep the cold out, how to fend off the microscopic invaders, and how to get along—more or less—with people who worship different gods.² None of this happened by chance, and none of it came for free. In fact, for most of our history, our stock of knowledge didn’t increase at all, and the people who dared to add to it were often ridiculed and sometimes killed.

Despite all that, we have made a lot of progress in the millennia-long project of diminishing our ignorance, and that’s why I get to eat focaccia and play Call of Duty, while my ancestors had to eat moldy bread and play a game called “hide from the marauding Visigoths”. But the project isn’t finished yet. It’s not even close. There’s still so much suffering, and we could escape it if only we knew how.

That’s why we fund science. We all pitch in to hunt down the knowledge that can’t be found any other way. We don’t seek the knowledge that will turn us a profit tomorrow—that’s what businesses are for—but the knowledge that will support a permanently better life. We do science that is speculative and strange because that’s where the breakthroughs will come from, the frontiers of knowledge where our intuitions stop working, where predictions fail, and where the things that seem sensible are unlikely to be important. We do this with public funding because it produces public goods. The things we discover are too important to be owned; they must be shared.

So yes, it’s beautiful, but that’s not why we do it. Yes, it’s practical, but not right away. And no, it’s not charity. You don’t “save” money by skimping on science, just like you don’t save money by sending second graders to the coal mines instead of the classroom. You could think of it as investment because it does pay off in the long run, but even that undersells it. Pooling our resources to discover new truths about the universe so that we can all have better lives, to strike back against disease, suffering, poverty, and violence, to reduce ignorance for the benefit of all—that’s literally the most badass thing we do.

THE THREE SIRENS OF ENTROPY

Our past wasn’t inevitable and our future isn’t guaranteed. We have to choose to keep increasing our knowledge. That choice might seem like an easy one, but we have to contend with three tempting but false arguments for choosing to do opposite, three Sirens of Entropy trying to seduce us into running the ship of civilization aground.

The first is rejection. Knowledge comes with tradeoffs—the chemistry that cures can also poison, the physics that builds space rockets can also make cruise missiles, and so on. Plus, the past always looks idyllic—as long as you don’t look too closely—and so it always seems like history has just recently gone wrong. Anyone could be lulled into believing that these tradeoffs cannot be managed or improved and must be avoided entirely, that the solution to our problems is less knowledge, rather than more. This view does require you to ignore or deny things like the near disappearance of extreme poverty, the end of child labor, the historically low rates of violent death, etc., but there will always be people willing to rise to that challenge, because the appearance of one problem will always be more salient than the disappearance of another.

The second is complacency. When the lines keep going up and to the right, it’s easy to assume that’s just what they do, and to forget that every increase ultimately comes from our expanding stock of knowledge. You can slip into thinking that our living standards rise of their own accord, that death and disease recede because we want them to, that the GDP fairy puts 4% growth under our pillows to reward us for being such good boys and girls. When you think that progress is a perpetual motion machine, you’ll see no need to top up its gas tank.

And the third is the pie apocalypse. Every time we grow the pie that we all share, we also have to figure out how to split it fairly. We get pissed off when the products of progress are disproportionately captured by the rich, and well we should—it’s like playing Hungry Hungry Hippos against someone who gets to mash two buttons instead of one. But it’s easy to focus on the pie-splitting problem and forget the pie-growing problem entirely. We can thus descend into an all-out war for pie, where they only way to get a bigger slice is to steal someone else’s. Meanwhile, the pie shrinks and shrinks, and we end up fighting over crumbs.³

EASY MONEY

So it’s a good idea to get smarter, and we can all contribute to that mission. But that doesn’t mean we’re doing a good job right now. People are right to be mad about the state of science funding in America: the fraud, the waste, the low ambitions, the dogmatism, politicking, and rent-seeking. Maybe this chaos is at least a chance to sunset some of the most outrageous parts of the system, so long as we’re committed to figuring out the best way to spend our science dollars, rather than throttling or lavishing funds the way a king dispenses dukedoms and decapitations. (“Sussex for my real friends, no necks for my sham friends.”)

Fortunately, this ain’t hard to do. There’s so much low-hanging fruit that we’re tripping over it. Here are three of the easiest, most obvious moves that we could make right away, and that we should have done half a century ago.

1. STOP PROPPING UP FOR-PROFIT PUBLISHERS

For-profit publishers make their money by privatizing public goods. The government pays the scientists to do the research, then publishers paywall it, and finally the government pays again so the scientists can read what they wrote. This gets obfuscated because publishers don’t charge the government directly. Instead, universities fork over millions for journal access, then charge it back to the government as part of the much-hated “indirect costs”. The taxpayers foot the bill for all of this, but they don’t even get to read the studies themselves unless they have a .edu email address.⁴

Everybody knows this is ridiculous. If this business didn’t already exist and you tried to pitch it on Shark Tank, Mark Cuban would laugh in your face. It only worked out this way because some schemers realized that academics are a captive market, so they bought up all the journals in the 70s.⁵ Individual scientists have tried to upend the system through self-immolation—that is, by refusing to publish in or review for journals owned by Elsevier and the like—but it hasn’t worked, because it’s hard to convince everyone to immolate at the same time.

This is the kind of coordination problem that only the government can solve. We could score such an easy win by paying for the minor costs of publishing directly (hosting, copyediting, the pittances occasionally paid to editors, etc.), rather than paying middlemen to do them at a ~40% markup. I’ve got plenty of issues with scientific journals, but if we’re going to have them, we shouldn’t also set money on fire. So if you want to smash things, smash this one!

2. DON’T REQUIRE THINGS THAT YOU DON’T WANT TO PAY FOR

By one estimate, principal investigators spend 44% of their time applying for grants. It takes an average of 116 hours to fill out a single NSF proposal. Most applications get rejected, and so most of that time is wasted. These costs don’t appear in the budget because nobody can say “if you pay me, I will spend half of my time trying to make sure that you will pay me again in the future.” But that’s what they’ll do, because failure to secure a grant is death for an academic. So whether federal agencies realize it or not, by making their applications so laborious and competitive, they are paying people to spend almost half of their time trying to get paid.

And that doesn’t include the cost of grant panels who have to sift through those applications, the bureaucracy that has to process the paperwork, the mandatory Institutional Review Boards that take six months to tell you whether it’s okay to ask someone if they own a lawnmower, etc. The government requires all of these things, and since the government is funding the whole enterprise, it also pays for them.

You might think we adopted all of these policies because we had evidence suggesting they would make us better off, but we didn’t. We adopted them because they sounded good at the time, and why would you check something that sounds good?

3. LITERALLY TRY ANYTHING ELSE OH MY GOD

Whenever people complain that a lot of government-funded science ends up un-cited and unused, or that it’s hideously ideological⁶ or pointlessly incremental, I gotta laugh because those are the projects we picked. We got exactly what we asked for; it’s like ordering something from Amazon and then being angry when it arrives. The problem isn’t the product—it’s the picking.

If we want better outcomes, we should pick different projects. The whole point of funding science is to discover things that wouldn’t get discovered anywhere else. Pharmaceutical companies can make plenty of profit turning molecules into medicines, but they can’t go looking into the mouths of gila monsters, asking “Any good drugs in here?” Thats how we got GLP-1 agonists, which are now used by millions of people.

So public funding should go to projects that are foundational, speculative, long-term, useful but unsexy, or big if true. Some of those projects can be identified by committee, but many can’t, and so we should pick them some other way: lotteries, golden tickets, trust windfalls, fast grants, bounties, prizes, retroactive public goods funding⁷, “people not projects,” and moonshots that are actually moonshots, to name just a few. We should be placing some of our bets outside the scientific consensus so that we don’t waste billions on one idea that turns out to be wrong. And we should really try to figure out how one guy funded almost every single person who won a Nobel Prize for molecular biology between 1954 and 1965...19 years before they won. It would be cool to do that again!

Many of these methods cost less than the standard “solicit one million pages of applications” procedure. If we tracked them long term—not “did this get into a good journal”, but “did this end up mattering”—we could figure out which ones work better for what ends, and we could get more science for less money. It is a national embarrassment that the agencies who fund experiments do basically zero experiments themselves. Would you trust a pulmonologist who smokes?

BYE BYE HEISENBURG

I understand why people think we can balance the budget by shrinking science, and I understand why we spend our limited science funding in such a cowardly way. We want everything to be accountable, and it’s hard to tell the difference between actual accountability and the mere appearance of it. Nobody gets blamed for “saving” money, even when it costs more in the long term. And nobody gets blamed for doing things by the book, even when the book turns out to be fiction.

But paying for science is different from paying for other things. When you pay for a bridge, you get a bridge. When you pay for Medicare, you get Medicare. When you pay for an F-35 fighter jet, you...pay an extra $500 billion, and then you get an F-35 fighter jet. In the short run, though, you can’t know what your science dollars are going to get you. That’s the whole point of doing the science!

The more we fight against that fact, the more we demand legibility in the form of applications and metrics, the more we try to squash and slash and cut just for the hell of it, the less we get the thing we’re actually trying to buy. There’s a sort of Heisenberg Uncertainty Principle at work here: you can’t spend your money wisely and make sure you’re spending your money wisely at the same time. It would be cool to only run experiments that were guaranteed to both work and teach us something, just like it would be cool to only buy stocks that are guaranteed to increase in value. Until that becomes possible, we’re gonna have to take some risks.

In the long run, however, you know exactly what you’re going to get. The only thing that lifts our boats is the rising tide of knowledge. One of the basic functions of the government is to help make that happen. It requires some patience and some money. In return, it gives us literally everything.

1. I’m fascinated by this logic. The government pays a lot of people to do a lot of things, so why are researchers uniquely indebted to the American public? Should your local police officer also send you a thank-you note? Should the Secretary of State? Do we all deserve a thoughtful box of mixed nuts from the guy who trims the bushes in front of the DMV?

2. It’s fitting that in Cixin Liu’s Three Body Problem series, hostile aliens try to ensure our defeat by halting our scientific progress.

3. This same tension between growing and splitting, by the way, is one of the core problems of negotiation.

4. The Biden administration tried to fix this by requiring government-funded research to be open-access, but they got outfoxed by the publishers, who started charging authors for publishing rather than charging readers for reading. The last paper I published in one of these journals made us pay $12,000 to make it “open-access”. Publishing my papers on Substack, on the other hand, costs me $0.

5. As one publisher recalls, “When you’re building a journal, you spend time getting good editorial boards, you treat them well, you give them dinners. Then you market the thing and your salespeople go out there to sell subscriptions, which is slow and tough, and you try to make the journal as good as possible. That’s what happened at Pergamon. And then we buy it and we stop doing all that stuff and then the cash just pours out and you wouldn’t believe how wonderful it is.”

6. Although we’re probably not funding all that much hideously ideological science, see this analysis and clarification

7. I’ve only seen very online/crypto-y people try to do this, but there’s no reason we couldn’t use it for regular stuff