American Science, Shattered. So says is the title of Part 1 of a ten part series in STAT News. Since STAT articles are usually paywalled, I will summarize them here as they appear. Unlike most accounts of the current state of science in the United States, the authors of this article have found scientists who are reflective, who realize that while we (once a member, always a member at some level) are currently on a one-way street to perdition, actions and inactions of the scientific community helped pave that right-of-way from sidewalk to sidewalk:
For a substantial group of U.S. researchers, 2025 will be remembered as the year their path to a career in science was closed off, their dreams dashed. For others, it will go down as a chaotic game of red-light-green-light that left them constantly unsure of what work would be funded or halted, but that they managed to survive. For nearly everyone, the last 10 months have revealed that the research enterprise that catapulted the country to the technological fore was much more brittle than expected.
Sure, the courts have stepped in to restore billions of dollars in terminated grant funding to colleges and universities. Yes, the National Institutes of Health, despite layoffs and seemingly endless hurdles, managed to spend its entire budget for the fiscal year. And Congress, in a rare rebuke to the president, has so far refused steep cuts to the NIH budget in 2026 as well as a White House plan to consolidate its 27 institutes. But in the larger scheme of things, the Trump administration has, with shocking speed, ripped up the longstanding social contract that existed between scientists and the federal government.
This social contract made American science the world leader in basic research in all disciplines beginning with the vision of Vannevar Bush after World War II. This has been covered here before. The realization of Bush’s vision from the mid-1950s until recently spread the scientific wealth around the country. Yes, the research universities and private research organizations that received the most money at first have remained at the top of the funding lists (National Institutes of Health, National Science Foundation, NASA, Department of Defense) but virtually other institution that is willing to support research has received support. Contrary to the apparent views of the Current Administration, this outreach into the heartland has been a good thing.
Vannevar Bush wrote Science: The Endless Frontier as a report to the president in 1945 as a plan to make American science great. It did just that:
“Science: The Endless Frontier”…is the foundational document to which the past eight decades of technological winning can be traced. If you’ve enjoyed living in an era where doctors can find cancer with scans and DNA tests, where HIV is treatable, where lasers can correct your vision, GLP-1 drugs fight your food cravings, and premature babies don’t have to suffocate and die within their first few hours, you can thank its author, Vannevar Bush. The internet? GPS? The core learning algorithms that form the basis of artificial intelligence like ChatGPT? Bush. Bush. Bush.
Plus, virtually every other drug and medical intervention we have benefited from for the past eighty years, funded by NIH, and the research in other scientific disciplines funded by the National Science Foundation.
The one rule that made American science was the requirement that research proposals be evaluated by other scientists, who pledged to make awards as objectively as possible to fund research that was “meritorious.” This, of course, can be somewhat “subjective.” Ask any scientist who has ever had a research proposal not funded (that would be every one of us). But every scientist also knows that failure comes with the life of a research scientist. As long as the batting average stays substantially above the Mendoza Line (20%), survival is possible. Based on my long experience writing and reviewing grant proposals, about one-third are worthy of support upon first submission and another one-third should be funded after revision. The others are likely to remain unfunded for the duration. That success rates are now below 20% is an opportunity cost that is incalculable but large:
Perhaps the biggest rupture arrived in August in an executive order giving political appointees sweeping new powers over the awarding of research grants. The move directly undercut a core principle of the Bush blueprint — that projects should be supported based on scientific merit, not ideological whims. While the pendulum has swung back and forth over the years between scientific independence and political influence, never has it swung this far toward the latter.
Since the end of the government shutdown last month, several moves suggest the grip of political appointees over the NIH will only grow. An internal memo, obtained by STAT, reads: “Discretionary awards must, where applicable, demonstrably advance the President’s policy priorities.” Simultaneously, the White House and top administration officials have become more involved in NIH processes than in the past, and in October a friend of Vice President JD Vance was made the head of the National Institute of Environmental Health Sciences — without a research background in that area. The day the government reopened, it also placed an NIH staffer who was a vocal opponent of the Trump administration’s handling of the agency on leave. Then in recent days, health secretary and longtime vaccine critic Robert F. Kennedy Jr. and his subordinates at federal science agencies escalated their campaign to impugn vaccine safety, bypassing normal procedures to revise website language stating vaccines don’t cause autism and assert that Covid shots caused the deaths of at least 10 children.
“My fear is that they don’t even realize what they’re doing,” said Elias Zerhouni, who led the NIH under Republican President George W. Bush from 2002 to 2008. In recent months, Zerhouni has repeatedly reached out to members of the Trump administration. In each of those conversations, he told STAT, he has found administration officials earnest in their desire to strengthen American scientific competitiveness, but lacking both a firm grasp of how the research system fits together and a cohesive plan for how to overhaul it.
“They connect this complete chaos to an intent to make it stronger, not realizing that by doing what they’re doing — and not coordinating what they’re doing in the right titration — it’s basically taking an axe to the system,” Zerhouni said. “They’re destroying it and they don’t really understand that when you tell them.”
The deal, as Bush saw it, was pretty simple. Newly created federal scientific agencies — staffed by scientists — would set scientific priorities and then vet investments in those priorities across the nation’s universities and academic medical centers. They would cut the checks and stand back, confident that scientific breakthroughs would, in time, translate into improved health, economic growth, and global technological dominance.
But it is also true that Vannevar Bush of MIT was a bit parochial, as those of us in the provinces might say. And that led to conflict with certain politicians, one of whom was Senator Harley Kilgore of West Virginia:
Kilgore wanted politicians, who would be more attuned to the desires of the taxpayers funding the research, to set priorities. Bush, who spent much of his life in the academic bubble of Boston, thought that only scientists would have the expertise to judge where money was best sent. Kilgore also thought that some funding should be dispersed geographically, which Bush saw as inconsistent with a merit-based system. (NIH listed “geographic balance” as one of the criteria in its grant strategy announced last month.) Kilgore also wanted the government to hold the patent rights on advances that were publicly funded, which Bush thought would hamper innovation. Just this year, the Trump administration has threatened to take control of patents held by Harvard that were the product of federal funding.
Vannevar Bush went behind the back of Senator Kilgore, who was not wrong on the politics. About thirty years later, the Bayh-Dole Act of 1980, sponsored by two US Senators, did great damage to basic biomedical research in the United States by formalizing the, well, theft of “intellectual property” that would not have existed without public support (another battle lost, so far, with so-called march-in rights remaining unused for the most part).
And all the while, the transmissible viral infection of “expertitis” among scientists often got in our way:
That ivory tower mentality also fostered a “tendency on the part of scientific experts to talk down to the public,” and exclude them from discussions about funding, said Nicholas Dirks, a historian at Columbia University and president and CEO of the New York Academy of Sciences. “That has really backfired big time,” he added. “We’ve been very blind about how that leaves the public outside of those committee rooms, not knowing how these decisions are made and what science is really doing and whose interests are being served.”
Yes. And during the early days of the current pandemic expertitis was fulminate. But scientists are not always so obtuse. A review panel I served on last April included lay members of the community. They asked some of the best questions during the meeting. And politics has always had an effect on scientific research, as it should outside of the laboratory. Richard Nixon declared War on Cancer in 1971. For a long time, cancer stayed ahead of the scientists, but because of generous support for basic and clinical science, cancer is in retreat in many places. Childhood leukemias are often not the death sentence they were before Sidney Farber and others started the long, incremental project to understand the molecular causes of cancer. Millions, including yours truly have, benefited from this research. AIDS became a manageable chronic infection for the vast majority of those infected with HIV who have access to and can afford treatment (another matter altogether). Vaccines have made the childhood diseases I had (rubella, rubeola, and chickenpox, all before I was seven years old) mostly a thing of the past. Ditto for diphtheria, tetanus, pertussis, hepatitis B (which eventually causes liver disease, including cancer, in virtually everyone infected) and polio. And going forward respiratory syncytial virus and rotavirus will disappear, unless the current Secretary of Health and Human Services succeeds in his time travel mission back to the nineteenth century (when his brain worm would probably have killed him).
The cluelessness of the academic scientific establishment on the matter of “indirect costs” (overhead) has not been healthy. This has led both the people and the politicians to believe that scientists “are getting rich off grant money.” No, actually. While there is the occasional thief or cheat who escapes notice, for a little while, for the other 99% of us every dime of grant money has been audited, usually in real time (with the audit paid for by overhead). When Vannevar Bush eventually came around, the idea of spreading science across the country made perfect sense. Universities, medical schools, and independent research institutes would house the researchers and the funding agencies would contribute to the costs of research for the people doing the research and the materials they needed.
The current overall indirect cost rate for research supported by NIH in medical schools is less than 30%. The rate is generally higher for individual grants to laboratories, which need higher levels of support. But this means that institutions other than NIH are paying about 70% of the total costs. This is a very good deal by any reasonable metric. As an aside, that private funding agencies limit overhead to 15% is irrelevant. Virtually no research laboratory without a research program funded by NIH or NSF or equivalent gets these grants, which are a very thin layer of icing on the top of a rich, thick cake. But they often lead researchers in new directions and are essential for that reason.
Research is expensive, and has gotten only more so during my career, usually for good reasons of safety and good practice, particularly in the care and use of research animals and the safe conduct of research using toxic materials and experimental model organisms that can be pathogenic. Still, there was a consensus building that the current funding mechanism needed modification:
Addressing that weakness means changing the incentive structures that induce academic institutions to become increasingly reliant on federal grants. Under the current system, the federal government reimburses grant recipients’ costs of building and maintaining new laboratories, creating an incentive for expanding their research capacities indefinitely. These “indirect cost” payments also reward universities for not paying the salaries of scientists to staff them, because salaries paid through a grant come with indirect funding that accrue to the university.
“The incentive system is just terrible,” said Bruce Alberts, a professor emeritus at the University of California San Francisco, and former president of the National Academy of Sciences. Known as “soft money,” grant-funded salaries are not guaranteed, keeping scientists in a constant state of needing to find more support, which draws their attention away from dreaming up ambitious new projects and focusing on what’s going to get them their next paycheck.
“Besides all those problems, teaching has been suffering from this, and public service too,” said Alberts. “If the institution’s not paying your salary, you’re not really paying attention to the institution’s needs either.”
Another thing that has been discussed but not acted upon, is capping the percentage of a salary of the Principal Investigator (the scientist who submits the research proposal and is responsible for its completion) covered by grant awards. But this, in contrast with the current administration’s strictly performative alternatives (tariffs, sanctions, bluster) to developing industrial policies that might work, would have to be phased in over a period of 5-10 years. This is doable and would be a rebalancing win-win for the people and the scientists who do the work that leads to scientific advances that benefit all of us.
Still, it is clear that the wholesale attack on the scientific establishment and its scientists, most of whom have viewed their work as a calling or vocation has done great damage:
Arguably the most insidious fallout is that many scientists who work at universities no longer feel they can count on the U.S. government as a reliable partner in the pursuit of research for the public good. “That’s the most devastating part of all this,” one NIH official told STAT. “Why would anyone trust the NIH ever again?”
“That social compact is being systematically undermined at the moment by a group of ideologues whose real target is not science; its real target is what they perceive to be the power and the arrogance of elite institutions, starting with the great research universities of this country,” said Shirley Tilghman, a molecular biologist and former president of Princeton — one of those universities. To onlookers like Tilghman, what has happened since January seems to be a tragedy of unintended consequences. “The intention was to punish elite universities, it was not to destroy the scientific capacity of the United States, but that’s what they’re doing,” she said. “It’s one thing to destroy something. It is quite another to destroy it and have nothing to replace it with. I think that’s the moment we’re in.”
So, where do we go from here? How can any still-working academic scientist advise an aspiring scientist to go for it when the likelihood of success is so small? I have no idea. But whatever the ultimate outcome, the past ten months have been a teachable moment for American science and its scientists. That this has followed the uneven-to-poor performance of the biomedical science establishment during the current pandemic is not an accident.
Sudip Parikh, CEO of the American Association for the Advancement of Science (the C-suite title is not reassuring):
describes the U.S. research ecosystem of today like a densely packed terrarium. Over the decades, new species have been added, intermingling their roots and growing up and on top of and into each other in a way that makes it difficult to trim areas that have become overgrown without sacrificing the whole. “All of this was ripe, maybe overripe in November of 2024,” he said. “This president has come along, and for better or for worse — many times for worse because it’s been done without a plan right away — and he’s taken that terrarium and shaken the hell out of it.”
Yes he did shake the hell out of it, because he could. Not because he has a plan to improve anything. Undoubtedly the president enjoyed the disruption he caused. In any case, Sadip Parikh was on the job early, in August 2024, when he:
starting holding a series of late-night Zoom calls with a self-appointed task force of more than 70 influential leaders across science, academia, industry, policy, and philanthropy, including CRISPR inventor Jennifer Doudna; (Kelvin) Droegemeier, the former OSTP director; and Noubar Afeyan, founder and CEO of Flagship Pioneering. The group came together to form a set of policy recommendations — intended for whomever came to power in the November elections — to push American science and technology to new heights. Bush’s vision, they had decided, was increasingly irrelevant to the pace and scope of 21st century science.
In February, the task force released VAST, a Vision for American Science and Technology. The roadmap makes 14 pro-innovation policy recommendations. Over the next year, this shadow science cabinet intends to pull in mid-career scientists from across a wide range of disciplines to start to identify discipline-specific policy needs.
Whatever they come up with, Parikh doesn’t want to call it a social compact, or a social contract. That framing puts scientists outside of their communities. And they need to be inside, as essential members, every bit as essential as the people who drive buses and fix toilets and bag groceries.
VAST is very polished indeed! And Sudip Parikh does hit on something essential for the science-adjacent Professional Managerial Class (PMC) to take to heart. We are not outside of anything in society. Our expertise is limited but our interests are universal. This, of course, applies to everyone, including the plumbers to grocery baggers Mr. Parikh mentions. However, we, scientists and non-scientists alike, do not need to be led by a blue-ribbon panel of the PMC who made this complete mess because they were not paying attention to anything beyond their next grant application. The vision of Vannevar Bush, without the snobbishness, will do just fine.
Science, in its vastness, is still the endless frontier. But so is every other human endeavor. Instead of a revolution led by a congeries of VAST Contributors, the only thing science writ large needs is to regain its sense of purpose. And that is to improve our understanding of the natural world, from the interior of the atom to the entire ecosphere of planet Earth, and beyond (not including mining the moon or terraforming Mars, however; Star Trek is science fiction, Mr. Musk).
Given what science and its handmaiden technology have “achieved” during the past 250 years, perhaps we should refocus our attention on the ecosphere. Yes, this will be expensive, but that all depends on how costs are accounted and benefits reckoned. Our children and grandchildren might even have reason to thank us, one day.
See you next week in the usual format!
Interesting stuff KLG…
I’ll note that the Dodgers hit a lofty .203 for the World Series and won despite clinging to the Mendoza Line, so it is possible~
Everybody forgets he was an excellent defensive shortstop and had a heck of an arm. You don’t get to play in the big leagues even parts of nine seasons without a little sumthin goin on. Defense can win games, too!
Dare I make a plea to the North American contributors to stop relying upon baseball analogies to make their arguments?
Outside Canada, USA, and Japan, nobody cares and therefore nobody knows about the fine points of that game, its terminology, and its star players — but people from the USA seem blissfully unaware of this state of affairs. This leads to absurdities like the following, in a column giving advice to researchers written by a world-famous computer scientist (Dennis Patterson, in CACM v.68 nr.2):
“As most research does not have a big impact — otherwise it’s not really research — people tend only to remember the rare home runs and not the other at bats. Since there is little reputational downside to missing a home run, if you want impact, why not swing for the fences? It strikes me that hitting out of the park is more likely if you aim to hit home runs than if you always try to bunt for singles. As Helen Keller put it, aim high as ‘the fearful are caught as often as the bold’.”
I have no idea what all this is about, except I gather one should take risks.
We live in a globalized world, in which the only game that is played, followed, and discussed with enthusiasm on every continent is not baseball, not cricket, not rugby, not volley-ball, not polo, not golf, not tennis, not buzkashi, not sumo, not pétanque — but football (soccer for you English-speaking people). If you really want to use sports-related metaphors when addressing an international audience, refer to football/soccer — this is the only game everybody knows something about. Forget about baseball.
Better yet: do not use sports-related metaphors in your writings.
Indeed. Though it is probably a lost battle if you ask me vao. The temptation to use those analogies must be there all the time even if many do not understand them. I also guess that USians will never “feel” football/soccer as it is felt in the rest of the world. Even if the next world championship will be hosted, not just by the US, but by the whole North America. Disgustingly, the FIFA awarded Trump with a “FIFA Peace Prize” for his “tireless efforts” to bring peace. This indicates how the “Western Hemisphere” still dominates politically the football sphere. Russia is banned of course. I believe that a political change in FIFA might be one of the first signals of a truly multipolar world.
FIFA has always been, and remains, deeply corrupt.
Govern your reactions to its behaviour accordingly.
Here’s the deal with soccer in the USA, kids are really into it and then lose interest steadily.
{awaits brickbats from that guy who goes to pro soccer games in the USA}
Good point, and that is some sloppy writing you highlighted, but in this case it’s not really a metaphor and more of a sports-based equivalent – Mendoza line = 20%. And’s it pretty obscure at this point – you have to be a pretty devoted fan of a certain age to get the reference.
But the first and only autograph I ever got as a kid was Mario Mendoza’s, so this particular sportsball reference gets a big thumbs up from me.
Thank you KLG.
For this and all the others I haven’t commented on.
Your fountain of knowledge is worth a long drink, every time.
I curse the Bayh-Dole Act with every breath I take. It is the prime example of the decline of science in the U.S. Since that time, inventions primarily or completely funded by public expenditure have all become proprietary closed systems. “March-in rights” ?– hah, what a fraud.
If ever we were going to see march-in rights used the way they were intended, it would have been with the C19 pandemic. But we clearly see that the Bayh-Dole Act has had its intended consequences, see rules no 1 and 2.
In related anti-science news:
I know I’m treading on the toes of some Ivermectin advocates that comment here, but so be it.
CDC’s new deputy director is vocal critic of vaccines, advocated for ivermectin CIDRAP
Third infant in Kentucky dies of whooping cough as national cases stay high for second year in a row CIDRAP
For discussion of the above and more see TWiV 1274: Clinical update with Dr. Daniel Griffin.
so i suppose a thorough review of Lysenkoism is in order, then?
as a predictive model?
so much WINNING!!,lol.
here, on the farm, i apply the science of others almost every day.
the biggest barrier that has materialised to this has been the swallowing up of all the research journals by that damned swiss company that starts wit an E, and sounds like a fancy plumbing fixture.
as Lambert always insisted..the ability of citizens…and yeoman farmer guys like me…to both apply the research of others…as well as to potentially improve the understanding therein by that application…is how things move forward.
all this wanton destruction and rapine…so sinferuenza!….by stupidly rich morons who never bought groceries…is just a fitting capstone to where we’ve been heading for all my 56 years.
glad i invested what i could in a Library of actual books.
and a bamboo grove for paper for the eventual scriptorium.
sigh.
The researchers I know – few in number – are pretty checked out as far as our larger community goes. Their research seems to take them abroad frequently, which might be a part of it. I don’t know. But they don’t seem to have a good sense of their funding being from the largesse of the larger community. They might be time pressed or some such constraint, I don’t know. But taking their positions for granted, certainly. Or else assuming their place the natural result of their ability and work, rather than from a compact.
I feel the grand edifice of science in the u.s. has lost its sense of calling, much as many of the scientists, and scientist/managers have lost their sense of a calling. When Trump began his second term, what is called science in the u.s. had degenerated into a specialized business enterprise and little more. Trump and his cohorts have done tremendous damage to and destruction of the remains of science. Perhaps such ruin is necessary to clear the ground and make room for a new foundation for promoting Science, or at least something closer to the ideal of Science. Neoliberal Capitalism is poisonous to Science, much as it is to Education, Medicine, Defense … Democracy.
This post mentions many of the milestones toward the destruction of Science: the Bayh-Dole Act of 1980, the setting of scientific priorities and vetting investments in those priorities, politicized science for advancing the President’s policy priorities, or the policy priorities of some other political or more often business interest …
Controlling, science has inspired the Managerial Demiurge to invent ways to steer science, incentives, that also lead to further decay of the money driven University systems in the u.s. Open-ended government incentives for expanding university research capacities lead to the proliferation of research centers peripherally affiliated with the universities. Rewards to universities for not paying the salaries of scientists to staff them, because salaries paid through a grant come with indirect funding that accrues to the university — helped create a the system of post doc and graduate student employment guaranteed to give second thoughts to anyone who felt a calling to Science. “This has led both the people and the politicians to believe that scientists ‘are getting rich off grant money.’” [Are these research grants or research contracts?] I think there is more to this belief. The graduate students and post docs doing the research receive a relatively small cut of the grant money. The rest is eaten in overheads. Worse, the post docs are paid with relatively small, ephemeral “soft money,” grant-funded salaries and must forever worry that their employment might end with the research contract they are working on. Although they are not getting rich off grant money, I believe the university administrators running the research centers, and the Principal Investigators while not getting rich are doing very well. Most graduate students and post docs are compelled to sign away their rights to any discoveries of monetary value but I know of at least a few professors who did become modestly wealthy as a result of discoveries their research teams made.
I took a look at the VAST Report. It reads like a glob of power-point presentations, incorporating all the finest business buzzwords from business mission statements and corporate reports. The VAST Report is full of “competitive advantage”, “innovation”, “national security”, “infrastructure”, “empower”, “adaptable workforce”, “reducing inefficiencies” …. After several passes over the report I was not able to identify the 14 pro-innovation policy recommendations although there were numerous aggregations of bullet points identifying policy. I could not make myself read the VAST Report. It brought back too many bad memories of must attend presentations by the managers in the firm I worked for. Though it does contain some wordy over-dressed common sense recommendations the VAST Report solidified my impression that Science has become the Business Enterprise of science.
Trump may give this enterprise its coup de grace but it had already become an ailing shadow of its former self.
I think we must agree that not all science turns out for the good of humanity, and Vannevar Bush was involved in the early days of the Manhattan Project as an example. The war would have been won without the bomb even if they weren’t sure at the time.
So yes it’s very political and some might justifiably ask if our primitive politics–things seem to be going backwards–can control our attempts to master and not just understand nature. A major topic around here lately is AI–controversial indeed.
Trump and the Kennedy guy won’t be around forever and it’s not just science graduates who are struggling apparently. If we want better things to happen we may need to fix our politics first. The Trumpies are the symptom rather than the disease IMO.
An early observation in regards to this; “… the Trump administration has, with shocking speed, ripped up the longstanding social contract that existed between scientists and the federal government.”
When the same process was begun with the Social Contract between the State and the Citizenry way back in 1980, did the “real” scientists reflect that they would eventually suffer the same fate? Making predictions from data sets is part of their discipline. The facts were all there. Somehow, a sense of entitlement overcame most worries. Now the bill comes due. When you serve an evil master, evil becomes your fate.
It is past time to restore Moral Philosophy to the curriculum.
Stay safe in these perilous times.
Scientific research in the USA may have also been tarnished by the “replication crisis”.
While this may have been associated with the social sciences and not the hard sciences, it may have made many citizens more skeptical of scientific research.
The researchers involved were funded by some entity, perhaps the US government, unless they were working for free.
see https://www.ipr.northwestern.edu/news/2024/an-existential-crisis-for-science.html
“The reason why this issue is so important is trust,” she said. “If we don’t have a scientific literature that we can trust, then it’s not doing its job, it’s not useful, and it might actually be counterproductive and create other problems.”
It may have demonstrated to the public that research academia does not police its own.
And perhaps the hard sciences were tarred as a result.
A part of this post discusses the great expense of doing Science, all the “Research is expensive”…”care and use of research animals and the safe conduct of research using toxic materials and experimental model organisms that can be pathogenic.” This is true but I wonder whether this is the only Science that could or should be funded. For example, what kind of resources did Einstein require at what cost to write his paper on relativity? What kind of resources did Planck require to write his paper solving the problem of black-body radiation? The resources both papers required were time, access to and knowledge of a considerable amount of unexplained experimental results, a deep knowledge and understanding of the science and tools required to find a solution, and immense powers of concentration and imagination. I believe the u.s. had and perhaps still has a number of individuals trained in scientific specializations, skilled at understanding and using the mathematical and conceptual tools of their specialization, possessed of access to and the ability to sift through the enormous amount of research results already published, imagination, intelligence, powers of concentration, and perhaps already puzzled by some problem in their chosen field. One thing they may lack is time and the wherewithal to afford food, shelter, clothing et al. without spending the greater part of their time and energy working at some ‘job’. Would it cost so very much if the government were to employ the many trained intellectuals currently unemployed or underemployed to work on blue-sky research? I believe there are diamonds hidden in the research journals that might be gathered and crafted into jewels through the application of time, energy, and driving curiosity. Perhaps the government stipend employed scientists might dream up ambitious new projects. Would it cost so much to pay calling-driven scientists a living wage they could count on? Cutting out the science managers and project leaders, and all the management controls and steering would save a large part of the cost in employing scientists to do basic research. And would it cost so very much if some of the work required access to use of expensive equipment or supplies?
Keynes reportedly quipped that “The government should pay people to dig holes in the ground and then fill them up.” People would reply. “that’s stupid, why not pay people to build roads and schools” Keynes would respond saying “Fine, pay them to build schools. The point is it doesn’t matter what they do as long as the government is creating jobs”. I believe we live in times when less and less of basic essentials like food, shelter, clothing, energy, et al. will be available. Additionally, monopoly rents and the monopoly constraints on supply further help to drive the inflation of the cost for essentials. A Keynesian approach to growing employment and the economy — like paying scientists a small stipend [perhaps somewhat more than graduate students and post docs are currently paid] to do research would incur some inflation of the costs for food, shelter, clothing, energy, et al. However, some claw-back and redistribution of the monopoly rents and the monopoly constraints on supply that help in driving the inflation of essentials might alleviate a large part of the inflation problems. [And employing scientists should be a small part of the repairs to the damage done by so many decades of Neoliberal capitalism.] Of course the u.s. could never ever do any of these things.
Trump’s preferred approach to the economic problems facing the u.s. seems to be allowing more underemployment, unemployment, and more starving, ragged, and homeless. I am not sure how long Trump’s solution will continue to work.
My sympathies to Senator Harley K. Kilgore who, IMO, after reading this nice KLG piece and some extra context, had an interesting view on how to organise and fund public scientific research. I have always thought that scientific research should not be excessively centralized in scientific hubs but somehow dispersed, not only geographically, but in purely scientific terms, financing scientific fields which do not look fashionable but show merit and ambition in the proposals.
Instead of people like Kilgore (constructive mind) the US senate is now infested with people the kind of Lindsey Graham (destructive minds) fixated with sanctions. The EU Commission is equally infested with sanction people.
This is being a little disingenuous. Yes, academics are not paid a lot. But the academic sector as a whole is now extremely expensive, at all levels. University costs have been outpacing inflation for decades and it was never going to be sustainable. Zooming out past the immediate arguments and policies, one way of looking at the present political battle is as an dramatic, emotive theater surrounding the cold political calculation to downsize the academic sectors costs by fiat.
The hatchet is being taken first to the front-line staff, and not to the bloated administrative layer as it should be. But such is the way of all austerity downsizings. The bureaucracies involved fight to defend their offices first, not their staff or institution. I think the same battle will eventally be fought in the health sector too, among others. The US and the west no longer has the money to afford all this.
Your comment is misdirected. This piece is about the graduate-level operations that fund research, and not overall higher ed costs generally.
We have chronicled over the years how most major universities have become hedge funds with educational institutions attached. The bulking up in staffing and costs is the result of a self-licking ice cream cone, of highly paid staffers raising money, to then “invest” in naming opportunities from buildings to faculty chairs as well as overly plush plant (glamorous gyms!) so as to entice donors, which have virtually NOTHING to do with the educational “product.” particularly of scientific research.
To expand on Yves’s point, a discussion with the CFO of the local Russell Group university (I.e. top ten, research led UK university) disclosed that (mostly foreign) students pursuing taught graduate degrees are insanely profitable, (most domestic) undergraduate students are mildly loss-making and research bleeds cash like a stuck pig.
Obviously, this is the usual problem of attribution of revenue and cost vs value: none of the revenue of undergraduate and graduate students is attributed to research even though it underpins the reputation and the faculty that draws those students in the first place. The same accounting / philosophical games were played by Beeching and Lord King to cut regional feeder services and prioritise mainline services in British Rail and British Airways respectively: allocate the revenue away from one part of the network to another, making one profitable and the other loss-making. You manage what you measure….
It would be interesting to tie research funding to undergraduates and graduates, so that senior academics are rewarded by undergraduates who pass and the graduates who choose to join their groups. The peer review would then be a function of the choices of the next generation, not the previous. No more science advances one funeral at a time.
Instead undergraduates and grad students are there to cross subsidise research and grant funders are there to pay for the university asset base (the overhead) and the PI’s.
My point about overhead is that it is a good deal. Universities and medical schools build out the infrastructure in the expectation that some of the indirect costs of managing research will be paid by the funding agencies, including utilities, animal care facilities, auditors, grant specialists who keep track of rules and regulations and sometimes act as a buffer between scientists and the funding agencies. The entire edifice would collapse without these administrators that some call bloat. And overhead/indirect cost rates have decreased during my career. When I began the big players like Harvard and Stanford had negotiated rates approaching 100% for grants to individual labs (I do not remember all the details because that was beyond my ken at the time). Donald M. Kennedy got into big trouble for misusing overhead when he was Stanford President. Private institutions still get a higher rate but nothing like before. At my state university the rate was 48%; I think it is up to 52-55% now. Even at these rates no institution is getting “rich” from indirect costs. Could this regime change for the better? Of course, but not by fiat on a single day.
The administrative burden in the research part of universities and medical schools is not nearly as extensive as in other parts. But increases in this are also due to the requirements that did not exist when I began, such as animal care and use regulations. We treated our animals humanely but that was not necessarily a requirement. Biosafety Level designations did not even exist when I began because the first gene had not been cloned using a bacterial vector yet. For clinical research the requirements to protect subjects and investigators and the validity of the research are much more stringent than they were before, as they should be.
But back to administrative “bloat” in general. Medical schools must maintain large staffs to keep up with ever changing requirements for accreditation. The process is repeated every 7-8 years. Re-certification takes two 2-3 years of constant work by faculty and staff in addition to the regular staff involved in record keeping. The final report runs into the thousands of pages to address more than ten elements (I try not to remember exactly how many). Is this really necessary, if 95% of your medical students are graduating on time and “matching” into residency programs across the United States that are required for getting a license to practice medicine? These are absolute external checks and balances. And virtually all American medical schools are this successful…
Watching brilliant scientific minds at work has been a privelige. The drive to discover that led to the amino acid sequenator was a lifetime of work. Another relative has more patents at that local airplane maunfacturer than anyone. Academia and the business community has fostered these breakthroughs in spite of all the inertia, the bureaucracy, the jealousy, and general ignorance of politicians. We are now brooching the capacity barrier with machine intelligence making generations of inquiry nearly instantaneous. We are playing with fire, inventing fire, are we at the point that Joseph Tainter observed in “the collapse of Complex Societies” or?
We are so far beyond anything heretofore in terms of complexity, virtually everything we own and use on a daily basis, we don’t understand how it works, or how to fix it if it breaks.
Should the grid go down, one of the simplest things that hasn’t changed at all in the past say 150 years is guns and rifles.
Not one of them is powered by electricity…
And it’s not over yet. Because there is no full-year gov budget, at least some NIH institutes (I don’t know about all) are making significantly fewer awards than normally this funding cycle. Another (this time partial) gov shutdown in January (~30% chance at polymarket) may finish off a lot of research groups. And the NIH budget is in the same bill as the Dept of Education.
The so-called paylines are gone (a payline is the % of the applications ranked by reviewers that will be funded — a 20% payline means that the top 20% of applications will be funded). During the review process, the “bottom” 50% of applications are makred by the reviewers as such and have no chance of being funded. Without the paylines, someone at the NIH institutes can now select any application among the top 50%, which means that the reviewers have much less say in who will be funded.
See Implementing a Unified NIH Funding Strategy to Guide Consistent and Clearer Award Decisions on the paylines.
Is this a bad or a good thing that the paylines are gone and the reviewers have less say? My answer would be let’s wait and see. The previous system was not perfect either. What the reviewers often valued was not the proposed research but the reputation of the research groups and the research institutions. And the reputation came from a large # of papers published in the so-called ‘high impact’ journals (maybe KLG will disagree with me, but imhp, the Editorial Broad (a self-selected group of scientists) decides what will be published in those journals, not the peer revewers). The NIH made major changes earlier this year that were planned under the previous administration, to reduce the role of reputation in the application evaluation. If done correctly, removing the paylines may work in that direction even more. On the other hand, the decision process may become more political.
As to the “replication crisis”, KLG may disagree with me, imho, it stems from i) the widening gap between the resources and expectations — the resources are dwindling, the expectations are the same or higher — the size of the so-called ‘modular’ budget has not increased in decades; ii) worsening quality of the commercial biological materials, reagents and equipment — research depends on commercial sources of reagents and equipment, and crappification takes place everywhere — finding a good antibody is hard, PCR machines do not keep the temperature, etc; iii) worsening quality of support workforce (tech, grad students and postdocs) — this is an issue at ‘smaller’ research institutions, ‘top’ places can recruit from abroad and find better prepared workforce; iv) megalomania on the part of some researchers especially at med schools.
No real argument from me, NoN.
But a large part of the “replication crisis” is simply made up. But it does make the news. The probability that an experiment can be replicated with any precision is inversely related to the complexity of the experimental model. And in biological research, all models are very complex, even if the model is one cell type. The key is to ask questions that are only as general as the model allows. This is how and why the molecular bases of cancer have been characterized so well.
Prosaic example: The first series of experiments I did as a postdoc were repetition of the work of another group on the pH-dependence of the interaction with “my favorite protein” (MFP) with F-actin. Their protein was from one source. Mine from another. Same size. I never could repeat their results. Two years later I discovered the second version of MFP, is the result of a 400-million-year-old gene duplication in fish that actually dates back to our cellular slime mold ancestor. The proteins were different, only about 70% identical in humans, but who knew when animal genome sequences were still in the future? My result hypothesis was proved when MFP-2 was found on chromosome 15 by a cytogeneticist instead of chromosome-9, where we thought the gene was.
The paper that started the “replication crisis” was by JPA Ioannidis in 2005, and covers the misuse of statistics (standard business practice in science and beyond). It was found to have an error in an equation and was corrected in 2022. As far as I can tell, and I did not look very far, the error is inconsequential. However, this would indicate that nobody really read the paper. Ioannidis publishes between 50 and 80 papers a year, and I will leave it at that.
I think it is the so-called ‘post-publication’ review sites that are fueling the “replication crisis”. Websites like Retraction Watch, PubPeer and Forbetterscience (and the short-lived Paul Brookes’ site should be mentioned; there are others but I am less familiar with those). All of these sites began in the 2010s (I think). To what extent their founders were inspired by the JPA Ioannidis paper I don’t know. The founders of these sites are all trained as scientists and/or physicians. The founders of Retraction Watch are MDs, Bkookes is a professor, Forbetterscience is a trained scientist who is no longer in science, PubPeer is anonymous. From this, a whole new branch of — I guess we can call it — science sprang out — there are now fellowships, integrity consultants (Elisabeth Bik), Sleuth in Residence Programs (Retraction Watch), image analysis software (ImageTwin is a highway robbery in terms of cost), etc. The journals now require submission of the original images and data spreadsheets. There is money in this to be made — and there must a crisis to justify it. But this came out from within science itself and could be considered a self-correction mechanism.
The other source of this perception of a crisis seems to come from the industry. Because the biotech ultimately relies on the basic research findings, they need to know that the published work is valid. And they claimed that they could reproduce very few findings, at least in cancer (Begley CG, Ellis LM (March 2012). “Drug development: Raise standards for preclinical cancer research”). The Reproducibility Project: Cancer Biology comes from that. And of course, as you point out, the study systems and experiments can be very complex and there are always unknown unknowns. In the way, maybe Bayh-Dole is not such a bad thing, because a scientist that made a discovery can then launch a start-up that develops an application of this discovery — who could know better the system and technique than the discoverer?
A biotech I collaborated with once spent ~$16K and bought up every commercial antibody on the market for MFP, and one of many worked very well — a monoclonal. We managed to do some good work with it at first. But the second batch was not as good but still usable. The third batch was useless. I called the vendor and told them the antibody has no specificity any more and they need to remake it. And they tell me they will sell the rest of the current batch (the one that is useless) and then will thaw the hybridoma line and make more. I suspect they mismanaged the hybridoma line too (the antibody gene gets mutated with passage, because cancer cells have unstable genome) and need to start from the very beginning. Then, a grad student from a different institutions e-mails me that he can’t detect the protein with the antibody that we reported — I told him I can’t either anymore — the antibody is dead…
As one who became a scientific worker and then a scientist when virtually everything was analog, I remember well the advent of digital imaging processing. My first thought was, “Oh, hell no! This will not end well.” What it did was make data manipulation in molecular cell biology too easy for those so inclined.
When I had graduate students and technicians in the lab, they were required to defend their primary data to the entire group and then to me one-on-one. We were not perfect by any means, but I will stand by any data our group published. PubPeer has finally squashed the “make shit up while looking legitimate pathway” but at a very high cost to science. What the PubPeer sleuths find is not a replication crisis. It is a cheating crisis. Sylvain Lesné, finally learned the lesson at the cost of his career, along with many others. Then the “pay to publish anything” online open-access “publishers” came along to further dilute the literature with crap.
I am aware of the cancer research data, but it has always been a long way from cellular data to chemotherapy that works. The cancer research community forgot this and the Bayh-Dole hype machine covered it up as scientists forgot their calling and only wanted to “cash out quickly with a unicorn” as PlutoniumKun put it so well more than three years ago.
And commercial antibodies often do not work! Which is why unless those we needed clearly worked well, we made our own. At significant cost.
I enjoyed your article, KLG. One very minor proof reading point, para 19 “Yes. And during the early days of the current pandemic expertitis was fulminate” should read fulminant or fulminating. Fulminates are shock-sensitive metal compounds prepared from fulminic acid and used historically as detonators.
Thank you! Indeed. Spellcheck does not check stupid.