US Successfully Planned for the ‘Endless Frontier’ of Science Research in 1945 – Now It’s Time to Plan the Next 75 Years

Jerri-Lynn here. A timely reminder of how the US did science planning in the days before neoliberalism and its market ideology became dominant.

By Marc Zimmer, Professor of Chemistry, Connecticut College,  Originally published at The Conversation

The U.S. has been the most productive country for science and technology for decades. Many of the basic policy tenets that supported American prowess date back 75 years, to a document called “Science: The Endless Frontier.” Released by the first U.S. presidential science adviser, engineer Vannevar Bush, just two weeks before the Hiroshima bombing in 1945, it would become the blueprint for postwar science.

“The Endless Frontier” led scientists to successfully advocate for federal scientific funding and a separation between science policy and politics. They argued that if science could win wars, it could also help maintain peace.

The report advocated that governmental, industrial and academic research can accomplish far more in partnership than in isolation. It led to the development of the modern American research university, the National Science Foundation and increased government funding for science research, which rose by more than a factor of 10 from the 1940s to 1960s.

The science landscape has changed radically since Vannevar Bush was in the lab in the mid-20th century. AP Photo

But many facets of the plan aren’t working anymore, and the structural framework laid out in “The Endless Frontier” needs refreshing for 2020. Research funding, especially the government share for basic research, is being reduced, there is a narrower focus on short-term outcomes and U.S. federal agencies are cutting scientific advisory panels. How the country, the research community and the public respond to these changes will determine the United States’ geopolitical standing.

On the occasion of the 75th anniversary of the report, the National Academy of Sciences, The Kavli Foundation and the Alfred P. Sloan Foundation hosted a symposium to reflect upon the past, present and future of the United States’ scientific research enterprise. It brought together leaders from science, government, academia, business and philanthropy.

As a chemist in academia who’s just written a book called “The State of Science – What the Future Holds and the Scientists Making It Happen,” I was eager to see what symposium attendees had to say. The presentations and panels covered a range of topics, and four major themes – though not solutions – emerged.

Planning Further Ahead

Symposium attendees agreed: The U.S. needs a long-term federal science plan, one that spans many future administrations – a road map for science that is both protective and aspirational.

The consensus was that today’s level of federal funding and emphasis on market-driven research are imperfect. The path from lab bench to application has a significant gap – it takes time to convert research findings into a company, and venture capitalists are impatient. Many good ideas, such as large-scale affordable desalination, are dropped because the technology is complex and will take many years to commerialize.

Additionally, democracy is messy and slow. To stay competitive, it needs a long-term plan as was laid out back in 1945. As Rafael Reif, president of MIT, said, “Vannevar Bush envisioned a wild garden of scientific possibility…. At the current moment, [the scientific enterprise]… requires deliberate concentrated action to harvest the results. In effect, we need to farm for innovation.”

hina has grown to rival the United States in scientific strength. Mick Ryan/Cultura via Getty Images

Stronger Global Competitors

In large part thanks to strategies laid out in the 1945 Bush report, the U.S. has led the world in scientific innovation and research for over 70 years. But the ecosystem of the research world is changing.

According to Sen. Chris Van Hollen, D-Md., “the United States’ share of global R&D spending fell from almost 40% in the year 2000 to 20% in 2017. During that same period of time, China’s share rose from less than 5% to over 25%.”

China is a formidable scientific and technological competitor, particularly in its capacity to use science effectively and build partnerships with industry. Numerous speakers suggested that a lack of an American long-range coordinated plan that links academia, industry and federal research is letting the Chinese ascend.

Sen. Lamar Alexander, R-Tenn., told the audience, “The situation we are in is very much like being a very good football team, playing in a league that suddenly has gotten a lot better. There are a lot better teams in our league for the next 75 years than there have been in the last 75 years when it comes to science, technology and research.”

Presenters were torn about limiting American collaborations with China. Should the U.S. limit knowledge transfer or continue collaborations – the most effective way to advance scientific knowledge? Not surprisingly, most academics present favored more collaboration with researchers from China and other countries.

“Are we going to have the courage of our convictions to fund the science we need, at the level we need, to ensure our competitiveness? This is more important than knowledge seepage,” observed Ronald Daniels, president of Johns Hopkins University.

Gabriela González, professor of physics and astronomy at Louisiana State University, suggested that multinational partnerships are so important to modern science that graduate schools should teach students how to conduct collaborative science.

Outreach and education are working to make sure tomorrow’s scientists and engineers reflect the whole U.S. population. Ariel Skelley/DigitalVision via Getty Images

Expanding the Tent

The United States and the world as a whole have not been doing a good job at taking full advantage of the diverse pool of potential scientists. The STEM fields may have found and nurtured many of the future Einsteins, but we have fallen behind in cultivating new Marie Curies and George Washington Carvers – let alone the C.V. Ramans.

A 2017 National Center for Science and Engineering Statistics report shows that although white men make up only one-third of the U.S. population, they constitute at least half of the country’s scientists. Countries, companies and academic institutions are handicapping themselves if they don’t make use of all scientific talents available across a vast array of gender identities, races and ethnicities.

Thirty-seven of the 89 U.S. citizens awarded a Nobel Prize since 2000 were foreign-born. Most notably, all six American winners of the 2016 Nobel Prize in economics and STEM fields were immigrants to the United States. As Rafael Reif, himself an immigrant from Venezuela, put it, “immigrants are the oxygen” that let research survive and thrive.

Disturbingly, the National Science Foundation reports that the number of international graduate students coming to the U.S. dropped by 22,000 (5.5%) in 2017, though the decline hasn’t been as steep in the following years.

Communicating Science, Outside the Lab

Truth and facts are central to the workings of science. However, the increasing speed of scientific breakthroughs makes it harder to understand and communicate science and its complexities.

Actor Alan Alda, who founded the Alan Alda Center for Communicating Science at Stony Brook University, connected public misconceptions about scientific topics with the lack of good science communication. The public has to trust the research enterprise in order to support and learn from it. He said, “We don’t always have to agree, but we have to learn to hear each other,” and that requires good communication.

Maintaining and bolstering trust in science has never been more critical, as science has an ever greater potential to change or lives, for better or worse.

Print Friendly, PDF & Email


  1. GK

    Thanks for this post. It’s remarkable to consider how forward-thinking Vannevar Bush’s report was, written as it was in the midst of WW2. I also was struck by the following quote from the report’s recommendations for science education. That was 1944; today, we have wealthy parents shamelessly buying admission to college for their children. Where did we go wrong?

    We think we probably would not, even if we were all-wise and all-knowing, write you a plan whereby you would be assured of scientific leadership at one stroke. We think as we think because we are not interested in setting up an elect. We think it much the best plan, in this constitutional Republic, that opportunity be held out to all kinds and conditions of men whereby they can better themselves. This is the American way; this is the way the United States has become what it is. We think it very important that circumstances be such that there be no ceilings, other than ability itself, to intellectual ambition. We think it very important that every boy and girl shall know that, if he shows that he has what it takes, the sky is the limit.

  2. paros

    I think the path Vannevar Bush set out was detrminental and has directly led us to where we are with the current myriad of issues with scientific funding. The way he advocated for federal scientific funding programs to be structured led to increased administrative bloat and to increased resource capture by well-connected scientists rather than funding of the best scientific ideas.

    1. Craig H.

      The guys controlling the funding think it’s working great and DARPA is 20 years ahead. The margin of this text box is too small to contain the proof. Also it’s above top secret and if they told you they would have to kill you.

    2. Susan the other

      Let make the military into a new campus for more humanitarian science. Not just to promote the humanitarian, environmental side of science (we do that in our universities quite well), but to go to the heart of destructive science and change it.

  3. Jeremy Grimm

    I am not sure what message this post intends to convey. The Federal government is spending less on funding basic research, cutting advisory panels, and focused on short-term outcomes. The “level of federal funding and emphasis on market-driven research are imperfect” — imperfect? Then this is undercut by discussion of a world or stronger global competitors. Fostering diversity in Science is lauded — to garner political correctness points. The number of foreign graduate students has dropped. Finally Science poorly communicates with the public. No specific reforms are mentioned — I guess this symposium is about “vision”.

    On its conclusion I expect the symposium will recommend more funding for Science, especially basic research, more funding and support for minority and female scientists, and funding for an outreach program to encourage more foreign graduates to come to the U.S. for their schooling — maybe some proposals for a further sweetening of the H-1B laws.

    I think it will take a lot more than a “vision” to repair the U.S. systems of higher education, and the enthrallment of Science to the Neoliberal Market. The Vannevar Bush blueprint extended the linkages between Science and DoD as the U.S. Government promoted the Cold War. Although that coupling brought us many poison fruits, it also funded many true advances. But as Neoliberalism infested our Polity — Education and Science were reduced to handmaidens serving the Market. I doubt that will change as a consequence of this symposium.

    But Science already had other problems. As an undergraduate I was deeply impressed by Watson’s account of how the structure of DNA was discovered. The petty jealousies and back-biting revealed in that account which I attributed to concerns for who should get credit or the prizes for a discovery suggest another problem for the advancement of Science. Too much concern for personal advancement and prestige are built into the Scientific hierarchy.

  4. Rolf

    I am a veteran of academic research (earth science and chemistry), having spent the past 25 years post-PhD supported largely (although in no way lucratively) by research grants (soft money). Simply put, the state of funding of US science is a mess, and I am not even sure if it can be fixed. Most current problems in my field are enormously complex — all the easy stuff has already been done. What remains is hard and will require long term commitment. This should come of no surprise to anyone.

    The good news? In my area of interest (physics and chemistry of crystal growth and dissolution), the steady increase in low cost computing power, together with the proliferation of excellent software (often available at little or no cost) has meant that many previously intractable problems can be investigated via modeling and simulation. Simulation results can then be compared with “ground truth”, using selected (and expensive, difficult to fund) experimental observations made in the laboratory or the field. This cycle of collection of real experimental results and comparison with their virtual counterparts creates a virtuous feedback that enhances overall understanding, ultimately leading to improvement in theoretical understanding.

    The bad news comes in several pieces.

    First, the above activity is still hugely expensive in terms of time and equipment. Laboratory instrumentation and hardware continues to increase in sensitivity and capability, but these obviously greatly increase the cost of data acquisition (these capital costs are typically borne by the university of research institution). Instruments also do not last forever, are expensive to maintain, and their research utility falls once their capabilities are eclipsed by the latest revision.

    Second, I am old enough to have witnessed the slow embrace of a mindset that any funded enterprise in science must have a demonstration of results that are provably useful, and not just to scientists. At one point I recall a warning that NSF proposals for basic research must separate mere “curiosity-driven” science versus that which yields a useful (presumably easily monetized) result. But this distinction is not one that scientists typically make. Not to put too fine a point on it, but many key discoveries in science are accidental, serendipitous, and most scientists are indeed best driven by a basic curiosity of how nature is put together. That is not to say that science doesn’t have its method: careful hypothesis testing and comparison of actual versus predicted results is of course the way research is conducted. But most scientists are aware of the severe limits of their knowledge, and are willing to abandon or at least modify a treasured theory if it fails to explain results, even to the point of starting from scratch. My point is that the oft-voiced notion that we should constrain research to that which is most likely to yield useful results presupposes that we already understand the relationship between known and unknown, i.e., that we know the answer before we start. Obviously for key problems this is not the case.

    I believe this climate is much different from that prevailing post-WW2 up to the end of the Cold War, when much work was funded not so much because of recognition of strategic importance, but simply because even if we didn’t know the importance, we at least knew we wanted to gain that understanding before the Soviets did.

    Lastly, as any one leaving graduate school within the last 30 years knows, permanent positions in academic science are few and very far between, although a professorship is the position most scientists entering graduate school have in mind as a career goal, despite the fact they few will achieve it. (Research advisors are typically not anxious to disabuse students of these ambitions, either.) Europe is in worse shape that the US: faculty have to retire or die for a “new” position to open. In the US, most science PhDs can find a postdoctoral appointment that will support them at poverty wages for a year or two. But most schools forbid postdoctoral appointments to exceed a fixed time limit, and thus they must either be offered a permanent position or jettisoned — and it is usually the latter, in order to make room for new postdocs. After a few of these cycles, scientists find themselves with good papers but no job offers, large outstanding debt balances, no savings, while their friends have long since bought homes, started families, begun a real life. These frustrated, highly trained individuals are thus at risk of passing directly from the training segment to the retirement segment without actually entering the career phase of their professional lives. If we value science and scientists, how do we create an environment in which science is a realistic career path? This a very real problem, and one exacerbated with emphasis on STEM careers and anxiety over the US “failling behind”, but this really a demand problem, not a supply problem. Michael Teitelbaum (Harvard Law, ) has written extensively on this.

    I do not know the answers to the above — there may be none. But I think we need to simply come to terms with the fact that, as in art, music, literature, and other human endeavors whose products we value, we should appreciate the fact that investment in science and science education is money well spent, but it not an investment whose time-to-maturity we can estimate with any precision.

    1. Ian Ollmann

      Agree, the usefulness criterion is misplaced. A problem is that most academic scientists are a bit out of touch with industry despite much trying and wouldn’t know usefulness if it bit them. Since they are largely in control of the granting process too, it’s a bit of blind leading the blind. Industry can and does hire scientists to solve problems important to industry and industry can, should and does pay for that. If the government is paying, it should be blue sky research for the benefit of everyone, and in this area it is really not clear what is going to turn out to be valuable and what not. There is enough of a valuableness incentive through patent law and the potential to start a small company based on a discovery. I don’t think the public research funding needs to push in this direction too. Curiosity driven research is fine with me. Industry will need the results eventually. It just might be 40 years before they do, and you can never be sure which industry it will be.

    2. Jeremy Grimm

      You have written a moving statement about the condition of Science clearly describing the perils of managed Science. I wanted to serve Science but saw the perils you described and opted for an easier path in Engineering — only to find Engineering degraded by the same forces degrading Science.

      Science should not be enthralled to the Market. The only utility that science should recognize is the utility of new ideas for understanding old ideas and developing the furtherance of Knowledge and the creation of more ideas. We stand on the edge of many breakthroughs in Science that dwarf concerns over profitability or patents or other Market triviality. The only path forward is to begin building ideas from the many bricks of knowledge collected and lying about in the thousands of journals. The building of ideas is a slow process fraught with dead-ends and error — and if utility remains a criterion — promising rewards far beyond any “investment” in managed Science. And Knowledge, like Culture, has value in-itself.

      1. Rolf

        Jeremy, I agree, particularly about your point “building of ideas” and the heuristics involved. I think one thing that might help is to increase public awareness of the economics of scientific publishing.

        As many here may be aware, access to journals is controlled by several multi-billion dollar behemoths: Elsevier (aka the Dutch bandits), Springer-Nature, Wiley Interscience, and of course AAAS’ flagship, Science Magazine. Despite the fact that much of the original research was funded from a public purse, the published results are guarded behind private paywalls. This is true even for seminal papers published 50-100 years ago.

        How much money are we talking about here? The subscription cost schedules for university libraries are a closely guarded secret, but probably in the range of a few thousand to tens of thousands per year. Elsevier, for example, follows the “bundling” practice of cable companies, grouping key, top tier journals together with others of far lesser quality. This practice inflates the total subscription cost for a university library into millions per year. Cash-strapped universities are finally, at least in a few US and European cases, saying “Enough!”, and jettisoning the entire package. However this leaves faculty and students in the lurch, and has led effectively to a grey market for science literature (Sci-Hub, LibGen, Researchgate, and so on).

        Students, postdocs, and faculty are absolutely bound by the requirement of publishing their work in top journals, a necessity for job offers, tenure, advancement, grant money, etc. Grant income is key, but not for faculty salaries per se (at best faculty can ask for summer salary): post docs on soft money, graduate student stipends and tuition remission, and the overhead share retained by the university administration itself — typically 50% of the direct costs of a grant. Publications in top tier journals are thus obviously the coin of the realm in science, and without them (both in production and consumption), one’s career is over before it starts.

        What is more, these publishing houses have their product delivered by the authors for free. Offhand I know of no other industry in which this is true. The cost of peer review — a critical bulwark against fraud in science, is also borne entirely by faculty, researchers, graduate students, who pass judgement on and improve the quality of the final published paper. This is time-consuming, tedious work, but absolutely essential to maintain a standard of excellence. But these costs (mostly time) are never reimbursed by the publisher, although they could easily afford to do so.

        So, one way of reducing the inflationary costs of science — at least in terms of publications, is to put control of this vital link between scientists and the public (who paid for the research to begin with), back in public hands. Papers can be embargoed for a year or so, giving the commercial publisher time to get their (minimal) costs back with profit. But after that, these papers should become public goods, free to all. Elsevier et al. of course will continually try to create more journals (“we have to, scientists want to publish more papers!”), and pump up their journal stable. But their control of literature by copyright — and all authors must sign over copyright to the publisher as a requirement of publication, is far, far too long. As mentioned above, papers published in the 19th century are still locked away. For Gods sake, why??

        Again, the public probably has little awareness of any of this. Many people are completely confused as to how academic scientists make a living, and think the universities “pay for research”. No. The public pays for research, and universities are tax exempt. A lawyer once asked me how much of my income was from “publication royalties”; I was fairly stunned that anyone thinks scientists derive any direct income from journal articles. I also recently read that the Justice Department is pursuing a case against Alexandra Elbakyan​ (Sci-Hub), who they claim is working with Russian intelligence. This I highly doubt — this is scientific literature, not Facebook propaganda, and if anyone has a claim on this “intellectual property” it is the public. And if a lone computer scientist in Kazakhstan can figure out how to host terabytes of scientific literature on her own server for free, then surely the US Government can figure out a way to do this for nominal cost.

  5. rtah100

    This article neglects two of the starting conditions for that US pre-eminence:
    – European Jewish scientists emigrated to the US en masse
    – Nazi/German and Japanese scientists deliberately recruited en masse (Werner von Braun, I’m looking at you!)

    It is hard to see these two being recreated so the ambition of the 75 years will have to be different and less hegemonic.


Leave a Reply

Your email address will not be published. Required fields are marked *