Further reflections on funding criteria for science.

Indications are that some in Congress are rethinking funding criteria for science… though upon inspection those new criteria might not look all that different from the criteria in place now: requiring that (1) funded research contribute to the national welfare in any of several ways; (2) it be ground-breaking; and (3) it not duplicate existing work. What seems to be at issue is more a question whether an additional high-level certification attest that these criteria have been met.

As long as the topic has been raised, here’s some further perspective. It’s prompted in part by the administration’s major new research thrust to map the human brain and stirrings that portend a possible human voyage to Mars sometime in the early 2030’s (when Earth and Mars will be relatively close).

Both programs would appear to satisfy the Congressional criteria summarized above. But they’re large… they’re big science… and that raises the question: should such grand initiatives meet additional criteria, and if so, what are they?

A preliminary: what makes some science big? Ask people, and you’ll get a variety of answers, ranging from: a given dollar threshold. Multi-investigator. Multi-institutional. Multi-disciplinary. Multi-year. Multi-national. Some or all of these attributes do indeed characterize big science, but they might not be definitive.

Consider two additional, slightly different traits:

(1) so big that the host organism (that is, society…humankind…a nation…you pick your social unit) notices.

For developed nations, R&D expenditures generally amount to a few percent of GDP; in the U.S., this rate is about 2.8%. That rate is not as high as the rate for some other countries, but U.S. GDP is so large that this amounts to something over 30% of the world’s total. The dollar figure is in the neighborhood of $40B/year. That’s for everything: the physical, chemical and biological sciences; the Earth sciences; mathematics and computational science, and so on. Most of the projects imbedded in the $40B are small, the order of say $50-100K; that means something like 500,000 to one million projects. Take away any handful of these… or add a few, even multi-institutional, multi-year projects… and there’s hardly any discernible impact at the national or international level.

But go to Mars, or map the human genome, or the brain, or build a high-energy particle accelerator to look for the Higgs boson, and you’re quickly spending billions of dollars over a span of years. It’s impossible to tackle projects of this scope without having an impact on the rest of science. Money for these efforts typically comes at the expense of other work. Mapping the human genome provided an example. At the time, some biological scientists were suggesting that mounting a serious effort to understand the nature of human thought would be a worthy alternative. And it’s not just the money. For example, such big projects also make significant demands on the technically-trained workforce, who are then unable to work toward other ends.

That brings us to the second characteristic of big science:

(2) so big that order matters.

Back in the 1970’s as lunar exploration drew to a close, NASA and its stakeholders began contemplating manned missions to Mars (and beyond). But given the limited scientific and technological capabilities available at the time, the cost of such an effort would have been staggering relative to GDP. At the time, the US faced the need for expensive military outlays to conclude the Cold War and maintain thereafter a dominant global security position. Going to Mars was a bridge too far. Efforts to build the Super-Conducting Super Collider also fell victim to this calculus. Instead, the United States concentrated its research funds on smaller projects.

Literally.

Nanotechnology received emphasis as did high-performance computing. Biotechnology experienced rapid development. The results of these and other investments have been not just the creation of knowledge and understanding but the generation of tremendous wealth and economic growth… not just for the United States alone, but for and expanding set of trading partners worldwide. As a result, a richer world… a world also more advanced scientifically and technologically… can, three to four decades later, expand its horizons on the types and variety of explorations it might contemplate. Not just governments but also private-sector firms are getting into the act. By tackling these opportunities first, the United States and the rest of the world have actually hastened the day that human beings might set foot on Mars.

All this suggests that when it comes to the big projects, special attention needs to be paid to the question of whether they expand options for society (the way IT and biotechnology have) or whether they constrain or choke off future possibilities. That’s inherent in the NSF criteria, both current and contemplated, but perhaps not as explicit as it might be. When a single science project becomes so big that the opportunity cost is significant, then this needs to be taken in account in a considered way.

Want an example of a bad technological decision… one that foreclosed options for society? Egyptian obsession with bigger and better pyramids for the pharaohs comes to mind.

This entry was posted in Uncategorized. Bookmark the permalink.

4 Responses to Further reflections on funding criteria for science.

  1. A more contemporary example of a “bad technological decision… one that foreclosed options for society” might be the amount we’ve spent / are spending / will continue to spend of fusion energy. To me a general rule of thumb for big projects (need a definition of “big”) might be that if we can’t reasonably expect a payback in 10-20 years, we shouldn’t make the investment. The logic is simple: if an investment hasn’t paid off within 20 years, then most likely the money invested could have been put to better use. Fusion energy, with its many decades of funding, has certainly not justified any past investment.

    This gives rise to a second rule of thumb: the level of justification and scrutiny should get more intense the larger the investment already made. Too many government programs live on and on and on and are excused by “We’ve spent so much on them already, we can’t afford to quit now.” Any good investment counselor – or experienced gambler – knows you don’t throw good money after bad.

    If you think in these terms, I’d come to the following conclusions about current “Big Science” programs.

    • The National Ignition Facility at LLNL would be cut immediately. Hasn’t paid off, hasn’t worked, and no real “victory” in sight.
    • Climate Change research would be looked at much harder. Anything that purports to save a hypothetical world 50 to 100 years from now likely is a poor investment.
    • Weather research. Pour the $ in! Tremendous payoffs so far, and more in sight.
    • Solar power. Kill government funding for it. As a country we’ve poured in $billions since the 1970′s, with little real payback. The problems aren’t scientific but regulatory and economic. Developing higher efficiency solar cells will pay back very little. Ditto wind. Both of these already have captured most of the niche markets they can. Investment in them will not allow them to overcome the much bigger external problems they face.
    • Energy storage. The most important technical and economic problem with most renewables is their intermittency. Fund battery and other energy storage work to finally recoup something from the huge investments in renewables AND to increase system efficiency for all kinds of energy usage (e.g., cars).

    Obviously these reflect my “investor’s” eye; you might choose something else. The key, as with any investment, is to be as objective – and hard-nosed – as possible.

  2. James Correia Jr says:

    I interpret this to be all about a focused effort to innovate. One that is thought out in the aggregate over the long term. If we could focus on our efforts on something BIG (high risk high reward) what are the things we need to innovate to get there? in other words you plot a course knowing the payoff is far off, but that enough little things will emerge that make the investment worth it, even in the face of potential failure of the grand vision. This entails taking big risks. Time or dollar amount are relatively irrelevant. The exercise is one of pushing the boundaries, making bold and even riskier decisions, to provide a path forward and a singular focus for the project at hand.

    Going to the moon was a grand challenge. Not just technologically with the invention and focus of, essentially, the fly-by-wire computer but with the investment in people in all kinds of arenas. The benefits of that mission are still paying off. And no one realized just how much of that technology would be used in our daily lives 50 years in the future.

    WiFi came from studying black holes with interferometry in 1977 and yet a majority of the population in developed countries has it in their pocket now.

    What we lack is courage is to take big risks and see them through. Money is an excuse because our time horizons have shrunk to relatively immediate pay offs. The space program adopted farsighted vision. They wanted to hurry and beat the Russians so they adopted the fail early and fail often mantra. Now we want to build really big organizations that last forever, because it is comfortably stable. This is a symptom of shortsighted thinking and risk averse behavior. It has a role in business, maybe. But science isnt an organization. It is an integration between people and technology, knowledge and application, desire and need. And every once in a while it needs to be spurred to achieve things that are impossible. From the impossible comes the possible. Recent example of big risk: Sky Crane. Untested. Fly-By-Wire. Now we have an awesome robotic geologist on Mars.

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>