Wednesday, May 23, 2012

Putting our money in seed corn -- literally

Noah Smith has an excellent post arguing that spending our limited research dollars on bigger and bigger particle accelerators is, at this point in history, a bad idea.
Weinberg is a particle physicist, one of the heroes who developed the Standard Model. Thus it is not surprising that most of his article concentrates on particle physics experiments. Unfortunately, I think that appeals for governments to pour more money into particle accelerators are A) doomed to fall on deaf ears, and B) not really very convincing in the first place. Let me explain why.
First of all, the Standard Model of particle physics is good. Really good. In fact, we've never conducted an experiment where it makes an incorrect prediction at any level of precision!! In that sense, it is one of the most successful theories ever. Now, the Model may or may not fail at ultra-high energies (such as those that could be produced inside a black hole or a multibillion-dollar particle accelerator), or at galactic distances. But these are not environments that will ever matter for human beings on Earth. 
As Weinberg points out, the Standard Model is incomplete. It doesn't include gravity. But we have another theory, general relativity, whose track record is just as good, to describe gravity. Unifying these theories would increase our understanding of the nature of the Universe, but it's not clear whether it would improve our ability to predict our immediate surroundings. 
In other words, new particle accelerators may be able to answer interesting questions, but they are unlikely to produce much of technological value.
In fact, this has proven true for the last several generations of particle accelerators. We've discovered a zoo of new particles, and these discoveries have improved our theories greatly. But none of these new particles has been something we can exploit for technological applications. In the early 20th century, new fundamental physics led rapidly to applications like nuclear bombs, semiconductors, lasers, and GPS. But to my knowledge, nobody is even trying to make a device that exploits the properties of B-mesons or neutrino mass. 
To this, add another problem, which Weinberg discusses: We actually have no idea if the "next generation" of particle accelerators would find anything useful. In the past, we always had new theories that predicted stuff we should expect to see if bigger accelerators were built (for example, the Large Hadron Collider was built to search for the predicted Higgs Boson). As of now, new physics theories have made no new concrete predictions about what should come out of bigger and more expensive accelerators. If we build those accelerators, it will purely for speculative, exploratory purposes - to see what might be out there. 
Smith's piece was still fresh in my mind when I read this previously cited piece in the Washington Post:
On Wednesday afternoon, [Rep. Jim] Cooper rose to the defense of taxpayer-funded research into dog urine, guinea pig eardrums and, yes, the reproductive habits of the parasitic flies known as screwworms--all federally supported studies that have inspired major scientific breakthroughs. Together with two House Republicans and a coalition of major science associations, Cooper has created the first annual Golden Goose Awards to honor federally funded research “whose work may once have been viewed as unusual, odd, or obscure, but has produced important discoveries benefiting society in significant ways.” Federally-funded research of dog urine ultimately gave scientists and understanding of the effect of hormones on the human kidney, which in turn has been helpful for diabetes patients. A study called “Acoustic Trauma in the Guinea Pig” resulted in treatment of early hearing loss in infants. And that randy screwworm study? It helped researchers control the population of a deadly parasite that targets cattle--costing the government $250,000 but ultimately saving the cattle industry more than $20 billion, according to Cooper’s office.
My natural bias is pro-research and if the amount of money we spent on particle accelerators was unrelated to the amount of money spent on other research I'd probably say go for it, but in an era of tight money (or the perception of tight money), there are other areas that score higher on almost every non-ddulite criteria.

Smith suggests a number of projects related to replacing fossil fuels. Joseph would probably have a number of suggestions involving health and medicine. My first thoughts are agricultural. For starters I'd like to see something like the Human Genome Project for species that can have a potential impact (positive or negative) on our food supply.

It's easy to argue the economic benefits for this kind of research (discoveries like this can be worth $100 million a year which means, after a decade or two, you're talking real money). It's not so easy, however, to get journalists and politicians to give these fields the respect and support they deserve. Part of this comes from a combination of ddulite tendencies and scientific illiteracy -- the reporters love the high-tech stuff but really don't understand it -- but another (albeit related) problem is the tendency to approach the debate as a conflict between practical and pure science. The distinction is artificial and not particularly productive and it's less than clear which side the evidence comes down on.

I'd argue that, given the current state of the disciplines, you're more likely to make a major, change-the-way-we-see-the-world scientific advance digging for parasites in pig manure than analyzing results from the Large Hadron Collider. That's not to say that these results won't prove important, just that we're more likely to see bigger advances in the life sciences in the next few years and agricultural research is one of the best ways to pursue those advances (if not the best).

No comments:

Post a Comment