When fiscal hawk Sen. Tom Coburn set his sights on waste (funding humanities nonsense) and duplication at the National Science Foundation, there was outrage that a politician might actually look out for how taxpayer money was used.

Well, scientists were right to object even if they were wrong. Smart people in science know we wasted $72 billion in subsidies for energy companies with the right political connections in the last four years, I would rather have had that money used for basic research, and a Navy destroyer that costs $6 billion each is a head-scratcher, but that doesn't mean we should be funding people to play Everquest and call that science - nor should we have dozens of programs spending billions of dollars convincing people inclined to be doctors that they should instead be scientists instead when the whole thing has begun to look more and more like a pyramid scheme exploiting post-docs.

Waste or value has some intellectual wiggle room but it is hard to defend the ethics of anyone who got duplicate money for the same project and kept the cash or, worse, set out to get funded for the same project twice. 

The Virginia Bioinformatics Institute at Virginia Tech has found that funding agencies may have awarded millions of dollars to scientists who submitted the same grant request numerous times — and that scientists accepted the duplicate funding. The analysis led by Dr. Harold R. "Skip" Garner of the Medical Informatics Research Group, found that $70 million in funding this past decade - and maybe billions overall - looks inappropriate. 



If you submit a new proposal right after you just received funding from another agency, one to research ethanol-resistance genes and one to research ethanol-hypersensitivity genes in C. elegans, was that illegal? Well, no, but getting double funding from the army and from the NIH at the same time for a project that seems a lot alike is certainly suspect. Plus, budgets are finite, someone else got denied a million dollars in research funds because another researcher better knew how to exploit the government-controlled science system.

Garner, Lauren McIver and Michael Waitzkin systematically compared 858,717 funded projects using using text mining software and then they did a manual review of matches. 1,300 applications looked like overlap and their manual analysis showed 167 pairs were very similar. Garner is founder of HelioText, which provides such text analytics services, and helped create the text similarity engine used in the study. They got the summaries from NIH, NSF,DOE, DOD and Susan G. Komen for the Cure publicly-available data.

What they didn't have access to were the full grant files so all they could match was description similarity. That means, said Garner in his statement, the numbers could be far higher. "It is quite possible that our detection software missed many cases of duplication. If text similarity software misses as many cases of funding duplications as it does plagiarism of scientific papers we've studied, then the extent of duplication could be much larger. It could be as much as 2.5 percent of total research funding, equivalent to $5.1 billion since 1985."

Eugenie Samuel Reich and Conor L Myhrvold, writing in Nature, used Freedom of Information Act requests to get examples - including the ethanol example I cited above - and it is absolutely worth a read to see the extent of the problem.

167 researchers out of tens of thousands is not a lot, to scientists, but we live in a modern culture where millions of law-abiding gun owners are going to be penalized because of a few criminals, which many people in academia applaud, and an actual criminal committed suicide due to being prosecuted, which many people in academia expressed alarm about - it's the exact same government in both cases.

If researchers don't adopt internal accountability and circle the wagons around nonsense, the way they did about Coburn's report on waste and fraud, politicians will take control - and it will be the same sort of political theater, posturing and sound-bite yelling about science funding that we see about everything else.  

Funding agencies do sometimes catch duplication, but they are bureaucratically archaic - legacies of a Cold War mentality about science agencies that we refuse to fix.  The researchers that were part of the overlapping funding proposals fall back on the idea that they need 'clearer guidelines', which shows that many of the successful people in the government-controlled science industry know how to play the game as well as anyone else in politics.

To ethical people, the guidelines are obvious.