science commentary

couple years ago a paper came out in Science that claimed the oceans were rising due to global warming. Upon reading it, it was evident that the scientists had actually found that the sea levels were dropping, which violated the contemporary dogma. To get the required result, the authors had to add a correction factor for rebound from the retreat of the glaciers after the last ice age. The rebound, of course, could not be measured, so they ‘estimated’ it with a computer simulation.

Now a panel at NASA is repeating the claim. http://news.yahoo.com/global-sea-levels-climbed-3-inches-since-1992-201058029.html A few years earlier, Daniel Henninger, commenting on the Climategate scandal, famously had said that science was dying. I was skeptical, but when I saw that Science paper I breathed a sigh of despair: maybe Henninger was right.

Would that paper have been accepted by such a prestigious journal if the authors had not found a way to make it support global warming? That this question can be asked shows us the problem: more and more, science must be twisted to fit a political narrative in order to be accepted.

The Magic Circle of Science: More money = more cures, but politics puts brakes on everything.

Twisty science generates a feeling of malaise among scientists, who detest producing it, and among skeptics frustrated at seeing their tax dollars wasted. The malaise doesn't just affect climatology: scientists in all fields feel under siege. And everybody and his uncle seems to have a solution.

Science is a positive feedback loop: new knowledge, if replicated by other labs, leads to new theories, new ideas, and more knowledge (see figure above). New products and cures are a side benefit that leads to increased funding. But each step in the circle suffers when politics steps in, and we scientists are up to our earballs in politics these days.

Critics point out that we're producing fewer and fewer new drugs and new inventions. The critics' rallying cry is the “increasing disparities between expenditures and deliverables”^[1]. You might never see that expression on a protest sign, but when scientists hear the word ‘deliverables,’ they know what it means: more bureaucratization to ensure ‘accountability.’

This is not the Way of science. Science is an anarchy, and it can only thrive so long as it remains one.

Intellectual property (IP) laws in the USA selectively benefit large corporations; as the inventors of television and FM radio discovered, owning a patent does you little good when a company can effectively invalidate your patent by stealing it and hounding you into bankruptcy.

Knowing this, most scientists and inventors work for corporations or universities. Unless they strike out on their own, they receive little or no financial compensation for an invention that may produce billions of dollars of value; to the majority of scientists the only value a patent provides is as another line on their CV.

Industry conducts little basic research; it is what economists call a free-rider on academic labs, yet they complain that its ‘reproducibility’, however defined, is not sufficient for their needs. A few scientists become fabulously rich from starting their own company or patenting a discovery, but they are regarded as failures by their colleagues for having lost their way. The vast majority of scientists don't want to start a company and pity those who failed and became administrators, CEOs, and multibillionaires. A friend of mine who I knew as a postdoc is now the director of a well known research institute, and though he probably makes ten times my salary, those of us who remained in science speak of him as if he were dead. Another postdoc friend became a vice-president at Pfizer. Poor guy, I knew him well.

I have yet to meet a fellow scientist who values patents, inventions, or practical discoveries, which go by the derogatory term “methods,” more than publications. The goal of most of my colleagues is not to invent a new method, but to discover some important fact about nature. Until the H-1B crisis hit, most of them received a reasonably good salary for doing this and regarded it as liberation from the burden of worrying about sordid matters like money. Despite the claims by some companies that more and more H-1Bs are needed, the flood of foreign workers has devastated the career prospects of many. Over 60% of today's postdocs in the USA are here on temporary visas. Few of them will ever achieve tenure.

The source of funding encourages this mindset. For most academics money comes from federal grants; thus their true employer is the federal government. In bioscience the seat of power is in Bethesda, Maryland, where the broad goals of the feds are turned by an army of bureaucrats and administrators into specific program announcements or PAs.

These PAs are scrutinized by academic scientists seeking funding, and they represent the desires of the public. If the people want more cancer research, more funds become available for oncogenes and cell metabolism and more projects in biomathematics and kidney diseases get mothballed. If the news media exaggerate the risk of AIDS or new infectious diseases, the money flows away from neurology research.

In science the measure of value is not physical products produced, but the number of papers published and grants received, or in other words the number of resources consumed.

This was not scientists' doing. It is the fault of administrators and HR departments which are constantly searching for the best metric for hiring and promotion decisions. Their solution is ‘impact’. Impact = Number of Publications × Journal Impact Factor. This is an inaccurate measure of value, but it's easy to calculate.

Thus, our true deliverables are publications and resources consumed, and we've been successful at maximizing them. If someone wants to reduce the disparity between expenditures and deliverables, they need to consider how to change the metric used by the administrators. Don't bother complaining at us—we already know the metric is bogus.

The H-1B crisis made things worse for science by flooding the universities and corporate labs with cheap foreign imports. As job opportunities dried up, American students specialized more in non-science majors, creating a vicious circle. The flood of foreign scientists harms the foreigners as well, many of whom complain bitterly about the difficulty of finding a good position. When they're forced out of science, many return in disgrace to their homeland, where they are cooked and eaten (the guy who told me that had a thick accent, so I might have gotten that part wrong).

These problems are not created by scientists, but by the government. Yet since academics are de facto government employees, they are reluctant to criticize it.

As Paula Stephan wrote in How Economics Shapes Science researchers think almost constantly about money. Funding from nonprofits is minute by comparison and tends to be narrowly focused on popular causes like specific diseases. Funding from corporations is spotty at best and treated with suspicion by administrators who fear that the corporate sponsor will grab the lucrative IP rights. (Some nonprofits also grab IP rights, since it is their main source of revenue.)

Writing in 2012, Stepan noted that grad students and postdocs constitute a ready supply of cheap, temporary labor for faculty. Schools congratulated themselves in having placed the student, even though NSF data showed that only 12% ended up in tenure-track positions. The situation is far worse today, with postdoctoral positions increasingly scarce. Unlike in other professions, in science losing one's job usually means having one's career and one's dream destroyed and a decade of expensive and difficult training gone to waste. What positions are available go to cheap foreign imports: Stephan says in 2008, 60% of postdocs were here on temporary visas. One could reasonably argue that in today's economy only a fool or a fanatic would go into science. (Before anyone starts speculating, I fall into the latter category.)

Although the book was published in 2012, most of the data are from 1995 to 2005. Much of the conclusions are out of date, suggesting that Stephan had trouble getting it published. That might be because such a grim assessment is not what people want to hear. But if anything, she's over-optimistic.

The net result is the bargaining power of individual scientists has declined. Calls to reduce or eliminate tenure are music to the ears of school administrators who pad their offices with more bureaucrats and invent more rules and paperwork for the researchers to follow. Example: we are no longer permitted to put beakers or jars on a shelf higher than 52 inches, lest they fall off and break and hurt somebody. Yet the lowest shelf is 52½ inches above the ground.

Critics of tenure are trying to solve the problem of academic bias. But if they get their way, science gradually transmogrifies into advertising and the search for revenue. Eliminating tenure would kill academic research. A university gave me an adjunct professor position one time while I was working somewhere else, and I don't even put it on my CV.

Without tenure, there would be only researchers and adjuncts, and they would have little contact with each other. The power would flow from the individual researcher to the boss or administrator. The researchers would have to discover whatever their boss wanted, whether it was real or not, to avoid losing their job. They'd give the boss phony statistical analyses, data that have been cherry-picked to hell and back, and discoveries that can never be replicated. I've seen the social dynamic in action time and again. Where there's no tenure, short-term planning, quick results, and infighting are the norm. If you think credibility in science is a problem now, wait until tenure is abolished.

Those bureaucrats have power to screw up the Magic Circle of Science, and they use it by increasing the risk to individual scientists. As a result scientists become more reluctant to take risks in the science itself, which means more crowd-following behavior. This in turn leads to the rise in cliques like those in climatology and nutrition science, where politically connected scientists grease the skids for their pals and block publication for their enemies. Especially at nonprofits, the bosses are political appointees, and they run their institutes like fiefdoms. (I can't say publicly what scientists think of them.)

I know one scientist at a nonprofit who was invited to start a crowdfunding of his research project. When he mentioned it to his boss, the boss was enthusiastic. “By all means!” the boss said. “Write up a proposal on my project and let's see if we can get some more funding!” Needless to say, the scientist didn't bother.

In such places, the pre-eminent focus is the organization's (i.e. the boss's) project. Those who criticize it quickly find themselves unemployed. Publications from such institutions have the appearance of science, but are in fact little more than paid advertisements. Nonprofits are especially guilty of this, and research incentives in such places are low—in many places no share of any royalties goes to the inventor, so though they produce many patents they are often of low value.

The most corrupt branches of science—climatology, nutrition, and a few others— are those where making any statement has real-world consequences. If you give incorrect nutritional advice, people will die. If your climate model is wrong, people will make bad economic decisions. The status is low and the pressure to conform is high, so the best scientists avoid the field altogether. Outsiders are loath to intervene because of the cost of being wrong.

By contrast, the least corrupt branches are where the cost of being wrong is low. If a physicist claimed, for example, that a proton only had two quarks instead of three, or that photons were not really waves, the only consequence would be ridicule within the community. Those who say that physics seems (or up until recently, seemed) to have less fraud than biology because biology is a ‘softer’ science are mistaken. Unlike physics (with a few notable exceptions), biology has a direct impact on human life. If someone claimed that the conjunction of Mars and Jupiter caused cancer it could cost many lives. If they claimed that it caused the Big Bang, it would have little impact on our daily lives (except possibly for the second-hand smoke caused by being burned as a heretic).

All scientists know the difference between good and bad science. Political interference and well-intentioned reforms force many of them to do bad science. They are not happy about that. And the last thing the world needs is more angry, pissed-off scientists.

References

[1] http://www.pnas.org/content/early/2015/08/12/1504955112 Increasing disparities between resource inputs and outcomes, as measured by certain health deliverables, in biomedical research Anthony Bowen and Arturo Casadevall. Also cited in this PJ Media article.