The first principle of science is that you must not fool yourself and you are the easiest person to fool.
The passing of Martin Fleischmann on August 3 brings to mind one of the most famous (and infamous) recent episodes in the history of science: the cold-fusion fiasco. Those of us who were involved in science in March 1989 (or just paying attention to the news) could not avoid hearing about Stanley Pons and Martin Fleischmann of the University of Utah supposedly producing nuclear fusion at room temperatures. Announced with great fanfare and huge press publicity on March 23, 1989, Pons and Fleischmann claimed that they had run simple electrochemical experiments in the lab that had managed to start a nuclear fusion reaction. Conventional nuclear physics had always argued that fusion (squeezing two or more atoms together to produce another atom) could only be produced at extreme temperatures and pressures, such as the fusion reactions that occur now to drive the heat engine at the core of the Sun. This is why nuclear fusion has only been used for hydrogen bombs and is not yet practical for nuclear energy or other peaceful uses. If Pons and Fleischmann were right, their discovery would revolutionize nuclear physics and possibly solve our energy problems with a source of energy that was cheap to obtain and not as dangerous as the various proposals for fusion reactors. The media immediately ran with the story on a massive scale, so that nobody who paid attention to the news at that time could possibly miss the message.
Of course, when any scientist makes claims that are so far out of the mainstream yet seem potentially legitimate, it is a challenge to other scientists in their field to test their results. Most scientists were skeptical that true cold fusion had been achieved, especially since the announcement was made (apparently due to pressure from the President of the University of Utah, competing with BYU) without peer review of the paper, which had not even been published yet. Today, when someone makes a startling announcement, scientists expect the peer-reviewed scientific paper to be soon available so they can check the science for themselves—and this precaution was largely a result of the cold-fusion debacle. Back in March 1989, many scientists in many different labs around the world dropped everything they were doing to try to replicate the experiment. Even though there was no widespread use of the internet back then, scientists were chatting back and forth over the phone about their efforts and results (or lack thereof), and by the time the paper was finally published in Journal of Electroanalytical Chemistry (and NOT in Science or Nature, as the work’s importance would suggest), many of them were even more skeptical. Then when the paper came out, they jumped all over it, and within weeks there were numerous critiques and rejections of the research as fraudulent, sloppy, incomplete, unreproducible, inaccurate, erroneous and unethical. Within months after the initial announcement, the work was widely denounced among the scientific community and no longer taken seriously. By 1992, both Pons and Fleischmann moved from Utah to the IMRA lab in France, where they were less notorious and could do research in a private lab with corporate funding and less scrutiny from the scientific community. Sponsored by Toyota, the lab spent over 12 million pounds with no results before it closed in 1998. Fleischmann continued to work in private labs, or US Navy labs, where his research was still considered potentially important. In 1995 he returned to England, where he served mostly as a “Senior Scientific Advisor”, and by 2006 he was suffering the effects of Parkinson’s disease, diabetes, and heart disease. He died this past August 3 at age 85. His student Stanley Pons (now 69 years old) still lives in France and works in the same lab, even giving up his U.S. citizenship.
Although cold fusion was a failure, it is an important object lesson in the nature of science. Many of the mistakes and problems that we saw with the cold-fusion experience inform other similar problems in scientific research and its acceptance by a scientific community.
1) Never do science by press release. In many fields, there is enormous pressure from institutions and sometimes from publicity-conscious scientists to promote their research to the hilt. As much as scientists would like to see their results get the attention they deserve, too much PR and hype is a bad thing. It always raises a red flag in the minds of most scientists, and also serves as a red cape to bait them into focusing on checking your results. In most cases, the highly promoted discoveries turn out to be not only less than claimed, but often just bad science. One thinks of the recently debunked claim that there are bacteria in Mono Lake who are build their DNA not with phosphorus but with arsenic; the overhyping of the alleged “impact” 11,000 years ago and its supposed effect on mass extinctions of Ice Age mammals (more on that in a future column); or more recently, the big hullabaloo over the Messel fossil Darwinius, which turned out just to be a nice articulated specimen of an adapid primate, and not even closely related to anthropoids (also the subject of a future column). Typically, the PR and development people at a university or foundation are always searching for ways to garner publicity and get their scientists into the media. To some extent, that’s fine, but as soon as it becomes an overhyped frenzy of attention, it’s sure to enhance the skepticism of the scientific community and guarantee the quick debunking of the claim if it can’t stand scrutiny. In the cold fusion case, it was the President of the University of Utah who pushed Pons and Fleischmann past their comfort limits, and ended up with with egg on his own face as well as embarrassing his scientists and his institution.
2) Pathological science. Even though there are still diehards out there who still claim that they have achieved cold fusion, most of their work is not published in peer-reviewed journals, and is now regarded as “pathological science.” Such a status is typical of a hot scientific idea that achieves a burst of fame and attention. Then when it fails to produce results, most scientists turn away, leaving only true believers behind, who pursue their research outside the mainstream science community. If you look hard enough, there are remarkable number of symposia conducted and volumes published within the diehard community, all to maintain their faith that they were right all along and that science has rejected them unfairly. In this respect, they closely resemble the followers of various millennialist cults and religions, who sold all their possessions to be ready for the Rapture at a predicted date, only to be disappointed again and again. As psychologists have shown, the human brain works in such a way that even though proven decisively wrong in their predictions, confirmation bias and reduction of cognitive dissonance works so that the “true believers” become even stronger in their falsified belief. They use classic ad hoc rationalizations (“the Rapture came but it was invisible” or “our prayers saved the world from destruction”) to continue to follow their belief system, because the emotional connections with the believer community override the rational portion of their brain that tells them they were wrong. So, too, do the pathological science communities such as the cold fusion acolytes, continue with their delusions.
3) Science is always scrutinizing and testing its claims. Not every claim in science gets a huge amount of publicity or immediate cross-checking by other rival scientists, but the really important and startling ideas do, often right away. The Pons and Fleischmann claim was completely dead within months of its publication. Similarly, the outlandish claims of impacts causing certain extinctions (such as the Permo-Triassic or Triassic-Jurassic) were debunked within months by people who went back and did the dating and geochemistry that the original author neglected to do. In some cases, it may take a few years before outrageous claims are fully tested and debunked in the published literature (often due to the slow pace of much scientific publication), but eventually bad ideas are winnowed out.
4) Peer review works. The startling claims of cold fusion advocates were made without peer review, and the authors were heavily criticized for publishing their results with minimum peer review in a small journal with less stringent standards than the top-line journals. Eventually, when they published results that any scientist could try to replicate, the peer review system showed their claims to be impossible, and the idea was falsified. Not every bad scientific idea is weeded out by peer review (most are only corrected years later, and some may survive scrutiny for even longer), and sometimes peer review is too strict and unfairly rejects good ideas that are out of the mainstream. But science is a human enterprise, with human failings and feelings built in. Given that this is true, science has the unique distinction of being the only system that is so heavily self-scrutinizing and self-correcting. No religious, economic, or political theory of the world can make this claim.