M-c Shanahan

In a recent post on the Scientific American guest blog, I looked into some of the communications issues that arose as result of critical comments about the NASA research that were posted online. This post is a bit of a companion post, revisiting some of the questions that I asked there but looking at them from a slightly different perspective.

On January 6 at BoingBoing, Andrea James wrote about similarities that she saw between recent science news (such as NASA’s aresenic-DNA story) and Pons and Fleischmann’s announcement of cold fusion. She argued that the culture of ‘firsting’ – in this case, trying to be the first to get a new story into the public – can lead journalists to cast a less critical eye towards scientific announcements. This effect is amplified when science communication is carried out by press conference. Making a link between the cold fusion story and NASA’s announcements caught my eye because it is a comparison that has been rattling around in my head for a little while as well.

To be clear, I don’t think that James is comparing the quality of the science. The general retrospective view of Pons and Fleishmann’s work is that it can best described as pathological science  – exhibiting not necessarily fraud but delusion on a grand scale. The term comes from a talk given by Nobel laureate Irving Langmuir at the Knolls Research Laboratory in 1953 (and appropriately republished in Physics Today in 1989). Langmuir defined pathological science as “cases where there is no dishonesty involved but where people are tricked into false results by a lack of understanding about what human beings can do to themselves in the way of being led astray by subjective effects, wishful thinking or threshold interactions.”[i] That is a serious and damning view of the science of cold fusion. In making comparisons to the communication culture surrounding cold fusion, neither James nor I are attempting to make a similar claim about the NASA arsenic-DNA paper.

While the science may not be similar, the science communication issues that have emerged do seem to intersect.

Much of the early criticism of the arsenic-based DNA paper was communicated rapidly and disseminated widely through blogs, Twitter and other online forums. Some were written by scientists in the field responding directly on their research blogs (such as Rosie Redfield from the University of British Columbia) and others were written by journalists who sought out the comments of experts and reported their concerns. The authors of the paper and NASA spokespeople, however, argued that blogs and other online space were not appropriate for science criticism. They argued that any concerns should be addressed only through peer-reviewed letters to Science, the journal that had originally published the research. Hearing this gave me a twinge of recognition. I thought, “I’ve heard part of this story before…”

On March 23, 1989 the world’s press gathered for an excitement-filled press conference held by the University of Utah. Two scientists, Martin Fleischmann and Stanley Pons, claimed to have discovered a nuclear fusion process that could be produced simply and reliably in a regular laboratory – cold fusion.  Taking this finding to press conference before it was accepted for publication in a scientific journal was considered a breach of the usual practices of science but the researchers and their press officers argued that it was necessary because of the potentially lucrative patents involved and because there was another nearby research lab working on the same problem. While Pons and Fleischmann, and researchers from Brigham Young University, submitted their results to Nature for review, news reports continued to highlight positive results flowing in from around the world.

This success did not last long. By the time the American Physical Society met in May, several labs had announced that they were unable to reproduce the results and even worse that any positive results they had seen earlier seemed to have been caused by improperly calibrated instruments and incorrectly interpreted data. Pons and Fleischmann also withdrew their paper from Nature’s review process. The consensus against cold fusion grew from there, and the controversy was mostly dead by November 1989 when the Energy research Advisory Board released its final report recommending no further funding for the University of Utah’s cold fusion program[ii]. How did the star fall so quickly?

Well, 1989 might seem like the dark ages of online communication – no Twitter, no Facebook, no blogs – but it turns out that an informal network of electronic communications played a pivotal role in the quick and negative consensus. While the media frenzy continued and government committees were convened, scientists were using electronic back channels to communicate with each other. CERN physicist Douglas Morrison set up one of the many electronic newsletters read by enthusiasts and sceptics alike. His was called “Cold Fusion Newsletter” and it was sent by email to a network of colleagues and posted to the sci.physics.fusion newsgroup.

Similarly, announcements of supporting and contradictory findings were quickly distributed to relevant labs by fax. The Princeton Plasma Physics laboratory, for example, heard about Japanese findings matching those of Pons and Fleischmann when they were sent a faxed copy of a Japanese newspaper report with a short handwritten English translation: “4/1/89. Yomiuri Shimbun (Japanese biggest news paper). Tokyo Agriculture & Engineering University announced ‘The [Koganoi-shi] re-produced the Utah experiments’. They measured heat, gamma and tritium. Prof. Koyama.”[iii]

These informal networks were initially created to share information about the experimental details, providing tips and ideas for those trying to replicate it. Later they became the main distribution network for conflicting results and emerging questions about the plausibility of cold fusion. It was Morrison who used the Cold Fusion Newsletter to first turn his colleagues’ attention to Langmuir’s infamous lecture, suggesting there that cold fusion was a case of pathological science.

So how was this informal source of criticism received by the scientific community? While Pons and Fleischmann mostly avoided responding to these critiques (they also didn’t attend the American Physical Society meeting), the research community depended on this informal network to come to a consensus about cold fusion. The only serious communication criticisms were directed at Pons and Fleischmann for attempting to duck the peer review process and communicate their results through mass media instead. Tom Instrator, a nuclear engineer at the University of Wisconsin, was unequivocal, “that’s not the way you do science… If they want [other scientists] to not give them a hard time, they should give us enough to copy it. Or maybe they have given us enough, and it doesn’t work.”[iv]

This is something that I find interesting. It’s easy to understand why Pons and Fleischmann’s communication strategies were heavily criticised. They provided information almost exclusively through mass media and baulked when pressed to publish in peer reviewed journals. This resistance probably caused the electronic underground network. With this startling lack of technical data, it’s no surprise that scientists turned to other sources, but looking back I wonder why there weren’t also questions about whether this informal network was reliable. Electronic communications were relatively new, and unverified claims (such those of Japanese labs replicating the results) were passing quickly from person to person. The deluge of information from every angle (faxes, emails, newsletter) was described by many scientists as overwhelming – but no one suggested the it was illegitimate in the way that NASA suggested that the blogged and tweeted comments were. Why not?

I think the key factor is one of boundaries. The blogged comments were public, blurring boundaries that mark the usual sources and usual audiences for scientific critique. The cold fusion network, on the other hand, was insider only. To get access to the newsletters, faxes and emails, you had to be on the inside already. At the time, Morrison stressed that “they are an informal network, they are meant to be academically confidential…It’s only academic people. I don’t give it to the press.”[v]

NASA complained that the critiques of its arsenic-DNA study were inappropriate and that anyone with a serious critique would formally publish it through the peer review process. Morrison’s comment though suggests that it’s not so much the formal peer review process that matters so much but setting a clear boundary between those who have access to contributing and reading the critiques, something that blogs don’t do. This continues to make me wonder whether audience and author boundaries weren’t also the issue that NASA was addressing. Can we read between the lines to wonder if it wasn’t so much that the critiques weren’t peer reviewed, but that they were public?

Apporva Mandavilli makes a similar point in her essay in Nature, called “Peer review: Trial by twitter”. She points out these informal conversations have always happened but they’ve usually been in private – at conferences, in the lab and in correspondence between friends and colleagues. Putting them out in public, and in large quantities, can make them overwhelming, intimidating and difficult to filter. As public forums like Twitter and blogs continue to grow as a space for informal critique, there will continue to be cries of resistance and demands that the peer review critique process be respected. The case of cold fusion I think provides an important reminder about what this resistance is more likely related to – moving critique outside of the boundaries of the private community. I’m not arguing against the importance of peer review or of formal journal publications but just wondering about how we will make sense of the growing public media. To finding ways to work with and understand the impact and possibilities of online communications, do we have to dig a bit deeper to make sure that we know what the issues really are?