These scientists got sued for doing their job (Patreon)

[This is a transcript with links to references.]

Truth can be painful, but truth can’t be sued. So sometimes, people who’ve been hurt by truth will try to sue those who dug up the truth. This doesn’t only affect journalists, who often have legal backup from their employer, but increasingly often also scientists, who usually have no legal support. This worries me considerably. What are scientists being sued for, why is it happening, and what can be done about it? That’s what we’ll talk about today.

I witnessed this problem first about 15 years ago in a very unlikely place, a talk by a professor for phonetics in Stockholm. Phonetics is the study of how humans produce and perceive sounds, and how we articulate speech. You might think vocals and consonants can’t be all that controversial, but think again.

The Stockholm professor and a colleague of his from Gothenburg came across a company, Nemesysco, that sells voice-analysis technology. According to the company’s website, this technology “enables better understanding of your subject’s mental state and emotional reaction at a given moment by detecting the hidden emotional cues in his or her speech”. They claim that their method “identifies various types of stress levels, cognitive processes, and emotional reactions that are reflected in different subtle properties of the voice.”

The phonetics researchers looked at this and found that the type of speech analysis doesn’t even provide sufficient information to make such inferences. The case is particularly clear, they said, for the identification of lies, because this has been very thoroughly studied. They wrote a paper called “Charlatanry in forensic speech science” that was published in 2007 in the International Journal of Speech, Language and the Law. In their paper, they looked at the technology that the company was offering at the time and wrote that, quote, “these machines perform at chance level when tested for reliability”, end quote.

The company pressured the journal’s publisher to withdraw the paper and also threatened to sue. They said it wasn’t so much the content of the paper that disturbed them as the language. In particular they didn’t like being called charlatans, and you don’t need a voice analyser to “identify various types of stress levels” in that response.

In response, the publisher removed the paper from the website and instead put up a comment from the company. In an interview with the magazine Science, they publisher’s managing director said they simply don’t have the resources to put up a legal fight. Indeed, the journal has so few readers we’d probably never heard of the paper if the company hadn’t made all that effort to prevent you from hearing about it.

Then the Swedish research council got involved and a lot of people got very upset, and in the end nothing happened one way or another. No one was sued, the company continued selling their product and even that the journal removed the paper from their website had few consequences since other people uploaded the PDF elsewhere and you can still find it online.

Now, I know nothing about phonetics other than the German “ü” is supposedly very nice to sing which is why there are a lot of “ü”s in the German National Anthem. But for today the point isn’t who is right or wrong. The point is that those were scientists doing their job and a company didn’t like the result. The company had lawyers, the scientists didn’t. Though I’m now afraid that singable ü’s is the only thing you’ll remember from this video.

Around the same time that the Swedes ran into problems with the voice analysis company, a very prominent lawsuit unfolded in the UK, involving the science author and journalist Simon Singh who also happens to have a PhD in particle physics.

Singh had written a column for the Guardian about a book he’d just published with Edzart Ernst about false promises of alternative medicine. In his column, he took on chiropractors in particular. Chiropraxis is the idea that you can cure all sorts of illnesses by readjusting the spine. In his column, Singh wrote:

“The British Chiropractic Association claims that their members can help treat children with colic, sleeping and feeding problems, frequent ear infections, asthma and prolonged crying, even though there is not a jot of evidence. This organisation… happily promotes bogus treatments.”

British Chiropractors were not amused, and set out to demonstrate that you should not mess with people who crack joints for a living. Several hundreds of them complained to the British Chiropractic Association, an event that scientists on twitter aptly dubbed the “Quacklash”. The Chiropractic Association sued Singh for defamation and the Guardian removed his column. Again, they raised an issue not with the scientific case, but with the language, claiming that Singh had accused them, not of being wrong, but of being deliberately dishonest.

In a preliminary hearing, the court ruled against Singh. He appealed, saying he hadn’t claimed their dishonesty was deliberate. Singh won, the Chiropractic Association dropped the case, and the column went back up. You can still read it on the Guardian website. Never underestimate a physicist.

Now you might think that Singh’s major problem was really that he blasted out his opinions in the Guardian and that doesn’t happen all that often, but the problem that insulted parties come after scientists is more widespread than you might think.

Take for example the curious case of a German-Ukrainian scientist by name Leonid Schneider. He has a PhD in biology and writes a blog called “For Better Science”. Schneider writes mostly about biomedicine, and I don’t understand half of it. I really just read it because there’s quite some drama on his blog. Publishers and scientists threaten to sue him all the time. Sometimes, they do. For example, some years ago, he became unwittingly involved in the case of Paolo Macchiarini.

Macchiarini is the doctor who pioneered trachea implants treated with stem cells. In 2010, he was hired the Karolinska Hospital in Stockholm where I once spent a very unpleasant night, but which I am told is otherwise a top place. At Karolinska, Macchiarini performed a series of trachea surgeries.

Unfortunately, it turned out, his procedures were more experimental than his patients knew. Many of his operations resulted in complications and death of the patients. Macchiarini did not fully disclose the details to anyone, so no one besides him really knew what was going on. Not his employer, not his collaborators, least of all, the patients.  

In 2014, former collaborators raised the first concerns about some reports that Macchiarini had published, which they claimed did not accurately reflect reality. The first investigations cleared Macchiarini, but closer study of the cases increasingly raised eyebrows. In 2015, the Swedish Research Council stopped funding him. Several people at Karolinska, who had been involved in his recruitment or had vouched for his credibility, resigned. In 2018 the Karolinska Institute ruled that Macchiarini and six other researchers involved in the cases were guilty of scientific misconduct. Six papers had to be retracted. Macchiarini went on trial for involuntary manslaughter but was not convicted.

The lesser known part of the story is that Macchiarini had former collaborators in Germany, Heike and Thorsten Walles who also worked on trachea implants, though with a somewhat different technique, and that brings up back to Schneider with his blog. In late 2016, he mentioned that some papers of the Walles’s, who are a married couple, also seem to be problematic. They swiftly sued him, demanding that he removes some content from his blog, threatening a fine of 250 thousand Euros if he dares mention it again.

The lawsuit ended on a settlement, though Schneider was stuck with legal costs of about 12 thousand Euro.

Heike Walles was found guilty of research misconduct years later, in 2022, though in contrast to the Macchiarini case that made a big splash in the media, the Walles case in Germany was basically not heard of. Schneider has been sued several other times for drawing attention to problems with papers in the scientific literature.

Fast forward to today. A current prominent case is that of a psychology professor at Harvard Business School who has been accused by other scientists of multiple cases of fraud and academic misconduct.

If you saw the headlines two years ago about a paper about honesty that was retracted over false data, that was the beginning of it. The three psychologists, Leif Nelson, Joe Simmons, and Uri Simonsohn, who raised doubts about the paper write a blog called Data Colada. That’s not me mispronouncing collider, but data Colada as in pina Colada. The three are quite famous in the field, because they kickstarted in a quality revolution in psychology that I have been mightily impressed by.

One of the authors of the honesty paper is Professor Francesca Gino from Harvard Business school. The three psychologists found further irregularities in three more papers that Gino co-authored and raised concerns about this to Gino’s employer.

Harvard Business School conducted an investigation and put Gino on administrative leave. Right after that, in June 2023, the three psychologists published a series of blogposts in which they explained the problems with the papers in detail. For example, in one study there are 35 suspicious responses in the data table that supposedly come from study participants, had an entry just saying “Harvard,” and that push up the significance of the finding.

The professor then sued the psychologists who accused her of fraud. Their legal expenses will likely reach hundreds of thousands of dollars. You can support them on go fund me, link is in the info below.

Again, my point here is not who is right and who is wrong. This isn’t my field, though to me the evidence against Gino looks very damning indeed. The point is that you have scientists doing their job and they are being sued for it.

Another recent case is related to a prominent trial in the UK, that of Lucy Letby. Letby was a nurse in a neonatal care unit. In August this year, she was convicted of having killed seven babies and sentenced to life in prison. But before I tell you what this has to do with scientists being sued, I need to tell you about another case, that of a paediatric Dutch nurse named Lucia de Berk.

In 2003, de Berk was convicted of four murders and three attempted murders and sentenced to life in prison. A main piece of the evidence against her was the unusual high death rate of babies in her care.

Enter the statistician, Richard Gill. In 2006, Gill became interested in the case, looked at the numbers and found that such a cluster of cases could have occurred by chance. He mobilized colleagues, including the Nobel Laureate Gerard ‘t Hooft. The Dutch nurse was exonerated in 2010. Gill also helped work out the statistics for a some other such “cluster cases”. Naturally, he was very interested in the Lucy Letby case.

With the Letby case on trial in the UK and Gill in the Netherlands, he wasn’t directly involved but wrote about it on his blog, criticising in particular that the court did not draw on any statistical analysis, but he also commented on medical details and other evidence.

Alright, you might say, but there must be thousands of people on social media doing exactly the same that. Probably correct, but either way, the British court was not amused. Given that the UK left the European Union you might think there’s not much they could do about Gill’s blog anyway, but it turns out that Richard Gill is a British national.

So he got emails and letters from the British police, demanding that he removes certain content from his blog. The letters say that “the material is a flagrant and serious contempt of court” and that “if you come within the jurisdiction of the court, you may be liable to arrest”.

I know nothing about the case, or the science involved, but again this isn’t the point. The point is that Gill was making a scientific argument, and that was deemed “contempt of court.” This law exists because of the concern that such discussion might influence the trial, and yet such a law goes completely against the way that scientists are used to dealing with evidence.

Those cases are all very different, but I think they have one thing in common, it’s that we have a big clash between the way that scientists deal with evidence, which is to dish it up and discuss it – not always very politely – and the way that the jurisdiction deals with it, which is by telling people what they’re allowed to say and when. Personally, I think that if a court is concerned that someone’s blog might influence the judgement of the jury, then there’s something wrong with the court proceedings and not with the blog, but maybe that’s just me.

Underlying this is a bigger problem. It’s that we currently have no good way for integrating scientific evidence into either branch of the government. I talked today about the judicial branch, but the same problem exists for the executive and legislative branch. At the moment, we do this by a patchwork of committees and advisors and institutions and reports. Sometimes that works and sometimes it doesn’t. This isn’t a good procedure and I think that if democracy is to survive, we need to find a systematic way to build a scientific evaluation of evidence into our decision making.

The quiz for this video is here.