Picture















Brian Deer at Westminster Skeptics. Photo: blahflowers ( http://www.flickr.com/photos/blahflowers/4596824668/)




“scientists are no more trustworthy than restaurant managers – whose kitchens are randomly inspected to protect the public.”

Brian Deer made this assertion in the Guardian yesterday. I knew it was coming: he had mentioned it at a meeting of the Lewes Skeptics in the Pub last month. I was there, as were several senior research scientists. They tore into Deer over the comment.

The scientists were quick to point out that scientists have to be trustworthy because truth will eventually out. When you do research, your reputation will be destroyed by any fraud discovered – and discovered it will be: all it takes is for other people to repeat the experiment and report results.

Deer countered that scientists had not found out Andrew Wakefield: in fact, he said, they had colluded in the fraud, with the journal’s editor doing everything he could to block Deer’s investigation for the Sunday Times.

I was fascinated by the discussion because both sides made good and useful arguments, many of which I had already delved into in researching Free Radicals.

Andrew Wakefield, I would suggest, was much worse than a bad restaurant manager.

When restaurant managers cut corners on hygiene, they are not trying to poison anyone; usually they are simply practising strategic risk-management. Successful restaurants don’t need to take such a risk, but for a new or struggling restaurant, the money saved in the short-term can make a food poisoning outbreak a risk worth taking for the sake of the business. Restaurant managers don’t take risks to make millions, they take risks to keep their business afloat. Perhaps to ensure they can pay the mortgage this month. Or their staff.

Wakefield, on the other hand, was selfishly motivated. He knew his theory had no validity from intuition or supporting evidence. He simply stood to make a lot of money and didn’t care who suffered along the way. It was common or garden greed.

But it is actually hard to find cases of scientific fraud where money is the motivator – they are the exception. Scientists aren’t actually that interested in money; for most, it’s a practicality rather than a motivation. Scientists are largely after discovering the truth – and discovering it first (that’s all they have, after all; there are no prizes for second place). Ambition – not greed – is the source of most cases of research fraud.

In science, fraud is astonishingly prevalent, as noted in this paper where it is referred to as “Normal Misbehaviour”. But it is motivated by frustrations with the real world, where data is not clean, and use of intuition or hunches is essential to discovery but inadmissible as evidence (see this letter to Nature by Fred Grinnell (£)). Sometimes fraud is motivated by a scientist’s over-confidence in his or her abilities: Einstein heinously cherry-picked data for his determination of the gyromagnetic ratio – he wanted the value that matched his theory, which he assumed to be correct. His theory, as it happens, was wrong. (Interestingly, when the right value was discovered years later, no one made a fuss about Einstein’s “normal misbehaviour” – if you commit fraud in a certain way,  when you are exposed it just looks like you got things wrong, which is entirely forgivable).
It was interesting to note that all the scientists in the Lewes Skeptics audience said they would welcome “data-hygiene” inspections on their labs and notebooks, noting that their funding bodies reserve the right to carry them out. As long as the routine still allowed them to get on with their work most of the time, they saw it as a good thing.

But, unlike with restaurants, such inspections would do little to protect the public because in most cases – not Wakefield’s case, admittedly – these Scientists Behaving Badly (£) are not putting people’s lives and health at risk for personal gain. Deer is guilty of a mistake that the scientific process has taught us how to avoid: making generalised statements on the basis of limited evidence.

It’s understandable, though. Deer has judged science on the basis of its public face rather than its true nature. If scientists are upset with him, they only have themselves to blame.

Science has deliberately chosen to clothe itself in a brand image, applied after WWII (more on that in Free Radicals), that is whiter than white. Brand Science is Trustworthy, Reliable, Logical, Rational, a direct path to better lives for all. The reality is rather different, as any working scientist knows. Unfortunately, science has made the mistake of continuing to embrace this branding, rather than throwing it off.

As the spat at the Lewes skeptics meeting showed, scientists are human, passionate and often fierce. Science is for lions, not lambs – if you don’t believe me, try putting a newborn theory out there and see how happy scientists are to rip it to shreds (check Jenny Rohn on this).

It would be better for everyone if science was comfortable with exposing this humanity. With issues such as climate change on the table, Brand Science can no longer afford to hide behind its self-created mythology. There is too much risk of someone like Deer pulling back the curtain and making it easy for its critics to cry shame when there is really nothing to be ashamed of.
Ian Eiloart
11/5/2011 21:49:08

It's important to make the distinction between the questions "Are scientists trustworthy?" and "Is science trustworth?".

The answer to the first question must be "some are and some aren't", just as with any group of people.

The answer to the second is, in my view "yes, in the long term". But new results must be treated with caution, and evidence must be carefully scrutinised.

Reply

Your comment will be posted after it is approved.


Leave a Reply.