Friday 9 March 2007

Why experts need to speculate, without speculating

The Cumbrian train crash was just one in a long line of stories that saw me and my colleagues asking experts to speak in the media about what went wrong before anyone could possibly know what went wrong. And let me tell you these are the times when you really do need to take a deep breath before calling a scientist, because it’s almost inevitable that one of the experts we call will be seriously outraged at being asked to speculate in the media before the full facts are known.

One of the experts I spoke to on the liquid explosives scare last summer shouted down the phone that the police had asked people not speculate and he expected the Science Media Centre to comply with that request. Others get less cross but are no less strident in their refusal to join the frenzy of speculation that is now a permanent feature of our 24–hour–news culture. One of the train experts I called this week insisted that no self respecting scientist or engineer would take to the airwaves without access to the full facts on what had gone wrong.

This is particularly familiar territory for the Science Media Centre. After all, we were set up five years ago precisely because leading figures in the scientific community recognised that the absence from the media of the best experts on issues like BSE, GM crops and MMR mitigated against a balanced and accurate debate. So asking scientists to enter the fray before the full facts are known is kind of part of the job description. It also happens to be something that I now feel passionately about.

Breaking news is, by its nature, based on speculation and guess work –from the horror of six healthy men on the verge of death after a clinical trial went wrong to the images of the poisoned Russian dissident to the outbreak of bird flu in our turkey farms.

But speculation and guesswork is not what scientists do best. And of course you could argue that the scientists refusing to speculate are simply displaying all that is special and distinctive about science. The search for truth and respect for evidence and accuracy that drives the pursuit of knowledge by scientists is about as far removed as it could be from the media’s needs at times of breaking stories. I will always remember the sharp intake of breath in a room of 200 leading scientists at the Royal Society when Simon Pearson, news editor of The Times, calmly told them about the first rule of journalism: “to the question ‘do you want it good or do you want it now?’ there is only one answer”.

It also stands in sharp contrast to the culture of politicians, protest groups and NGOs who often see these huge breaking stories as an opportunity to get their messages over and raise their profile. The Soil Association certainly did not let the lack of full facts stand in the way of them using the outbreak of bird flu amongst Bernard Matthews’ turkeys as an opportunity to voice their opposition to factory farming. Similarly news of disasters at GM crop sites or nuclear power stations are seized on by opponents of these technologies with breathtaking speed.

And good for them - campaigners exist to raise awareness of their issues and put them on the political and media agenda. I know, I did it for many years. But it does make it all the more important that scientists and engineers enter these discussions as well.

One of the train experts who objected to my call (and not just because it was 9am on a Saturday!) argued that the integrity of expertise is intimately bound up with the refusal of real experts to imitate other groups by speaking before the facts are known. And there is a beautiful logic in this position. After all, surely the way to counter irresponsible and possibly dangerous speculation by non-experts is to not play the same game.

But it’s exactly because of this integrity and respect for evidence that I want the public to hear from this expert. It’s precisely because when he goes on the air he will refuse to speculate that I want him to do the interview. It’s specifically because what he says will be based on 10, 20, 30, often 40 years of expertise in this field that he will be a million times more qualified to speak on this issue than many of those who happily make themselves available for interview. At times when people quite literally fear that they will die – from Sudan I in our food, ricin on the underground or travelling on dangerous trains – I want to hear about the real risk from real experts, not self appointed ones.

When media headlines screamed out that terrorists were planning to wipe out hundreds by throwing ricin on a packed underground train, many of us thought we could be facing our worst nightmare. Yet the first ricin expert I spoke to told me that, in fact, we could swim in it, eat it or sleep in it with minimal effects and that to kill us it would need to be injected inside our bloodstream (as it famously was with the sharpened tip of an umbrella to kill a Bulgarian spy, Georgi Markov, in 1978). Getting this expertise into the media very quickly changed the shape of that story.

When six men are lying dying in a bed after a pharmaceutical clinical trial and I am already hearing patient groups and politicians on radio talk-shows saying this proves that animal research doesn’t work and that companies are taking risks with people’s lives, I want to hear from scientists who can explain why adverse effects don’t always show up in animal trials and who can balance a horror story with the facts and figures about how many clinical trials take place with no problems. Not knowing what had happened in this particular case needn’t be a barrier to experts sharing what they do know about monoclonal antibodies, first-in-human trials and the importance of both!

And here’s the rub – we never ever ask scientists and engineers to speculate. In fact we encourage them not to. When the media were claiming that Litvinenko was poisoned by thallium, our physicists made it clear that they had no more information about what was killing him than anyone else, but were happy to be interviewed on what they knew for fact about thallium, then radioactive thallium and finally polonium-210. However counterintuitive for the scientists, they did accept our argument that it was better for them to talk to the media at this time, rather than self appointed experts who could fuel scare stories with their partial knowledge.

So what I’m arguing here is that society needs to have access to the best science and engineering precisely at times when these issues are headline news and the public and policy makers are most engaged with them. The truth is that when the full facts are known about this latest crash it will probably merit a 100 word article on page 10 of The Times; no-one will be listening, no-one will be worrying. The time to inject expertise into the public debates about vaccines, clinical trials, train crashes and similar issues is when they are in the news.

The expert that starts an interview by saying, “I cannot speculate on what caused this crash and nobody should until we have the results of the investigation – but I can tell you this about the infrastructure of this model of train…” will be injecting much needed accurate information into a story as well as reminding the public that what is good about scientists is their integrity and respect for facts. The Science Media Centre's philosophy is that “the media will start ‘doing’ science better when scientists start ‘doing’ the media better”. We are not in a position to change the media – but we can respond to Simon Pearson’s first rule of journalism by offering to help the media to “have it now AND have it good”!

11 comments:

loidallen said...

Thanks for a wonderful and informative article.

John said...

The point is surely that in order to speculate, you don't actually need facts. Speculation is based on identifying theoretical possible options, and guessing which is most likely -- this can be, and usually is, done with scant regard for observation and facts.

The Bird 'flu outbreak was interesting, since the lack of any evidence did not stop DEFRA and other scientists speculating that it had come from wildbirds. This was despite the fact that there was little or no evidence that this particular strain was likely to have come from wild birds, or that wild birds were involved in its spread in Asia. Over the years government scientists have generally based attitudes to disease in wild animals on pure speculation. Even now the relationship of badgers and Bovine TB seems to be based on the speculation that it is more acceptable to blame badgers than to blame the other carriers of TB such as brown rats and field voles (or cows).

Andy Challinor said...

Interesting. I do not agree that "speculation and guesswork is not what scientists do best." As a climate impacts scientist, I have seen the reticence of scientists to speak unequivocally on climate change diminish over the years as the evidence has built up. Until the evidence base is complete, everything is, in a sense, speculation and guesswork. I would prefer, however, a less emotive term such as simply 'uncertain' - quantifying and reducing uncertainty is a key aspect of science.

Calls for comments on breaking news would perhaps be better phrased by simply asking "what kind of processes or problems can lead to this event?" The answer given is then a simple matter of fact, and avoids speculation on the specifics of the news event. The question is, would such a comment be sufficiently newsworthy? And how might the medium in question choose to present it?

Relatedly, on the issue of climate science in the media, see
http://news.bbc.co.uk/1/hi/sci/tech/6115644.stm
for some astute observations.

Mike Begon said...

I understand, of course, scientists' reluctance to speculate

but we will never stop others from doing so - publicly

so shouldn't we be moving towards a culture where the media - and through them the public - understand a scientist who speculates in the following terms

"there are three (or whatever) possibilities here as far as we can see at present

option 1 is possible in principle but would require ...

option 2 is more likely but only if ...

option 3 appears at present to be most likely but is far from certain and could be ruled out, for example, by ...

there may, of course, be further options that suggest themselves as further facts become available"

difficult

difficult for the public to understand

but isn't that the challenge? (ultimately, not simply to say this kind of thing, but to educate the media and public to understand it)

surely the alternatives are worse - oversimplifying, or staying silent to maintain scientific integrity (while shouting down the phone)

Pamela Goldberg said...

Fiona Fox - Science Media Centre - "should scientists speculate on breaking news stories", I think this is an excellent article which I will circulate. In the political arena Margaret Thatcher was a past master at saying how pleased she was to be asked the question ..... and then giving the information she wanted to communicate anyway. I think that this may be a political skill rather than something which scientists are either trained for or comfortable with. A little practice on questions you can't or don't want to answer could do wonders!

The more that scientists can talk about what they do and be realistic about the limitations of knowledge the better science will be understood.

Unknown said...

Great article. A few times I've been asked by journalists if I am willing to speculate on some aspect of a story - and they are always delighted if I say that I will... for us to have real impact on the news and hence the public, scientists and engineers need to be part of the excitment of a breaking news story, not the boring "no comment" bit that gets left on the cutting room floor. But we are, of course, afraid we'll make ourselves look like idiots by speaking beyond our level of knowledge. I feared criticism from my colleagues over my involvement in the Polonium story - but in fact got nothing but positive feedback from colleagues who, in some cases, know more more about the subject than me, but were not themselves prepared to talk to journalists... so our fears are perhaps sometimes ungrounded....

Keep up the good work!

David Fisk said...

If it helps the key principle in accident investigation is to collect all the evidence first and then speculate. This isn't being prissy but comes from experience. If you collect evidence with a cause in mind you see things that aren't there, miss things that are and can destroy the possibility of changing your theory. Columbus bits were assembled in a large hanger before anyone seriously started to unpick the catastrophe

e.g. the police first thought that Woolmer had had a heart attack because of his team's result. Plausible but consequently they lost time they should have used fingerprinting the guests

You need some line like 'speculation risks distorting the investigation at this early stage. The investigation will be giving a high priority to anything that suggests this is not a one off that means the rails.......

Paddy Regan said...

Great blog from Fiona, tell her I fully agree....my view has been if we (as professional scientists) want to have the 'title' of 'University Lecturer, Professor in whatever' etc. then when something of national interest comes up in that area we have a responsibility to at least listen to the questions posed and try to provide, where we can, useful back up information for the 'non-expert' media people who need such info. Hiding away behind 'I'm not sure yet so I won't say anything' is I think a pretty poor response from professionals in the field, particularly those of use who are also paid to be 'educators'.
You guys at the SMC do a great job, keep hassling us 'in the business' to keep helping you where we can.

Kate Carpenter said...

An excellent article, I totally agree. I'd rather hear intelligent informed speculation than guesswork by those with no real knowledge at all.

The bottom line is that without informed speculation by experts, there is only uninformed speculation by laymen, who reinforce societal prejudice.
For example, the anti-speed-camera groups will always supply a plausible-sounding 'expert' with bad physics, and if a decent specialist is not put opposite him, the anti-view appears 'fact'.

Keep up the good work.

Professor Richard James said...

I thought that the article was very interesting with a good justification for scientists being involved with the media.

For a completely unexpected consequence of me as a scientist being in the media you might like to look at

www.the-scientist.com/news/display/52851/

Ian Forbes said...

Fiona's argument - which is very well put - also works for issues, events and perspectives, arguably to a greater extent, which involve social science expertise. All too frequently, I hear anecdote treated as evidence and used to counter or dispute evidence from research, and one person's often colourfully expressed opinion set against the careful contribution of someone who has built up a lifetime's knowledge not just of one area, but of the way to approach, understand and explain social phenomena in general. Standing back from the fray may be the wise strategy for some individuals, but the science community has I believe a professional responsibility to ensure that the science perspective is always heard. As a social scientist I do not subscribe to the view that we can provide the answer, but those who do claim to know the answer must always be subject to challenge.