Last month a study was published that did a statistical analysis of 40 years of observations by volunteers of selected bird, butterfly and plant species native to Britain. The study was marketed as bolstering the Sixth Extinction Hypothesis, the notion that 'the biological world is approaching the sixth major extinction event in its history'. The last such extinction event occurred 65 million years ago when most dinosaurs became extinct.

Some are disappointed that the study didn't get much commentary. The usual suspects tootled the horn of doom briefly but didn't draw much attention from either supporters or detractors. In The Dog That Didn't Bark In The Night The Loom asks:

I've ... been searching for criticisms, but to my surprise I can't find a single mention of the study in outlets that have attacked these sorts of studies in the past. Could it be that these folks are hoping that this study just disappears if they don't call attention to it? Or are they at a loss for a rhetorical trick to misrepresent the findings? Or do they accept that this may be a sign of a sixth pulse of mass extinctions, but is simply not worthy of commentary?

Tony Gilland obliges in Spiked with Biodiversity: don't get in a flap.

When I spoke to the paper's lead author, Jeremy Thomas, from the Natural Environment Research Council's Centre for Ecology and Hydrology, he freely admitted that 'to go from UK butterflies to world insects is a massive extrapolation', and stated that while the authors of the paper 'are confident about the UK data, the rest is speculation'. It is worth asking why, then, so much attention has been paid to the speculation...

The key finding of the paper, and something that surprised the authors, is that, butterflies, in general, have declined more significantly than birds and plants in terms of their range - the number of 10km2 grids each native species has been recorded in. The authors found that 28 per cent of native plant species (from a total of 1253 species) have decreased in Britain over the past 40 years, 54 per cent of native bird species (from a total of 201 species) decreased over 20 years, and 71 percent of butterfly species (from a total of 58 species) declined over a 20-year period. Two butterfly species have been declared extinct.

Of course, behind the general statistics there is a lot more detail. For example, if you were to look at the data behind the general statistic for birds, you would find that for 84 of the 201 species there has been little substantial change in their range (plus or minus 10 per cent), and that 41 species have increased in range by more than 25 per cent, compared to 25 species that have declined in range by more than 25 per cent. So what is the significance of this high-level statistical analysis?

As Gilland notes the significance isn't clear, even to the authors of the paper, who justify their speculations in political terms:

'if you are advising policy makers you usually can't wait until you have the decimal points in place but have to go on the best available information even if it is tiny. Otherwise there is a danger that it will be too late to do anything'.

Gilland counters:

If nobody but botanists and biologists cared about the fate of Britain's butterflies and plants, there might be more merit to such a headline-grabbing approach. But given that there is an official obsession with biodiversity in the UK - and, indeed, a reluctance to believe anything but scare stories - this latest extinction hypothesis seems rather unnecessary and unhelpful.

In several previous posts, especially Waffling with Dignity and Winter Sports, the conflict between science and policy, and the role scientists play in the conflict, was examined. Those posts attempted to illuminate the damage to society and the perception of science by society caused by what Gilland calls the "headline-grabbing approach" which uses speculation to influence policy. The earlier post dnE ehT ... toN recounted the history of such behavior over the past few decades and notes:

Caring for the environment is an age old, deeply held value of the vast majority of humans but support for the environmental movement falls far, far short of this majority. By politicizing these issues the environmental movement has soured the public and harmed the environment. They have become our greatest environmental problem, the largest impediment to developing effective environmental policy.

Carl Zimmer at The Loom seeks to discredit skeptics and Tony Gilland at Spiked seeks to discredit sexed up speculations published for political effect. History seems to support Gilland's stance more than Zimmer's since humans are so very bad at speculation and science speculation has had such a shoddy record in recent decades, especially about environmental issues.

But what if the sky is falling? What if the doom mongers of the past were just premature in their fears or misdirected in speculation and we would be wise to alter our behavior to avert that doom, or some other as yet unfeared doom? After having falsely cried wolf so many times how can Chicken Little be taken seriously now?

I have argued repeatedly that scientists, science writers and the media in general as well as politicians that use doom scenarios as wedge issues are the cause of doubt, that they have ruined their own credibility, and that they must rehabilitate themselves by changing their behavior so that they can serve the useful social purpose of informing society. Shouting louder or seeking clever ways to disguise speculation doesn't overcome public distrust of science, it makes it worse. Gilland is wrong that there is "a reluctance to believe anything but scare stories", few believe them anymore than they believe other political assertions. It's just ammunition to attack political opponents and people continue to live in ways that demonstrate their lack of belief. Looking at what they do rather than what they say in the heat of battle reveals their beliefs.

We need some maturity in the science community, recognition that by presenting speculation as data they have diminished the ability of society to respond, slowed reaction time to legitimate threats.

For related posts see Unanimous Fallacies and Affective Forecasting Accuracy

UPDATE: 4/22/2004

In a letter published in Nature a group of researchers in the Biodiversity Research Group, School of Geography & the Environment, Oxford University echo Gilland's plaint and offer suggestions for improving the situation.