Roulette, insurance policies and loading the dice: What could the IPCC learn about risk? A Q&A with James Painter
- 26 Sep 2013, 10:30
- Ros Donald
How can smart use of the language of risk help those
communicating the Intergovernmental Panel on Climate Change
(IPCC)'s new blockbuster assessment? Oxford University's James
Painter tells us Carbon Brief how recommendations in his new book
can help science communicators and policymakers explain the IPCC's
Painter's new book, '
Climate change in the media - reporting risk and uncertainty',
analyses media coverage of climate science. Among the book's
conclusions is that talking about risk may help shift the
conversation away from less certain areas of science and start
people talking about what society needs to do to avoid the worst
effects of climate change.
What does the reporting of risk and
uncertainty mean in the context of the book, and how does it relate
to climate science?
In the book, we looked at what we already know about
how the media report risk and uncertainty in general, and then
looked specifically at climate science and how the print media in
six countries reported the uncertainties and risks around it.
To analyse the content of media reports, we used
indicators of uncertainty such as ranges of projections, use of
words like 'may' 'could' and 'might', and dissenting voices
including the different types of sceptics.
For risk, we assessed how much the media and
scientists explicitly use the word 'risk', or risk concepts such as
assigning probabilities and confidence levels to possible outcomes,
or every day risk language such as 'loading the dice', 'taking out
an insurance policy', or 'playing Russian roulette with the
One of the findings of the book was that nearly half
of all the 350 articles we looked at included a quote from a
scientist or science report indicating some manifestation of
uncertainty. Much fewer used risk language.
How have the media reported risk and
uncertainty in climate science, especially the work of the IPCC?
In 2007 [when the most recent IPCC report was
released] the media were not very good at explaining how the IPCC
measured uncertainty. About 40 per cent of the articles included
the IPCC concepts of likelihood and confidence levels but only 15
per cent included an explanation of what they actually meant
- such as 'very likely' meaning more than a 90 per cent
Journalists gave us lots of good reasons to explain
this - such as readers or viewers not liking or understanding
numbers - or the difficulties of concepts such as 'medium
confidence'. But one of the essential challenges remains to
explain to a general public that uncertainty [in a scientific
context] does not mean ignorance, and uncertainty can be
quantified, particularly to help decision-making.
How has this played out in the run-up to the
I think some of the reporting has been excellent, in
that it has raised the issue of how the uncertainties can be best
communicated and whether talking of risk can help.
These are very important issues to think about. I
would point to this by Alister Doyle of Reuters and this by Seth Borenstein of
AP, who are directly addressing the issue of uncertainty. It is
also positive that a recent speech by
Lord Stern, in which he uses
risk language [- he's quoted saying it would be "absurd" to
claim the risks of climate change are small -] has been widely
My concern would be that some of the right-leaning
press seem keen to discredit the IPCC and have misreported the
uncertainties, or their implications. Professor Myles Allen's
response [below the line] to a recent article in the
Mail on Sunday is exactly what climate scientists should be
doing to point out when journalists are making mistakes, or using
the revision of climate sensitivity estimates to downplay the
urgency of taking measures to reduce emissions.
What tensions are there between scientists'
concern about reporting their results accurately and the more
accessible language of risk?
Different scientists have different approaches to
this, and in part it depends on what scientific discipline they are
coming from, and in part on their perception of whether, or in what
way, scientists should wrestle with the need to communicate the
climate science to a wider audience.
For example, it is very understandable that the
authors of report on the physical science - which comes from one of
working groups, Working Group One - may shy away from using
They argue that they prefer to use likelihoods as it
is a more neutral and quantifiable language; they say risk on the
other hand is a more qualitative language and implies only bad
outcomes whereas some outcomes from climate change may be
beneficial to a limited extent.
What's interesting is that the Working Group Two
report [on climate impacts] will probably have much more risk
language in it, as indeed the IPCC report on extreme weather events
did. This is in part explained by the different science
disciplines involved, but also by the greater acceptance that the
impact of extreme weather events are likely to be negative, and
therefore 'risky' in the sense of a negative outcome.
How can scientists use what you've found to
move the debate from dwelling on the uncertainties to what they do
This is a real challenge. For example, the draft
version of the Summary for Policy Makers for the Working Group One
report has the word 'uncertain' in it more than 40 times.
Journalists will no doubt report the uncertainties
extensively, and some media organisations may play them up to
follow an editorial line of justifying not taking action on curbing
greenhouse gas emissions.
Some of the climate scientists we interviewed in the
book stressed that at least in interviews with the media, they
found it helpful to start with what they were very sure about - for
example, that temperatures are going up, that it's largely caused
by human activity and that many of the impacts are likely to be
negative - before explaining the areas of uncertainty.
How can policymakers use the language of risk
to convey the need to reduce emissions to the
Several politicians and policy makers are already
using this language, such as Michael Bloomberg in the USA, and in
the UK, the Lib Dem Ed Davey and the Conservative Lord, John Selwyn
Intuitively you feel that risk language could work
with a wider audience, who are accustomed to risk from the worlds
of insurance, betting, health and investment. However, there's a
need for more research.
Which other conclusions from the book do you
think are valid in the context of the IPCC
Just one point - there's a lot of work being done in
general in the media on the use of infographics.
Several experts and journalists stressed that there
are already some really good examples of representing risk and
uncertainty- and much more could be done. Visual
presentations of uncertainty can be better than text.
James Painter is head of the
Reuters Institute for the Study of Journalism's fellowship
programme at Oxford University.