April 4, 2012, by Brigitte Nerlich
Languages of uncertainty
Communicating scientific uncertainty
There has recently been a lot of discussion about communicating uncertainty in science in general and climate change/climate science in particular. Many scientists, including Sir Robert May and Sir John Beddington have talked about how uncertainty is intrinsic to science and have advocated being more open about uncertainty, with the latter stressing that we have to make “evidence, and associated uncertainties, accessible and explicable”. But how do we do this? Again there is no shortage of advice, from a special issue of Climatic Change devoted to this topic in the context of the IPCC, to Adam Corner providing a very sensible summary of recent psychological and sociological research on this matter, to others who recommend that communicators who know their “audience, will minimize uncertainty and use a warm, more friendly style of communication“.
Uncertainty in ordinary language
What I found astonishing when looking at these discussions (which have intensified since ‘climategate’), was the lack of any reflection on how the word ‘uncertainty’ is used ‘in the real world’, in ordinary language. Making ‘uncertainty’ ‘accessible’ is not as easy as translating the scientific notion of uncertainty into user-friendly visual images or graphs or words like ‘likely’ or ‘highly likely’. We have to know about the contexts in which such ‘translations’ occur and how the concepts we want to make ‘accessible’ interact with the words and meanings that already are accessible out there.
A brief look at risk
When looking at the word ‘risk’, the situation is quite different. Here linguists like myself have reflected on the use of the word in various contexts. When studying risk talk using corpus linguistics, we found, for example, that when ordinary people talk about risk, this happens mostly in the context of health. Other studies have followed our lead and looked at environmental risk from a linguistic perspective. Some have studied the difference in ordinary language use of the words risk and danger. And some of the insights achieved have been incorporated into studies of climate risk and climate change communication.
Uncertainty, confusion and anxiety
But what about uncertainty? How is this word used in ordinary language? I have not undertaken any in-depth research of this issue yet, but I have had a quick look at two sources, WordNet® and The Bank of English. According to WordNet®, ‘uncertainty’ has two senses: ¨uncertainty, uncertainness, precariousness — (being unsettled or in doubt or dependent on chance; ‘the uncertainty of the outcome; ‘the precariousness of his income’)” and “doubt, uncertainty, incertitude, dubiety, doubtfulness, dubiousness — (the state of being unsure of something)”. Looking at various collocations and concordances within the Bank of English, it becomes clear that ‘confusion’, ‘doubt’ and ‘controversy’, as well as ‘worry’ and ‘anxiety’ are words closely related to ‘uncertainty’. ‘Economic uncertainty’, ‘political uncertainty’ ‘considerable uncertainty’, ‘total uncertainty’, ‘huge uncertainty’, ‘confusion and uncertainty’, ‘disruption and uncertainty’, ‘dark cloud of uncertainty’, etc. are phrases found in texts, from books to spoken conversations (see the extract from ‘British Magazines’). Overall then, uncertainty has highly negative connotations and it seems that it needs to be either ‘tolerated’ or ‘removed’.
Outside the texts I surveyed in the Bank of English, uncertainty is now also used in phrases like ‘sowing uncertainty’, ‘manufacturing uncertainty/doubt’, ‘producing uncertainty’. These are phrases used by people who examine a communicative scene where uncertainty is actively and politically used as a weapon in debates about climate change for example, something rather novel that needs a separate blog (and probably, in due course, a new entry in the Oxford English Dictionary).
Uncertainty and the progress/process of science
This is quite different to the use of the word uncertainty in science, where, on the whole, uncertainty doesn’t induce existential angst, anxiety and stress; is not seen as a threat; and is not used as a weapon. All measurements in science contain elements of uncertainty, which can be estimated. There might be doubt, but even that is, generally, regarded as a good thing. Reduction of uncertainty is the aim, but total removal would be regarded with suspicion rather than relief. Uncertainty is what makes science work and provides work for scientists. It is, on the whole, a good thing; it makes scientific life interesting. This is an aspect of uncertainty that scientists have focused on from Feynman to Firestein.
Uncertainty and climate change
However, in the context of climate change uncertainty should perhaps worry us more than we might like and in different ways. Until recently ‘uncertainty’ has been used as a ‘stick to beat’ science and scientists. The argument went that because science was too uncertain no action on climate change needed to be taken. This argument seems no longer to stand up to scientific scrutiny and it appears that “uncertainty should make us worry more than certainty, because uncertainty means that things can be worse than our best guess.” This means that, in the context of climate change, scientific uncertainty (related to sensitivity, probability, prediction etc. – all things ordinary language does not touch) also gains rather negative connotations. The question then becomes: How do you communicate uncertainty in the context of climate change without opening the floodgates to anxiety, stress and worry, that is, without conjuring up a ‘dark cloud of uncertainty’ that leads to paralysis and inaction? I am not sure, in fact I am quite uncertain!
An interdisciplinary group of academics led by Professor Sarah Metcalfe will explore the challenges of ‘Representing and communicating uncertainty: climate change and risk’ in two workshops funded by an AHRC, Science in Culture Exploratory Award (work linked to the University’s Natural Hazards and Disaster Mitigation Working Group). For further information on the AHRC award please email: Lowri.Jones@nottingham.ac.uk
PS, added 4 January 2013: Analysis of various types of uncertainty in the context of climate change (mitigation).
PSS, addes 19, February: on uncertainty and risk communication after the l’Aquila earthquake, good article by Helene Joffe
Brigitte Nerlich, Institute for Science and Society, School of Sociology and Social Policy, PI ESRC climate change project
Interesting discussion of uncertainty, which puts me in mind of Shibutani’s old study of rumour. The latter draws on the ordinary language meaning of uncertainty and argues that ‘authoritative’ sources of information – science and news media accepting its authority – as the means by which rumours are brought to a close. However, once science begins to promote uncertainty as integral to its practices, it becomes difficult to see what could be the basis of trust. Controversy comes to be the medium of scientific debate and rumour alike, with the dissolution of trust in public knowledge one of the consequences.
Thanks for your comment. I am not sure though I understand what you mean by science promoting uncertainty as integral to its practice. As far as I understand things (and when it comes to uncertainty in scientific discourse I am still learning), uncertainty IS part of scientific practice. My point was that this practical side of dealing with uncertainty is quite different to the way we ‘deal with uncertainty’ in ordinary life and ordinary language. The gap is actually very deep, which makes communicating across that gap so very challenging.
Maybe we should distinguish between (what I have just named) synchronic uncertainty and diachronic uncertainty. Suppose I arrive home feeling uncertain what is in the fridge for tea. No problem – as soon as my curiosity becomes unbearable, I can have a look and the uncertainty goes away. It is controllable and easily resolved. This is synchronic uncertainty. But if I think that my pension pot may have been devalued by a stockmarket downturn when I come to retire, there is nothing I can do to find out. I have to wait and worry. The resolution is (necessarily) in the future – ‘diachronic uncertainty’. That (in spades!) is the problem with climate change.
Just checking the weather forecast for travelling and came across this nice example of a (diachronic) uncertainty representation – notice how the uncertainty range changes the further you look into the future: http://twitpic.com/95ticn
@ Brigitte. Yes, I was making a slightly different point (poorly expressed). Controversy/uncertainty is one ‘moment’ at the frontier of scientific change, but change also produces what Randall Collins and Stephen Turner call temporary ‘settlements’, which alsmanage consequences for policy. By characterising science in terms of controversy this provides ammunition to those who wish to undermine the achieved settlements as the basis of policy. So, in relation to global warming, initial controversy gave rise to a temporary settlement and it is that area of consensus that has been subjected to deliberate attack by special interests. None of this is helped by a general neo-liberal approach to science policy and universities.
I see what you mean. The tension between (and political use of) controversy (in the guise of uncertainty) and consensus in the formation of climate change policy is certainly at the nub of many problems, as outlined by you.
I was having a similar interesting discussion about the use of uncertainty in science the other day. There was a view that the term uncertainty is often used as a euphemism for ignorance, which is actually closer to the reality, but less palatable! I thought it would be interesting to look at the two words together in this context?
Yes, absolutely. I think we have to think much more about the issue of the euphemistic use of uncertainty and the euphemistic and other uses of its quasi-synonyms, such as doubt, ignorance, imprecision etc. And again one has to be very careful in not conflating the use of uncertainty in one sphere of discourse with that in another or to build arguments on such conflations. Here are a few reflections on conflation of uncertainty with ignorance: http://www.carbonbrief.org/blog/2012/04/climate-scientists-take-on-lindzen
Yes, Brian Wynne has long made the point that uncertainty in climate models is not the automatic friend of sceptics. Based on his sociological work with colleagues in the 1990s, he has argued that the standard climate-sensitivity range of 1.5-4.5 degC (this was extended upwards in the 2000s), was a social judgment shaped partly by assumptions of what policymakers could handle. A ‘settlement’, as John suggests (though only in Europe), but one that may have under-stated the risks of abrupt change even as it becomes subject to the ‘dogmatic amplification of doubt’ to use Wynne’s words. All of which means the challenge is not only about how to communicate uncertainty in the scientific sense and avoid conflations with everyday meanings, but to get across the point that we shouldn’t have to rely on science alone to justify policy or public action.
Many thanks for this comment! I vaguely remembered the Brian Wynne thing when writing the blog, but not quite enough! Of course it’s not just a matter of mediating between various languages of uncertainty, although, as a linguist, I think that’s important; the challenge is, as always, to find political solutions based in part, but of course only in part, on science. Decision making is tough, individually and collectively. And things will probably get tougher with increasing (and interrelated) social, scientific and climatic uncertainties.
[…] we can learn from museums Making Science Public: Hype, honesty and trust Making Science Public: Languages of uncertainty Making Science Public: Waiting for […]
[…] two projects related to climate change; so I blogged about the weather and climate, scepticism, uncertainty and so on. Then, in May this year, our Making Science Public project started in earnest and I began […]
[…] to confusion, as the meanings of the word in scientific jargon and in ordinary language are quite different. I tried to dig a bit deeper this time and used an archive of ordinary language use which collects […]
[…] Uncertainty […]