Category: policy

The Sceptical Compass

First, thank you. I have been overwhelmed by the response to this blog, and privileged to host the conversation of ninety five individuals on my first post. Here is a Wordle of the comments (not including my own):

Second, some thoughts on terminology. Over the last year I have started to talk with people who do not agree with the majority view on climate science. And there is no homogenous “sceptic” viewpoint. No binary grouping, Us and Them. I do use the terms “scientist” and “sceptic” for convenient shorthand (more on this later), but whenever I talk about public engagement I bring up the same points:

a) there is a continuous spectrum of viewpoints;

b) a large number of the unconvinced have numerate backgrounds (off the top of my head, physics, chemistry, computing, engineering, geology and finance seem to come up most frequently);

c) for various reasons, they have lost trust in the way we do, or the way we communicate, our science.

This week I’ve been thinking that the ‘spectrum’ description can be pushed further. If you’re familiar with the Political Compass, you’ll know that it extends the usual left-right political spectrum to a two dimensional graph of left-right and libertarian-authoritarian (if you don’t know it, I recommend you do the quiz). Here’s my proposed equivalent.

The horizontal axis is sceptism: the degree to which one critically evaluates evidence, does not accept arguments of authority, and updates ones viewpoint according to new information. This is the ‘Approach’ axis.

The vertical is the resulting ‘Conclusion’ axis: the degree to which one is convinced that humans are causing climate change and (if there is some degree of human cause) the scale and speed of that change. The sceptic/scientist shorthand I use corresponds to this axis. I have also started to use the less well-known upholder/dissenter and convinced/unconvinced.

The compass doesn’t include policy preferences, of course.

I’ve marked some examples. I don’t think it is a simple categorisation: like the Political Compass, people can move around through their lifetime, can be in different locations for different topics, and may be ‘smeared out’ vertically in the case of large uncertainty. I am not trying to label anyone here, and these are not rigidly defined regions. This is purely illustrative.

Convinced: horizontally, scientists and many non-scientists aspire to be sceptical; vertically, people in this region are convinced by the majority of these statements (for example, the majority of climate scientists).

Lukewarmer: horizontally, as previous; vertically, somewhat convinced (for example: concluding that humans cause some change but the rate is likely slow or very uncertain).

Unconvinced: horizontally, as previous; vertically, not convinced (for example, concluding there is warming but the human influence is small or negligible).

Believer: horizontally, uncritical and trusting of sources they consider authoritative; vertically, convinced of rapid, intense climate change and impacts caused by humans.

Unbeliever: horizontally, as previous; vertically, not convinced (for example, concluding there is no warming).

For the Bayesian nerds, I’ve just noticed the horizontal axis could be considered the width of one’s prior, and the vertical axis the mode of the resulting posterior.

I’ve chosen to put the dots at the vertical extremes for the uncritical side (Believer/Unbeliever) to reflect the fact that people who are not critically evaluating each statement, only trusting in another source or opinion, may be more likely to agree with the extreme ends and see the issues in black & white. I’ve chosen the Sceptical dots to be more moderate in the vertical (Convinced/Lukewarmer/Unconvinced) to reflect the fact that critical evaluations may lead to a more nuanced view with shades of grey. But I think of this as a continuous space.

There are no value judgements intended here. There are several reasons why there is not a one-to-one relationship between critical evaluation and conclusion: access to evidence; availability of time or technical expertise to evaluate it (reliance on judgement of others); general fallibility of humans. Scientists have differing opinions and interpretations of the same evidence, and we are not perfectly critical, so we can be at different levels on the vertical axis. For example:

– a scientist who models the physics of ice sheets might judge the statistically-based (‘semi-empirical’) methods that predict a rapid sea level rise as “not credible”: they would therefore be lower down the vertical scale;

– a scientist might search for an estimate of the current health impacts of climate change and, for lack of time or another reason, use a non-peer-reviewed estimate that reported severe impacts: they would therefore be higher up the vertical scale and further left horizontally.

I’d be interested to hear if people think this is a useful framework. If you don’t like it, please (kindly) suggest changes.


Third, the scope of this blog. I said to Peter Gleick that my aims were: to communicate my own research, because I am publicly funded, and because it gives the research greater exposure; to engage sceptics (see above!), and to practice writing for a general audience. This post is already too long, and the time too late, for me to list every topic I intend to cover but it will become apparent as I write posts. Some things I cannot do on this blog:

a) answer every question asked: this will depend on my knowledge and the extent to which I have time to answer (both can be improved by postponing to a later post);

b) address everyone’s problems with climate science: I am only one person, an early career researcher with a lot of things to wrap up by 31st July, and although I try to read outside my area I cannot promise to have the expertise or time to address every issue;

c) comment on policy choices.

I suppose this is just a restating of not pleasing all of the people.


Fourth, a comments policy.

So far I have let through every non-spam comment and automatically allowed previous posters to comment. I would like to trust people to be sensible with this and not have to start moderating out comments.

Therefore I ask you to comply with the following:

a) civility is essential;

b) accusations are not to be made;

c) the words denier, liar and fraud are not permitted (this list may increase): see (a) and (b);

d) generalisations are to be avoided;

e) if you have a particular bugbear or issue with earth system model uncertainty that is not related to the post topic please invite us once, perhaps twice, to discuss it in the very suitable Unthreaded section of Bishop Hill;

f) if you have a particular bugbear or issue with some other topic, or with policy, please discuss it elsewhere;

g) interpret comments in good faith: each is from a person, with limited free time, and frazzled nerves, and good intentions;

h) liberally sprinkle your comments with good-humour, honesty, and ‘smiley’ or ‘winky’ faces, to keep the tone convivial.


Thank you.