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.


  1. Ben Pile

    The problem with a political compass — and even just an axis, or ‘continuous spectrum’ — is that it is a model for understanding something far more complex than coordinates can represent. A good example of the failure of representing opinion in this way is the political spectrum. The left-right axis was perhaps more useful when the world was more neatly divided. But as early as 1941, Orwell noted the redundancy of the categories, in The English Revolution.

    One couldn’t work backwards from a position on the compass, to say what somebody’s beliefs, values, understanding, or objections are. And neither is there a sensible way of arranging claims (and for that matter, counter claims) along an axis, towards some general principle or another, according to their magnitude. Quite simply, there are far more than two axes to the debate.

    There are a constellation of claims made in the climate debate. And it is a mistake to believe that the debate itself does not exist within a much wider context that is reflected within it. A more damaging problem is that people end up arguing with the coordinates of the debate: the categories themselves, rather than with the substance of it. For instance, the climate debate frequently descends into claims about the regrouping of crypto-communists (the ‘believer’ on your compass), or the ‘ideologically-motivated’ libertarian at the other (perhaps the ‘unconvinced’). As well as arguing with categories, people end up fighting ghosts.

    If there is an axis in a debate, it must reflect the structure of some proposal or another. That way, an axis moves from the origin as its first premise, to its conclusion at the maxima, and the extent to which an individual agrees with the argument determines his or her place on the axis. This means unpacking the debate in question, and as we know the climate debate is far more complex than just one question (thought it is frequently and unhelpfully reduced to one between ‘science’ and ‘dXXXXl’). Let us imagine such an argument, within the climate debate:

    i) CO2 causes global warming; ii) global warming causes climate change; iii) increased precipitation is caused by change climate systems; iv) increased precipitation increases opportunities for mosquito breeding; v) the rate of malaria infection increases.

    Now, one can fully concur with this chain of reasoning, but completely reject it. (Let’s forget for a moment that rates of malaria infection have fallen). What is not explained in this argument are its presuppositions. There won’t be increased malaria in the Southeast of England, not least because it has already been eradicated, but because we could easily mobilise resources against it. Where there is less wealth, the model holds more true. Yet, even where it is a more accurate description of an ‘impact’, it is only consequential if we cannot say that some other form of intervention (or, for that matter a withdrawal of an intervention already in place) would be more effective at dealing with, or preventing the ‘impact’. It is possible to do very good science, but from very poor assumptions. I believe that impacts are possible, but that they are principally determined by social conditions, not by the magnitude of climate phenomena.

    In most cases, the climate debate (where there is one) is about presuppositions. They are not reflected in the compass. Apologies for the long comment.

    • Anteros

      Very well said and argued.

      I was particularly struck by this –

      I believe that impacts are possible, but that they are principally determined by social conditions, not by the magnitude of climate phenomena.

      So very true. Some groups of people are vastly more vulnerable to climate than others. And even if a change in climate is measurable or noticeable, the vulnerability is still to climate more than its change, and is predicated on the prior degree of vulnerability.

    • Arthur Delaruelle

      What about a sceptic convinced that global warming is good for him ? Does not fit in the quadrant above.
      I live in Europe at 50 degrees North. My house is near the highway, I see each year millions of cars driving South, in search of the sun … Are they climate refugees ?
      One point common of the believers is that they NEVER mention any benefit of AGW. That is why I do not trust them.

      • Vince Whirlwind

        It isn’t good for you. You exist in the context of the economy at local, national and global levels, which in turn are all inter-linked.
        Climate change creates large costs to national and global economies (even Lomberg admits this, athough a more trusted source of data would be something like which in turn affects the local economy.
        You won’t be better off.

  2. Theo Goodwin

    Excellent work. Very promising. Should I use coordinates to identify myself on your graph?

    • Theo Goodwin

      My position on your graph is (100,0). I take this to be the position held by scientists at the beginning of an inquiry into a topic that claims to be scientific. I expect the science to go beyond “a priori” ruminations on first principles to embrace some well confirmed physical hypotheses which can be used to explain and predict some physical phenomenon that is posited by proponents of AGW. I expect a strong empirical component to the science based on passive or active experimentation. I expect scientists to do their utmost to make their results reproducible. I do not give credence to my own ideas until some other scientists have reproduced my work. Given these standards, I expect to find once again that climate science, so-called, is in its infancy if not in the birth canal.

      Our topic is computer modeling. Models are wonderful analytic tools for discovering things about physical theories that might have been overlooked. Models cannot be used to replace physical theory. Claims that models produce scientific evidence reveal a fundamental misunderstanding of models and physical theory.

      As a (100,0) sceptic, I do not intend to push a thesis about AGW. I do intend to keep us honest. The standards of honesty are provided by scientific method and the art of computer modeling. Some have told me that my standards are too high. Yet the works produced by Svensmark or Kirkby receive a rating of 100% on my standards. Svensmark and Kirkby are engaging in genuine physical science and their work will live in the annals of science; however, their work might fall short of providing an adequate understanding of AGW.

  3. Aussie

    Tamsin you make a very interesting point about the people who tend to fit into the sceptical side of things…. that they are numerate and include people with a finance background…. well that is true…. in my case I am a Bachelor of Economics and Commerce.

    There is one difference, and that is, some of us also go by observation and experience over a period of years. In fact many people who do not fit your categories and are sceptical actually use their life’s experience to counter what is being told to us. On top of this, most of us also use the experience of the country where we are living to determine our attitude. Again, in my case, living in Australia, I experience weather extremes in summer and winter due in particular to location. This experience has been enhanced by living overseas for a short time and experiencing even more extreme conditions (very cold conditions).

    This explains why, without having the mathematical background many of us will simply say that what we are being told is hype and fearmongering. In other cases it might be a medical study that has concluded with remarks that are simply not based upon experience… e.g. the alleged rise in Lyme disease in the USA that is supposed to follow warming and in Australia the alleged possibility of an increase in mosquito borne diseases…. in one particular claim the disease is not carried by the mosquito unless it has bitten an infected human that has brought the disease back to Australia after travel to tropical places such as Bali!! This is why there are so many who are on the sceptical side of the fence.

  4. Jonathan Jones

    It’s a useful and slightly unusual categorisation. I would normally describe myself as “climate agnostic with a tendency towards lukewarmer”; on your diagram that puts me fairly firmly on your “lukewarmer” spot.

    Like Aussie above, my entry into scepticism was primarily triggered by a sense that the field was badly over-hyped. Having worked in a wide range of scientific disciplines I am very aware of the role of hype in driving much of modern science; it’s a consequence of our funding and promotion systems, in which a willingness to make simple, clear, strong statements is favoured over nuanced careful assessment. In that sense the failings of much modern climate science are not particularly unusual, though they do seem to me to be a fairly extreme example.

    • Richard Betts

      Jonathan Jones:

      my entry into scepticism was primarily triggered by a sense that the field was badly over-hyped.

      Hi Jonathan,

      Is it therefore a logical extension of that to say that continued hyping of the field will only maintain your scepticism and encourage others to “enter into scepticism”? ie: continual banging-on about the urgency of the situation, instead of discussing the science at face value (and indeed trying to stop the science being discussed at face value!) only serves to create scepticism not counter it?

      So those who bang on and/or worry about things being “misused by the sceptics” only score own goals?

      • Richard Betts

        Sorry, formatting italics went wrong there!

        Tamsin, if you get time (ha!) please could we have a way of checking posts before they go live, like on Bishop Hill?

        Sorry to create more work…. 🙂

      • EdBhoy

        It was precisely the obfuscation of inconvenient data and claims that the debate was over that first set climate science alarm bells ringing in my ears. If we can’t expose the science (theory, data and methods) to considered criticism we are not employing the full power of the scientific method and ultimately doing a great disservice to the public. Good science stands up to scrutiny and is only strengthened by full disclosure and discussion – why would any well meaning scientist wish to stifle debate?

      • RealArthurDent

        I agree with Professor Jones, his journey to the luke warm side and mine were similar and I agree with your premise Richard encapsulated as So those who bang on and/or worry about things being “misused by the sceptics” only score own goals?

      • HaroldW

        Dr Betts —
        Yes, I think that conclusion is correct. The more that scientists act as advocates, the more they will be binned with politicians and lobbyists, and viewed with the ever-increasing distrust which those occupations enjoy. Those who are in the middle on the “approach” axis will gravitate to opposing viewpoints.

      • Jonathan Jones

        Yes. I would single out the RealClimate crowd, and people like Bob Ward as particularly likely causes of continued scepticism. As I said before Lying “to avoid being misunderstood” rarely ends well.

        And I’m afraid I would consider the recent Met Office post on the pause in global warming as a good example of how not to do things. It is perfectly right and proper to say “This pause is probably not significant: similar pauses have happened in the past, and we have good physical reasons to believe that warming will soon resume”. It is not OK to imply that the pause didn’t actually happen.

        • Mike Jackson

          And you can add to that the WSJ letter and the responses to it, in particular Laden’s somewhat partisan comments.
          While both the graphs he shows (‘How “sceptics” view global warming’ and ‘How “realists” view global warming’) can be said to be accurate neither is in fact correct (or do I mean the other way round?). Both sides of the argument are guilty of cherry-picking in their own way though I must confess that while I have seen several “realist” graphs that draw a trend line like the one he shows, I have never seen a “sceptic” graph which is so patently idiotic as his example.
          So while we know that the warming has been flat or near-flat over the last decade no sceptic is going to be convinced by the blatant advocacy of those who try to claim otherwise. I have used the word “stall” because I don’t know whether warming will resume or not ; to say it hasn’t even stalled is simply to deny reality.
          This obsessive need to adhere to the GW narrative regardless of any variation in what the climate is actually doing is hardly calculated to “win friends and influence people”. The reaction is increasingly likely to be the Mandy Rice-Davies response: “They would say that, wouldn’t they?”

          • BBD


            The problem here is that the unbelievers have been vociferous in misrepresenting the last decade as ‘evidence’ that AGW is a ‘myth’. This tends to make subscribers to the mainstream view very keen to demonstrate otherwise. Typically by focussing on evidence that warming hasn’t stopped (eg GISTEMP; BEST) and studies like Forster & Rahmstorf (2011).

          • BBD

            Jonathan Jones

            First, that is a serious misrepresentation of what I said (popular with unbelievers, is misrepresentation – see above). Second, you have breached the commenting rules:

            c) the words denier, liar and fraud are not permitted

            I hope you aren’t going to carry on being tiresome.

        • John Russell

          Define ‘pause’.

          We can agree that the rate of warming is momentarily not as steep as it was during, say, the 1990s, however — as many scientists keep telling us — we shouldn’t assume it’s a pause until enough time has passed to be able to distinguish the signal from the noise: and there’s a lot of noise. If we climb a mountain blindfolded do we assume that we’ve reached the top when it appears to start levelling out — or do we wait to declare that was the top at the point that a definite downward trend has appeared?

          The problem seems to me to be that the first group want to believe warming has plateaued, while the second group can’t believe it will plateau while ever CO2 keeps on climbing at an accelerating rate. While I can accept that an objective scientist will be content to wait and see what transpires, might not the human inside him or her worry about what will become of his children if the worst that models predict actually transpires?

          So I’m definitely in the second camp. That might mean that I would make a poor scientist: however it does make me risk-averse, which I would contend is a worthy human trait.

          • Jeff Norman

            The way I think about the current “global temperature” information (not what I would call data) may be a little different.

            Various models were used project temperature scenarios out to the year 2100. These scenarios were then used to suggest consequences. The scenarios and consequences were presented by the IPCC and taken and treated as fact by people at the believer end of the spectrum.

            I now look at where we are now to see if we can get to where the IPCC suggested and the believers believed we were going from here. More and more it looks like we cannot.

            This suggests to me that the models were/are in fact wrong. I look forward however to learn how the are useful. This usefulness should be tempered against people doing the wrong things for the wrong reasons because they believed what they were told.

          • Faustino aka Genghis Cunn

            “We shouldn’t assume a pause” because “there’s a lot of noise.” But you seem to assume a rise in spite of the noisy data; by your own argument, temperatures might be falling, and the interpretation really is that “We don’t know what is happening to global temperature since the turn of the century, as the data isn’t good enough to determine any trend.”

            As for risk-aversion, I could counter-argue that we shouldn’t impose certain significant economic costs because of uncertain possible warming and uncertain possible net negative consequences of it. There are always risks and unknowns, our race has progressed enormously by dealing with them.

        • geronimo

          I did the test and came out slightly to the right of Ghandi, when I feel slightly to the right of Mrs. Thatcher, perhaps because the questions, I don’t know.

          Anyway Jonathan, I agree with you about the Met Office pronouncements, and it has made me come to the conclusion, rightly or wrongly, that anything coming out of the Met Office is for the benefit of Greenpeace, Friends of the Earth, WWF etc. So to set Tamsin and Richard straight it is the way the Met Office communicates the science that makes me a sceptic, their latest foray into politics was to point out to the government that the last decade was the warmest on record, which is true, but they “hid” the fact that there hasn’t been any significant warming in that decade, and indeed since 1998. To me, in a world where poor people are paying more for their energy to support, what every engineer on the planet not building them, renewables that are totally useless to provide our future energy supplies, and starving because large swathes of land have been turned over to growing biofuels, the fact that there has been no significant warming should be flagged up, especially as in the same period CO2, the centre of gravity of the CAGW claims, rose by 6-8% in the same period.

          I promise you that if this happened in an engineering discipline, pencils, drawing boards and slide rules would have by now been revisited for an explanation, instead of which the Met Office chooses to hide this significant fact by telling politicians that the decade was the hottest on record. For everyone on the planet it was their oldest on record.

          • Tamsin Edwards

            “anything coming out of the Met Office is for the benefit of Greenpeace, Friends of the Earth, WWF”
            ‘they “hid” the fact’

            *cough* – policies (c) and (d), geronimo. Putting it in quotes doesn’t help you 😉

      • Alexander Harvey

        Hi Richard,

        I believe that certain individuals, groupings, and one blog in particular (RC) have acted to effect a radicalisation of some or many away from a sceptical neutral position towards a one hostile to theirs.

        Perhaps, more alarmingly, the effect has been to radicalise some with an hitherto sceptical acceptance of the overall concensus postion into the creation of an “awkward squad”. People paradoxically pulling on the other end of the rope; for methinks they go to far. This I know to be true for it closes on my own position and I know of others who push back for a variety of similar considerations.

        It is important that those who feel strongly hazard their reputations but some I percieve to have bet not just their credibility but that of the science itself by taking brittle stances and evoking some mythic scientific certainty were only a likelihood was warranted. Such is my prejudice.


      • Aussie

        Jonathon I would also add the desire to please… that desire can sometimes have dire consequences, especially when the scientist concerned has ended up disgracing himself/herself by falsifying data…. and no I am not referring to climate scientists, but to a well-known medical researcher whose big discovery had consequences, and then his desire to help someone caused him to falsify his findings in another case!!

  5. John Costigane

    Science should be about increasing knowledge, which is available to everyone. Instead, we have a political/advocacy movement, of the self-chosen, which seeks to ‘change the world’. The best antidote to this is good science, and climatology has the potential to be better than a good science.

    Modelling should include all scenarios including dangerous warming, dangerous cooling (Ice Ages) and very dangerous cooling (Snowball Earth). All 3 are possible in the future.

  6. Jeff Q

    Another promising post Tamsin!
    I like the axis approach, using that I reckon I’m just below the line of lukewarmer. I also very much like the fact that you (and others) have acknowledged that it is possible to be both rational and hold a view that AGW may not be as it is painted by the MSM.

    I’m looking forward to learning a lot here, who knows I may change my position! (although that would mean I’d have to send that massive cheque from big oil back!)

  7. Sean Inglis

    I agree that there are more dimensions to the debate than can be shown on a diagram like the one proposed, but I think sticking to two dimensions only is a good start.

    If you then wanted to introduce some other dimension, I’d prefer to see it as a separate compass graph with different axes.

    The point that we can occupy a range in either dimension is well made and I guess would be represented by the size (and intensity) of a particular blob identifying a particular respondent.

    One thing that strikes me about the example positions and names is that there is no occupant at four of the points where there is a zero value, ie at the intersection of the two axes etc.

    How could you refer to people falling at those points on the diagram? Can some of those positions even be occupied without inevitably referring to someone who declares that’s the case in a derogatory way?

  8. Nick M

    This is a great idea and a nice simple model based on the empirical evidence of expressed opinions. However “all models are…” well, indeed, and this one’s main problem is that it’s missing one component of the underlying theoretical framework. Paraphrasing your first commenter: In [many] cases, the climate debate […] is about presuppositions. I.e. many opinions are not arrived at by assessing the evidence and opinion then processing this into a personal opinion, but often the reverse; suiting lifestyle, politics, employment (also commercial interests [does this count as an accusation?]).

    Anyway, nice blog, nice ideas, keep up the public engagement!

    • Nick M

      I think the lacking component is a representation of the individuals whose axis is “for” or “against” and the level of evidence/opinion accepted depends whether it agrees of disagrees with this position. That is, a third axis, with your flexible framework working as a slice at zero. There is confusion if you just project onto the above plane.

  9. Barry Woods

    With respect to Richards and Jonathans comments:

    It might be enlightening to have 2 such graphics, one with respect to ‘climate science’, and the other with respect to ‘climate chnage science’, or the media/polictial/lobbyist/science advocate(version) of climate science..

    And see where people would identify themsleves on both, as I would be in 2 very different places on those graphics. And this is the problem, most sceptics I think are reacting to the latter ( ie, those that think ‘misused by sceptics’ !)

    I hope on the former graphic, that it would be fair to say, Jonathan, Richard and myself my sit in a broadly similar spot? (me tending towards luke warmer, overhyped a bit). YET on the MEDIA/Environmentalist/policy advocate version of that graphic, where would we all sit then?

    • Anteros

      Barry –

      I agree – the more you focus on a particular piece of science or evidence, the less disagreement there is. The further you head towards ‘implications’ and ‘meaning’, the more disagreement there is – from the same pieces of evidence.
      And I think you’re right to identify the place where the two career apart – the place where somebody mentions ‘change’!, because then you have all the emotional baggage about guilt and the worry over human agency and responsibility. At that point imagination can take over…

      I think a clear demarcation between what science can speak about and what it can’t is very helpful. Where value judgements about ‘danger’ and ‘risk’ and ‘change’ are concerned, climate science has nothing to say. So, many sceptic viewpoints are not anti-science, they are a-science. Which is perfectly appropriate.

  10. Rick Bradford

    An alternative approach to the 2-axis Cartesian visualisation is to give attributes their own space in a RGB color plot.

    For example:

    The attribute “must take action now” could range from Red 0 (no action required) to Red 255 (immediate drastic action required)

    The attribute “man is causing climate change” could range from Green 0 (not true) to Green 255 (overwhelmingly true)

    The attribute “I am informed about the science rather than belief-driven could range from Blue 0 (belief-driven) to Blue 255 (very well-informed).

    So someone who neither knows nor cares about climate would be RGB (0,0,0) => black. Someone who passionately believes that AGW is real and that action must be taken, but is not scientifically informed, would be RGB (255,255,0) => yellow; a well-informed skeptic about AGW concerns and actions would be RGB (0,0,255) => blue.

    Ideally, the attributes should be completely independent — in this case, they aren’t quite, as there probably aren’t that many people who believe passionately in the notion of AGW but disdain action; which would be RGB (0,255,0) => green.

  11. Paul Matthews

    Ouch. I realise it’s only shorthand but the “scientist/sceptic” shorthand that you mention twice is an awful categorisation.
    I am a scientist and a sceptic.
    Indeed every scientist should be a sceptic (as you suggest later).
    To clarify what sceptic means (the term is often misrepresented in the climate debate) it means that if someone claims something like “Most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations”, I want to know the full details of the assumptions, arguments and calculations that support that statement.

    The convinced/unconvinced categorisation also seems to me be to skewed. Many of us in the lower right region of your diagram are quite convinced – for example that the IPCC and some climate scientists have distorted the science, or whatever.

    I think of things as just a one-dimensional spectrum. So I completely agree with (a), but I’m sad that you omitted mathematicians/statisticians from (b)!

  12. Tamsin Edwards

    Ben – I would classify your mosquito intervention example as a question of decision-making, rather than scientific assessment, and that it relates to Richard’s point that mitigation is not the only decision to be made: there is also adaptation. Would you agree?

    Sean – I could have drawn the axes with the origin at the bottom left 🙂 Then the location you refer to would not look special. The axis crossing-point does not refer to some specific state, such as neutral / agnostic.

    Nick M – Welcome! I think I cover this by saying that the horizontal axis is the width of one’s prior probability distribution. If your prior opinion is infinitely confident (has infinitely small width: a delta function), then no amount of observational evidence can update that to a different result. If your prior is “open-minded” (large variance) then the observational evidence will heavily influence your conclusion. Not that I think of everything in Bayesian terms, of course *cough*.

    By the way, I did try to get comments numbered but the plug-in hasn’t worked. In the meantime people can use names and datestamps. Richard – yes, preview comment would be very useful.

    • Nick M

      OK I think I get it now. So sceptics (in the general sense: i.e. scientific thinkers) are more likely to move vertically depending on the current integrated evidence, whereas people on the left uncritically take on board evidence and opinion when it supports their stance, and vice versa. You can ignore my previous comments, early morning and all that.

    • Ben Pile

      Tamsin, I don’t agree with what you seem to have read from my argument. (However, I am at least as much an insomniac as you, as you can see from the timestamp, so I take responsibility for it.) My concern is that the policy is assumed by the model, in the notion of ‘impact’.

      I appreciate that you want to delineate policy and research, but I’m suggesting this is implausible — impact cannot be understood purely in material, scientific terms, either in terms of consequences for humans, or for the natural world. The compass problem shows why. Even when we make an ‘axis’ mean something (I.e. a chain of reasoning, a series of positive claims that progress), we see that we have to assume something to make them meaningful before we even begin. The same is even more true of a compass with more nebulous categories determining the axes: any coordinates will only speak to prejudices/existing values.

      To get to the bottom of a disagreement, there is only one method… A continued dialogue. Creating special geometry and taxonomies through which to understand the debate and its players is more likely to defeat progress, if not, irritate people. The debate already has terms, concepts, and claims which are disputed. But I see more effort made in the clumsy attempt to squeeze those mini-debates into categories, or to impose coordinates over them — on both ‘sides’ (oh dear).

      • Ben Pile

        I think it was a revolutionary of some kind who made the observation that once you give something a name, you don’t have to argue with it.

        • Dean_1230

          You raise an interesting point about labeling people. Why is there a need to even try? We all have our preconceived notions of ourselves, we all understand what and why we have come to our conclusions, but by putting a label on us, all it does it invite people to argue the label rather than the substance.

          For example, in some discussions I’ve had, I’ve been labeled a “denier”. When I’ve pressed those who have called me that on what exactly I’m denying, they invariably list the common things that “deniers” are supposed to believe: CO2 isn’t a greenhouse gas, the temperature isn’t rising over the last 200 years, CO2 levels aren’t rising, etc. The problem is that I don’t deny any of those, and yet I am still “labeled” as a denier and even in some cases, an idiot.

          As the comments here have shown, everyone grades themselves differently on the axes you’ve shown. What some call skepticism others call uncritical. The axes are totally subjective and all they do is bin people into categories that can easily be dismissed by the “other side”.

  13. lapogus

    I agree with Ben Piles’ comment. All I was going to say is that the axis diagram is far too simplistic (a bit like climate models!)

    On your comments policy; I can understand that you wish to proscribe the use of the words ‘denier’, ‘liar’ and ‘fraud’, but we are all humans and it is a fact that a subset of humanity do deny plain and simple facts (and I am sure on on both sides of the climate debate), and that some individuals are dishonest, and some do commit fraudulent activity. The idea that there never have been dishonest scientists and are none today in climate science (and other fields) is at best optimistic and to be frank laughable. If there is good evidence of a scientist being dishonest, or to take the individual out of it, dishonest science, and this blog will not permit discussion of it (if it is relevant to the thread) then this blog will not achieve much in terms of establishing the truth and furthering science.

    • Nick M

      Interestingly, all these words refer to people who, for reasons or others, actually lie on the left of the diagram (even though they may claim otherwise). This implies that the negative language is directed towards narrow-minded individuals. Of course those same people will always be more likely to assume their opponents are also…

    • John Russell

      I understand completely why Tamsin wants her comments policy to be as she describes.

      In practical terms: if some tells an untruth you have a choice of saying that they are a liar or that what they say is untrue. If you accuse them of lying that leads to a unpleasant spiral of name calling which is of no interest to other readers and will mean that in time no one will bother reading the comments. So if some tells an untruth, then the correct response is to say that what they have written is not correct and immediately provide the reasons for why you think this, preferably with a link to back up your position.

      The rules are simple, though to follow them requires self control. If a person has no self control then perhaps they’d be best posting on some other blog that caters for their personality type.

    • Tamsin Edwards

      A quick reply from my phone. As Richard said on Bishop Hill, I merely wish to keep this blog focused. If you think there is something doubtful about a method or result I discuss in a post, please do voice it. If it is about other science/scientists, please do not.

  14. HaroldW

    Interesting! My own categorisation is also two-dimensional, but differs from yours. One dimension corresponds to your “conclusion” axis — that is, whether the likely anthropogenic effects are negligible (or zero!), small, sizable, or catastrophic. The other axis is policy response, although this can not be neatly fit onto a linear spectrum. While there is a correlation — e.g. those who believe in catastrophic anthropogenic effects nearly always insist on immediate, global, high-intensity mitigation efforts — those in the middle of the conclusion spectrum (“lukewarmers”) have diverse opinions on research, energy, mitigation, and adaptation. [Of course, it’s true that the conclusion axis is also not perfectly linear, and one’s approach to policy depends on whether, say, a moderate anthropogenic effect is considered to be predominantly from greenhouse gases, or has a significant non-ghg component.]

    Although my personal taxonomy doesn’t include it, there certainly is a spread in “approach”. Notably, your chart doesn’t indicate any population in the middle. My own experience is that there are many who don’t feel confident of their own abilities to independently evaluate the scientific evidence, so avoid the right half-plane (“sceptical”). [To be sure, the discussion does tend to get technical quite quickly. And there’s a lot of utter garbage about, which has the superficial appearance of validity.] Those middle-approachers tend to attach themselves weakly to a trusted source of authority, but are genuinely open to arguments, as opposed to the “uncritical” masses. I welcome any effort which has the effect of moving people to the right on the “approach” axis!

  15. amabo

    Wouldn’t it be more accurate to use the term disbeliever rather unbeliever? Technically, everyone on the chart except the believer and the convinced is an unbeliever.

  16. Jack Hughes

    You forgot to explain how this model can predict anything or help to understand the outside world.

  17. Paul Boyce

    Hi Tamsin,

    Interesting graph. I’d classify myself as an “unconvinced”, but if I were being honest I think I’m more a “convinced” actually.

    By the way, I’m a sceptic, with a maths degree, appalled at what is being done in the name of (Climate) science.

    But there might just be hope for me. I find Richard Betts very persuasive, for example.(Personally I think a Richard Betts is worth a 1000 Paul Nurses to “The Cause”, but that’s just me). And if you were to join the Richard Bett’s camp – well who knows …

    I hope you will enforce your rules of engagement, so to speak. There’s far to much gratuitous insulting going on, on both sides. Completely unnecessary, and irritating to boot. In particular, I do hope you will not allow the use of the “D” word – no, not the “Delingpole” one, the other one!

  18. Jack Hughes

    For a future post you could cover your own work on computer modelling. Maybe a quick overview of the approach that different models take and the reasons for building these models.

  19. John Russell

    Interesting start to the blog.

    Given the choices available, I have to put myself high up on the vertical axis and well over on the right of the horizontal axis. However, by definition, because I am well over to the right (speaking axially) my views are not entrenched and I am prepared to consider any evidence; right up the point where the author demonstrates they are manipulating the facts to suit a pre-conceived agenda. Of course, once someone has done this repeatedly they tend to be mentally put in a box and ridiculed, and anyone who cites their work immediately runs the danger of being tarred with the same brush.

    I think one of the things your ‘climate compass’ shows, Tamsin, is why the term ‘sceptic’ is not a very good description of those who are entrenched in their opposition to AGW. In other words, both the AGW convinced and the AGW unconvinced will place themselves right over on the sceptical end of the horizontal axis and I’m guessing that absolutely no-one will admit to being left of the centre line. The question is, of course, not where we would like to see ourselves as individuals but where the objective observer would put us.

    The big service you can do us Tamsin is to help us arrive at a title for those who dwell in each corner of your graph: a title that’s not only accurate — that’s the easy bit — but a title that’s acceptable to those that inhabit each quadrant. How do we describe someone who clearly inhabits the bottom left hand corner, without using the — I contend accurate — term ‘in denial’? The problem is, you see, that if I suggest that someone is an ‘unbeliever’ (your term) it immediately suggests that I’m a ‘believer’, rather than the true opposite, which is ‘convinced’. Apart from anything else, ‘believer’ and ‘non-believer’ have too many negative religious overtones.

    • Tamsin Edwards

      “The question is, of course, not where we would like to see ourselves as individuals but where the objective observer would put us.”

      Yes – I meant to make that point more clearly. I put the three dots on the far right because that is where we aim to be. We may all sometimes stray leftwards…

      Religions don’t have a monopoly on believing things 🙂 I didn’t think the connotation was strong enough to avoid using it; or at least, I couldn’t think of a better alternative at 2am…

      • John Russell

        As you know, Tamsin, the phrase ‘true believer’ is well used in this debate as a term of abuse. As a consequence the word ‘believe’ has acquired emotive overtones and is therefore avoided by everyone but the innocent bystander. As a writer I’m very aware that once a word is appropriated there’s rarely any way to undo it and consequently it has to be avoided.

        Remember: what one intends by a word is irrelevant — all that matters is how it’s perceived by the reader.

        • Sashka

          Yes it is a form of abuse but it’s not just emotive. Most people believe in a lot of things that they cannot independently check.

        • Aussie

          John there are some excellent books on this subject about beliefs and our belief system. The way that some people behave and advocate on the subject is the reason why they get labelled as “true believers”. In other words, there is a sense of religion about their advocacy.

      • Sashka

        “Religions don’t have a monopoly on believing things :)”

        I saw the smiley but I’d like to reply seriously. I think it depends on how you define religion. To me, any unfounded belief is like a religion (or superstition if you prefer). Whenever people stick to a dogma that cannot be toppled by observable facts, it’s a form of religion.

  20. James Pope


    Good stuff so far, as per support on twitter I like the use of the quote. This looks really promising and hope it develops as you hope it will. I do have one MAJOR ISSUE with your plot….

    No uncertainty estimates 😉

    Certainly, in the last year the engaging with lukewarmers/sceptics has made me realise that some terms such as “denier” are not helpful in the most part, I still consider denier a useful term for someone who denies that greenhouse gases/greenhouse effect is not real. While I would a year ago have grouped all climate sceptics with deniers and used the terms inter-changeably, an attitude I have changed, as it is unfair, I still think there are deniers out there, I just recognise that these days most sceptics are doubtful of either the sensitivity of climate or the impact of the warming.

    Don’t know how others feel on this view, certainly not looking to offend.

    • Anteros

      James –

      A helpfully honest Comment. I picked up on this –

      most sceptics are doubtful of either the sensitivity of climate or the impact of the warming. </blockquote.

      There is something interesting hidden here. Would you say that the consensus is 'certain' about the sensitivity or the impact? I assume not, therefore the two positions you postulate – lukewarmer and sceptic on one side and 'unnamed' on the other – are in fact very similar. Both are uncertain about both sensitivity and impact. Perhaps the root of the disagreement is something else – something deeper – and this is obscured by either side veering towards its nearest extreme? Perhaps the disagreement is actually about how people feel about the future, and risk and change and human agency.
      I think Tamsin’s axes create a slightly false picture of how human beings conceptualise the future, and by that I mean it has precious little to do with reason – in comparison to how we persuade ourselves we think about the future. I would contend that anybody sure of the ‘impact’ of 2 feet of sea level rise for people 60 years from now has been using imagination rather than reasoning.

      • James Pope


        Yep, preview needed and now added I see!

        There is an uncertainty in the estimate of sensitivity (IPCC range 2-4.5 K), something I’m sure Tamsin will delve into, in time on this blog, likewise there is uncertainty into the scope of impacts, not sure if she will focus on this, I’m sure we’ll find out!

        I would say, there is a consensus (amongst climate scientosts) on the causes of anthropogenic climate change, that the world is warming, that we are causing it. Most people would agree that sensitivity is at least within the IPCC range for sensitivity (2-4.5 K for a doubling of CO2), some would argue it is higher. There are some (more outside of climate science I find, but some who are climate scientists) would argue it is lower, I generally see these people as the lukewarmers, they accept the above (causes and drivers on ACC), just think the climate isn’t as sensitive as the IPCC estimate and the models use (I apologise if I have generalised here, just my POV).

        Tamsin, I’m sure will delve into these sorts of things though!

  21. Mydogsgotnonose

    This representation isn’t the key issue. Climate science makes four basic scientific errors.

    No other scientific discipline claims that it can buck the 2nd law of thermodynamics [‘back radiation’ from an imaginary heat source]. No process engineer accepts this because we use the theory for a living so have to get it right. If necessary, I can prove why Arrhenius got it wrong.

    The concept of 100% thermalisation of absorbed IR has to be chucked – Will Happer, an IR specialist, warned of this in 1993. Thermalisation occurs at second phases, particularly clouds; the idea of the ‘hot spot’ in clear sky is impossible.

    Hansen’s 33 K present GHG warming includes lapse rate. Real GHG warming, if you accept such an average, is far less.

    The aerosol optical physics of clouds is very wrong for about half low level clouds which have bimodal droplet size distributions. The incorrect physics is used as a fudge factor to offset the incorrect warming prediction

    Until these basic errors are fixed, no climate model can predict climate and the pressure from the rest of science for climate science to accept it has got some of the basics very wrong will increase. In essence, the reliance of the subject on the ideas of Fourier, Tyndall and Arrhenius, good at the time but now disproved, condemns it to being an historical relic no matter how good the climate models otherwise behave: GIGO.

    • Tamsin Edwards

      I’m afraid back radiation is off topic. I believe Judith Curry has discussed it.

      There is no point in me starting a blog if it reproduces conversations that are going on elsewhere. This space is about how we try to assess uncertainty and usefulness in complex, imperfect models: not about making a priori assertions that they are useless. Perhaps a climate model developer would like to answer your queries, but if they do not I would like to refer you to comment policies (e) and (f). Thanks.

      • Mydogsgotnonose

        Fine, but I hope I have made the point for all professional engineers and physicists, the educated core of the sceptics, about this claim which is unique to climate science! It cannot be allowed to continue……..

      • billc

        I’ve chosen to nest my reply here rather than at the bottom. If you don’t like that, please move it to the bottom and i’ll follow suit in the future.

        Good post, I like the originality and it made me take the test (Economic Left/Right: -3.75
        Social Libertarian/Authoritarian: -4.87). Got a test for climate? It would be fun to come up with the questions.
        Until then, lukewarmer all the way.

        I applaud your link to Curry’s blog and want to declare my support for both of you…and I think the modeling focus of your blog is a great complement to hers, of which I’m a big fan. I’m with your comment policy all the way (echo). That said I’m looking forward to posts more technical than today’s. I also want to note I find Isaac Held’s modeling blog rather obtuse, though the content is good. IMO it is directed at an in-specialty expert audience, rather than a generally technical audience.

        As far as blog style I like the way Curry’s blog has turned out, with the exception that the limitation on nesting (currently at 3 I think) could potentially be increased, but with Anteros already in residence here, that would be like giving a mouse a cookie.

  22. Rupert

    Great blog Tamsin! Couple of comments.
    1. I think this post is too long for a blog post, sections 3 and 4 could have been left for seperate posts.
    2. I am confused about ‘sceptical’. You say that a ‘sceptic’ is at the bottom of the vertical axis yet you label the right end of the horizontal axis as ‘sceptical’. Surely the opposite of uncritical is critical rather than sceptical? I am a scientist and I don’t endeavour to be sceptical, I endeavour to be critical and objective.
    I think this compass is useful. Yes, it is a simplification of reality, but simply categorising everyone into either ‘sceptics’ or ‘warmists’ (which seems to be the default these days) is surely worse. So I think the compass is definitely a good idea, so long as it is used in moderation along with the knowledge that peoples view cannot really be so simply pigeon holed.
    Regarding 4h, and being someone largely devoid of humour of any kind, I shall forgo my ‘sprinkling’ of good hhumour and smilies and compensate by expanding my ‘sprinkling’ of truth to a flood.
    On a more serious note, there are a lot of items in your comments policy list. I instantly agree with 4a, 4e and 4f, but if you could justify the others in a future post that would be helpful to me.

    • Tamsin Edwards

      I meant sceptical in the sense of critical and objective. I did think of using Critical, but wanted to highlight the “scientists are sceptics too” point and also to be able to call it the Sceptical Compass 😉

      • Theo Goodwin

        You did the best anyone could do. Our language is not exact on these matters. Science is the critical and self-critical discipline “par excellence.” Scientific method has criticism at its heart. However, if you had gone down this path it would have confused a lot of people.

  23. mondo

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

    I think that this doesn’t quite capture the issue, from my viewpoint at least. What I am unconvinced by is that there is any real evidence that anthropogenic CO2 emissions are causing potentially catastrophic global warming. Roger Pielke Sr argues that man IS having an impact on local and regional climate, but mainly through land-use effects. It is also evident that there are natural cycles to be taken into account. And add into that equation the effects of changing UHI effects. My point is that there is no clear evidence (as opposed to model assumptions) that anthropogenic CO2 effects are greater than land-use, or natural effects, or UHI contamination of the record. In my view, I fail to see how the Convinced scientists can claim to be sceptical. Where is the evidence?

    • Tony Ratliffe

      This responds to “Mondo” at 10.57. He said “What I am unconvinced by is that there is any real evidence that anthropogenic CO2 emissions are causing potentially catastrophic global warming”. I think this gets to the heart of the whole controversy. Unless it can be shown to be more than vaguely possible that there is a simplistic causal chain from anthropogenic CO2 to catastrophic global warming, than most if not all of the proposed political actions in perceived response are mis-conceived. Discussions of practically everything else in the climate science area are interesting (always!) but essentially academic at this time.

      Regards, Tony.

      • Theo Goodwin

        Spot on. Have climate scientists produced some well confirmed physical hypotheses that explain how rising atmospheric CO2 changes cloud behavior and other such variables to yield a higher average temperature for Earth? But I think our topic is models. Maybe the question will be whether models can substitute for the desired physical hypotheses.

    • Tamsin Edwards

      I think this view would move you closer to the lukewarmer position that I described. But the boundaries are fuzzy.

      “Where is the evidence?” is rather too broad a question for me to answer! Let’s see if future posts sound convincing to you.

  24. John Costigane

    2 dimensional graphs are useful, but 3-d are better: think of this post with the actual climate. A 3rd axis, difficult on paper without 3 figure location, could be Knowledge on climate (ignorance to full knowledge). This could be the knowledge of individuals or the depth of scientific knowledge.

  25. Maurizio Morabito (omnologos)

    * “disbeliever” should be on the bottom extreme right of the graph. It’s all those people who have been convinced there is little to worry about, by the combination of their sceptical approach plus the bombastic claims of believers and convinced. And the more of those claims accumulate, the more their disbelief will become entrenched, so they will move even further to the right. It’s people who apply Sagan’s “extraordinary claims” idea to the max, in the area of climate change.

    I’m afraid most skeptics with numerate background are in that category already.

    * “catastrophist” should be on the centre of the X axis, extreme upper part of the Y axis. It’s those people who on the basis of scientific findings believe that CAGW is going to befall upon us.

    A great number of debate-challenged, fudge-prone climate scientists belong to that category.

    • Nick M

      Someone with a critical approach should never be swayed either way by “bombastic claims” of any kind, only by the overall weight of evidence. Those who (truly) are on the RHS of that graph should treat information coming from the LHS as biased and ignore it.

      A fair test of ones actual horizontal position would be how likely they are to change their position if a new piece of strong data changes the overall weight of evidence.

      • Sean Inglis

        Everyone would claim that they would change their position, but you would then engage in horse-trading about the meaning of the term “strong data” and how it should be weighed alongside existing evidence.

        On an individual basis, it would be interesting to have pairs of respondents with apparently diametrically opposed viewpoints plot where they think the *other* lies on the graph, and then discuss who’s placing who more accurately (and why).

        Perhaps a bit too much of blood to clean up afterwards.

        • Maurizio Morabito (omnologos)

          The position-changing event should be declared and detailed before it happens. IOW the question we should ask each other should be, “what would make you move from eg unbeliever to convinced?”. All those unable to answer would then ipso facto placed themselves out of the debate.

          This alone would IMNSHO reduce the noise in climate blogs below visibility.

          As for the power of bombastic claims, nobody can seriously avoid considering the fact that they are bombastic, therefore likely used in situations where the evidence is shaky to non-existent. Cry-wolf and all that.

  26. pouncer

    Is there a place on the chart for those who look at the debate from a political-process point of view rather than a statistical-process view?

    After all the UN-IPCC is political and claims no originality WRT the science. The various NGO’s such as Greenpeace advocate policies but don’t do research. The US EPA is concerned with formulating remediative policy more than developing new models or measures.

    Reacting to those realities, some of us are “skeptical” that rules of evidence, open meeting, open record, due diligence, and other POLITICAL processes are working according to the standards which generally apply to most other comparable fields. We compare the political processes in “climate science” to those in medicine, hazardous material transport, civil engineering, what have you, and find the “science” wanting — not so much from a scientific viewpoint, about which we have little opinion, but from the process. How does the matrix address this approach?

  27. oldtimer

    Using your chart, I put myself into the unconvinced category. One contributory reason was the failure of the climate scientists to make a parallel run of their land based temperature data sets for the pre and post 1990 period. It was around that date that the number of stations used dropped from c6000 to c1200, with only 200 common to the pre and post 1990 period. Others have pointed out that this change also removed all stations at higher altitudes and included a noticeable drift from the poles to the equator. The Muir Russell and Oxburgh enquiries into Climategate were made aware of this, indeed Muir Russell clarifed that the reduction took place, but did not investigate the implications. It has been suggested that it would now be impossible to make that parallel run. So we are left with the situation that no one knows, nor can they discover, the implications of this change. In the world I used inhabit, the financial world, this would have been utterly unacceptable. Chances are that the authors of the change would have been fired.

    Comments by many others on statistical methods used (such as by Matt Briggs and Bishop Hill) and comments revealed by the climategate emails cast further doubt, in my mind, on the quality control of the data and arguments presented to us. Yet here in the UK, they are the foundation for £billions of expenditure on schemes which themselves fail to deliver the reductions in C02 which its advocates proclaim to be necessary.

    • Tamsin Edwards

      I’m afraid the observational record is off-topic too, unless there are specific links to model assessment. Earth system science is too big a research area for me to cover it all. Hopefully soon I can write about some science so it will focus the discussion…

      • oldtimer

        OK. I come from another world where changes to the data set need to be understood and explained before they are used for modelling, forecasting, projecting and reaching conclusions thereon.

        • Tamsin Edwards

          It’s more that I’m interested in the general principles of “challenging” models with data. Specific questions about those data (whether historical or palaeo) are appropriate – eg whether their uncertainties have been appropriately quantified – but only under a post about that data. Not under one about terminology. If anyone can bring up any topic under any post, then the comment threads will become unmanageably long and people will tune out.


    • steven mosher

      old timer. that question, the great thermometer dropoff. , has been studied completely by zeke nick stokes and me. there is no problem. we looked at it with 3 different methids and several analytical approaches including the one u suggest

  28. AlinVA

    Doesn’t this depend on what is meant by “Intense & Rapid Impacts”? There are those who believe that the oceans will rise by 3 meters, and those who believe it will rise by 1 cm. They may both be “convinced”.

  29. phinniethewoo

    is this goin to be a science communication site or just a site like the blackboard, with lots of pseudo intellectual clucking ?

    generalisations are a sharp knife in the scientists’ drawer, btw.
    eg Newton spoke about all all objects when not hindered keep moving or stay still. He didon’t speak about objects for the proud gays move such, objects for all too long oppressed females move henceforth, etc.


    ALL objects he was talking about.
    generalisations are the stuff laws are made from.

    • Tamsin Edwards

      I’m not sure the point are you making by mentioning sexuality or gender. Please use more neutral examples in future.

      Of course laws are made on generalisations, but I was referring to people. I can clarify this when I make the policies a permanent post.

  30. Steveta_uk

    Not sure I understand the link between climate and the width of your posterior. Perhaps I misread.

    • Tamsin Edwards

      I meant that the y axis is ‘Conclusion’ based on available evidence and other factors such as methods and assumptions. This would correspond to the posterior distribution – the mode, or (smeared out blob) the width – obtained by updating a prior belief with observations. Not sure if that helps…

      • Gras Albert


        Thank you for the blog, I look forward to continuing my climate education with the help of your posts and the informed comment they promote, such a contrast with the science advocacy which masquerades as information presented by some of your colleagues.

        However, I do feel obliged to point out that you may have missed the point of Steveta_Uk’s post at 12:55, I interpret his contribution as being more reflective of your blog rule 4(h), perhaps paraphrasing, was he not asking does your bum look big in this?

  31. John Shade

    This new blog sure has flair and style, as well as, I hasten to add, the promise of substance to come.
    The simple model with two axes being of course ‘wrong’, I felt encouraged to try my hand at this. Here is an even simpler model, crude as well as wrong (!) since it just has 4 possible states defined by binary labels A and B. Defining the population that might be spread around it is one of the weakest points of this model, but it somehow includes those convinced of doom in the near future, by CO2 or otherwise, and those not so convinced, stratified further into those who believe additional CO2 has a major role to play in the climate system and those who don’t.

    A. Propensity to spread fear about imminent catastrophe: Lo, Hi
    B. Propensity to give CO2 a big role in climate change: Lo, Hi

    (A,B) State Example sites

    (Lo, Lo) Bemused

    (Hi, Hi) Alarmed

    (Hi, Lo) Alarmed,

    (Lo, Hi) Concerned

    ‘Bemused’ means puzzled by the level of policy changes underway and proposed and by the strength of the convictions of those pushing them using CO2 as a key lever

    ‘Concerned’ means wishing for more actions based on climate projections and puzzled by the lack of similar conviction in others about the impact of CO2

    ‘Alarmed’ means sounding the tocsin at every opportunity, convinced of imminent doom

    • John Costigane

      Lo John,

      I was going to do “Hi” but that has sinister connotations, like sustainability. Pachauri has jumped ship to the sustainability quango from the IPCC, maybe not all bad!

        • John Costigane


          Maybe all sceptics should use ‘ lo’ as part of their post intro. That way most readers could read the views they like.

  32. Dead Dog Bounce

    Thanks you for starting a blog, Tamsin. I’m an enormous fan of civility, and also think that good fences make good neighbours.

    Can you please explicitly state within the blog rules how you think questions of ethics should be treated? As this is “your house”, you are entitled to rule that this matters are not to be touched. This is certainly one interpretation of rule (c), but clarity would be welcome.

  33. Nick M

    One final thing. If the evidence is strong, clear and consistent, then there should only be a single point on the right hand side (I only know one person who disbelieves badgers exist, for example). It would be interesting to construct an x-y density plot, using some suitable proxy for x.

  34. Anteros

    Interesting post.

    I very much agree with Ben Pile’s first comment. With respect I think your two axes frame the debate very poorly and not only rule out many reasonable positions (including mine) but make some fundamental assumptions that are at the root of the disagreements. I actually think this kind of attempt at 2 dimensional framing makes dialogue even more difficult.
    While I’m being critical (sorry!), I agree with Paul Matthews about the false – and damning – distinction between science and scepticism. You could make a better case for distinguishing between sceptical scientists (ie real ones) and those prone to fearful imagining and leaping to negative conclusions. Those who, perhaps, say things like “coal is the greatest threat to civilisation and all life on earth”. I think for the majority of those who are labelled ‘sceptic’, the genuine science is just fine.

    Where they disagree with the ‘orthodox’ position is mostly what the science says, what it means, what it can tell us about the future and most importantly what the implications are – which I know is outside the scope of your blog, but as Richard Betts has said so persuasively, is also outside the scope of both climate science and science itself.

    The nesting and usability of the blog are already greatly improved and I think you got your blog policy spot on. Also, I think you’re making a very wise move to be quite specific about what your blog is about – and what it isn’t. Unfortunately it means my own interests [the philosophy and psychology of perceiving the future aka armaggedonology..] are always going to be slightly O/T but that’s my problem not yours!

  35. Eric (skeptic)

    The horizontal line can be thought of as a brick wall, requiring some effort to cross or even look over. The concern of the Northerners is simply physical: climate change. Southerners often have philosophical, political and sociological concerns about applied science and the free market. That is a mismatch that makes it hard to reconcile many arguments because the unconvinced skeptics consider the physical concerns to be mostly speculative and convinced skeptics will devalue concerns that are merely philosophical or blatantly political. There are exceptions obviously such as James Pope above who peered over the wall and accurately characterized the south side.

  36. Barry Woods

    Here is a great quote from a great commuicator of science 😉

    “I would personally be infuriated if I was dismissed on account of the behaviour of a group of people I talk with. Every single person I talk with has a different viewpoint, and I learn a lot about how better to communicate climate science by listening to them. If we dismiss swathes of people by association then our attempts at communication become futile: we end up only ‘preaching to the converted from an ‘ivory tower’, as it were”.

    Of course, if communication of climate science is not your aim, then it is your choice if you prefer to communicate with nobody! – ?
    that summarised how many sceptics feel, I think….
    Guess the author, before you follow the link….

  37. Alexander Harvey


    This is just a thought suggesting that you might be wise to use the same terminology here as you would normally. For instance the term models: all simulators are wrong in a way quite different from any notion that an emulator (particularly of a simulator) could be said to be wrong. The term model means many different things to many different people. If you use those terms distinctly, and few seem to, do so. Similarly expose your science in your prefered terminology rather than “helpful” metaphors. It is just a thought.

    I will further suggest, and kindly so, that you consider the importance of frames, and do without those that you do not really need. You asked if we liked your framework, personally no I don’t, How might it be improved? By not going there, would be my first consideration.

    I will try to explain my position. Your initial points a) b) c) were clear, your concluding points of that section were clear, in between I found a framework which I presume intended to add structure and clarity to the whole. The hazard being that those that dislike the frame elect not to see the picture (oops metaphor alert). 🙂

    I do try to at least understand the wisdom of frames and narratives but that understanding relates to the notion of there being a message. I temper that understanding with my observation that the process of trying to communicate the science resulted in a frame-wreck. I hear some opine that there has been a failure to get the framing right: “if only we could get the framing right and the narrative right and one more heave” etc. I will be happy to argue from a presumed minority of one, that framing and narrative, even messaging are very last century and that the imposition of frames has no role in the communication of science beyond some initial motivation of its importance. Science has or is its own frame.

    Climatic change, global warming, climatic distruption, global wierding are imposed frames. I can think of no part of climate science that would cease to exist in their absence. They may offer ample motivation to study science but they can be checked at the door.

    However I think that your choice of the George Box quote is a useful motivational tool (OK it is a frame). I will not deny that scientific understanding and evidence feed into environmental frames and other frames but I am prepared to argue that there is a hazard when a single group or individual tries to present both framework and content. I do not presume your intentions but you did kindly ask and I kind of replied.


  38. Pharos

    In my view each specific scientific enquiry should follow the ‘a posteriori’ path of rational observation, analysis and then conclude with a critical debate on possible causes. I think, largely due to environmental/political framing and funding policy, the academic study of climate science, particularly the freedom to reach counter-consensual conclusions, has been severely handicapped by the ‘a priori’ assumption that climate variability is abnormal, and the current condition is ideal.

    Human impact is always presumed to be harmful. Some ‘lukewarmers’ may conclude that within their expected range of enhanced temperature and CO2 concentration, the impacts will be significant to high on enhanced cultivable area of arable acreage, growth season and the CO2 fertilization effect on crop yields.

    • Faustino

      I hope that no-one thinks that “the current condition is ideal,” because we live in a world (and universe) of constant change, in which there can never be an ideal, constant, state. While I appreciate that Tamsin (very sensibly) wishes to constrain her blog to certain topics, there is perhaps a little license in this introductory, framework-setting post; so I’ll add that many on the warming side, some of them scientists, do seem to assume that we should seek to maintain aspects of the world in a very limited range: this is where a lot of controversy and debate/dispute arises. As for modelling, the Australian government has claimed one reason for it’s carbon tax is to save coral reefs from the disaster that modelling suggests will follow warming; recent extensive research shows that coral reefs thrive in warmer times (sorry, ref not to hand, paper released in last few days).

  39. Francis Turner

    I like the rose idea but along with a number of other commenters in the numerate/lukewarmer skeptic area, I think you need more/different axes. Fundamentally there are at least 3 somewhat related theories/hypotheses here and it is the combination of the three that is the problem.

    The reason why this topic is such a political one is that politicians and environmentalists (and some scientists) are worked up about CAGW – Catastrophic Anthropic Global Warming. There are 3 separate concepts in that and each one deserves an axis.
    1) Global Warming
    In the context of the last century or two I’d say most people accept that this has occured but there is certainly a question about the degree of GW and when it started/ended (if it has ended). This is probably the most concrete theory and it is quite easy to quantify so you can have an axis that goes from say 0 to 2deg C or so and anyone can put their value on that (with error bars)
    2) Anthropic
    This is the theory that (1) above is primarily caused by humans and not the sun or other long-run natural cycles. I’d say you need a 0-100% axis of how muchof GW is caused by humans and again some error bars. In my case I’d say the error bars are fairly big but I certainly think humans have had at least some effect on global warming but how much is unclear
    3) Catastrophic
    The hypothesis that (1) is going to lead to a catastrophe. I will note there is no need for (2) to be true or not for (3) to be true.
    This again is a 0-100% distribution of probability (100% == it is certain that there will be a disaster, 0% == it is certain there will be no disaster etc.) and again it seems reasonable that you have a range and/or error bars.

    I think you will find a lot of us would put oursleves at about 1 deg C on axis 1 but below 50% on both points 2 & 3 – and probably move down different amounts in both cases. I personally think that (2) is entirely possible – though I don’t think it is proven or even close – but that (3) is anywhere near as certain. I can certainly understand how people could think the opposite [(2) unlikley (3) quite likely] or think that neither is anywhere close to being proven or indeed other combinations.

    In fact I think I see how to do this, Since 1 is a number and 2/3 are probabilities you can do your rose for 2 & 3 and stick a number for 1 in the middle.

    • Anteros

      Francis –
      I agree with your analysis overall. One slightly technical point – you say that the A influence on GW should run between 0 and 100%. I think that is a mistake. Given competing influences it is quite possible [and likely in some circumstances] to be either a negative number or greater than 100%.

      For instance, if the natural variability and extraneous forcings were negative during the second part of the 20th century, the A influence would have to be greater than 100% as there was a net warming.

      This explains why Michael Mann can suggest that the observational data are consistent with a climate sensitivity of between 1.5 and 9 degrees C per 2XCo2.

      FWIW, I suspect there are a few climate scientists who would say that if the data is consistent with 9 degrees, it is also consistent with a number lower than 1.5. But, if there aren’t I’ll be very happy to admit i was wrong!

      P.S. The ‘negative number’ I include for the sake of the example – I don’t suggest there is any evidence for it.

  40. Roddy Campbell

    I think sceptical and experienced people just can,t believe the impacts models. Any more than they believed the much more tractable And less complex impacts of bird or swine flu or all the other scientific impact projections, which didn’t even have all the policy and politics baggage of Agw. My sceptical problem is that I found we weren’t doomed after all.

  41. Ron Manley

    I’m not happy with some of binary pairs that have been suggested. Convinced/Unconvinced and believer/unbeliever both suggest that one one attitude (convinced or believer) is correct and that everyone else will one day come to see the light.

    • Ben B

      I also see binary pairs but in the other dimension. I am not interested in talking to either Believers or Unbelievers, but am interested in talking to people in the sceptic/critical dimension (be they unconvinced or convinced).

  42. Green Sand

    Hi Tamsin, I trust all is going well.

    As requested in your comments policy I have posted a specific “bugbear” at “the very suitable Unthreaded section of Bishop Hill;”

    Would appreciate your comments, or Richard’s, when you get time, understand that you are all busy and it is not urgent, but I do believe that it is salient.


    Looking forward to your future posts.

  43. David Watt

    I think there should probably also be a time axis to your presentation as many of us I should think change our views with time.

    In the 1980’s I was sceptical. Then in the 1990’s impressed be the rapid temperature rise during the decade and the 400,000 year correlation between CO2 and temperature shown by the GISP ice cores I became a believer.
    In the 2000’s the temperature rise waned, I discovered that the ice cores showed the temperatures rising a few hundred years before the CO2 (so that CO2 rise was an effect rather than a cause of temperature rise) and just how iffy the IPCC estimate of climate sensitivity was. Climategate didn’t help nor did the Donna Laframboise revelations about the green activist corruption of the IPCC.

    I am now as I would now have to admit pretty far to the bottom left of the graph – but hang on a bit who can tell what the next decade will bring?

  44. Anoneumouse

    Oh yes, the sceptical compass, clearly indicating the directions to a common purpose.

    This article demonstrates quite clearly the colloquial vernacular verbiage of the Fabian.

    jolly hockey sticks

  45. Steve Bloom

    “I discovered that the ice cores showed the temperatures rising a few hundred years before the CO2 (so that CO2 rise was an effect rather than a cause of temperature rise)”

    It’s fascinating to see this point continue to be stated even though what it indicates is a fundamental misunderstanding regarding a rather obvious relationship that was even predicted (in a paper by Jim Hansen) some years before it was detected in the ice cores. (Of course the propagandists who fist raised it had no such misunderstanding, but correctly judged it to be something that could be used to confuse the public.)

    Here’s a thumbnail, David: On very long time scales, atmospheric CO2 levels are controlled by plate tectonic effects (the relative rates of emission via volcanism and sequestration in ocean sediments, which is related to the configuration of the continents). Over the course of the Phanerozoic, this basic relationship has been buffered by biological effects (e.g. sequestration via creation of fossil fuel deposits). On shorter time scales, Milankovitch (orbital) cycles kick in (important to note that they’re driving a fluctuation, not any sort of change in the base level), which is why when CO2 dropped enough for things to get quite cold a couple million years ago we started seeing regular glaciations rather than a smooth progression of climate. The key point is that in all cases CO2 is being driven by something else, and that if it had been the case that the CO2 change preceded the Milankovitch-induced warming it would have upended rather a lot of physics. So, for present purposes perhaps it would be less confusing to think of humans as the forcing and CO2 as the feedback.

    Ray Pierrehumbert has pointed out that it would be even less confusing to think of all of this as modifications of the Planck feedback, although that would require explaining the latter to people.

    Anyway, the details are readily available on the internet.

    • David Watt


      I have been doing geology long enough to know where my Phanerozoic is and to know that even the Milankovitch effects operate on a scale of tens of thousands of years. The effect shwn on the ice cores is much shorter term and acts over at most a few hundred years.
      It looks like an oceanic soda pop effect. When you heat the ocean up like the bottle it exhales CO2.
      Whatever causes it, as always the cause (oceanic warming) comes before the effect (raised atmospheric CO2).

  46. Steve Bloom

    “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.”

    When convenient, Tamsin, perhaps you could explain why someone like me (a non-scientist who follows the science closely, reads lots of papers, etc.) would benefit from participating here? Re your own research, can I expect to have much added to my understanding give that I already e.g. closely follow the blogs of Annan/Hargreaves and Easterbrook? Other than that, would it just be a “skeptic”-watching/tussling exercise? If so, I have to say it doesn’t seem like a productive use of my time, although there’s certainly some sociological interest in seeing somebody try to do what Judy Curry claimed to want to do but didn’t. And of course an improvement in communications skills is its own reward.

    Just to note, actively promoting participation in the blog of the author of “The Hockey Stick Illusion” is a little weird. It seems either naive or reflective of an inflated view of your own powers of persuasion.

    • John Costigane


      Could you please expand on your comments about Judith Curry. I find her an excellent blog host who allows a vast range of material on climate to be aired, not to everyone’s liking. When was science ever a popularity contest?

      • Steve Bloom

        Way too much bad and questionable science, John. All opinions may be in some sense equal in public discourse, but the same is not true of opinions about science. In addition, there are far too many gaps in Judy’s own knowledge of the science for me to put much weight on her own statements contra the “consensus.”

    • Jonathan Jones

      Steve, if you have specific comments about the accuracy or otherwise of Andrew’s book I’m sure you would be welcome to raise them on his blog. That is probably a more appropriate location than here.

  47. Die Zauberflotist

    I’m all for civility in a blog if, when it hits the fan, we get to make dikes out of the d… du.. du.. d.. “sceptics” to help staunch the rising seas.

  48. Tamsin Edwards

    Die Zauberflotist – please be civil rather than sarcastic.

    Steve Bloom – thanks for contributing, but I would say that “It’s fascinating…rather obvious…” etc is verging on patronising – please write with a little more understanding. I’d like people to assume good faith in “my house” – i.e. that people have genuine misunderstandings and disagreements – rather than that they are stupid or wilfully stubborn. Thanks.

    As to whether this blog is a good use of your time – I ask you to give me a few posts and decide for yourself, rather than ask me to advise you. There will be some overlaps with James Annan and Julia Hargreaves’ research areas, yes, but also differences in approach and application. There will also be topics they do not cover, I imagine, because I am a different person living on a different continent (e.g. if blogging about seminars, conferences and meetings).

    Andrew’s ‘Unthreaded’ space is an open forum where any discussion is allowed. I don’t see anything wrong in directing people to it. They are fair to me and other scientists (almost all the time) there. Have you read the HSI? (I’m just about to, finally, and will probably blog about it).

    • Steve Bloom

      Yes, Tamsin, it was a bit patronizing, but do you really think that’s inappropriate give the age of the claim (about 15 years IIRC), its simple nature, the bazillions of refutations, and its repetition by someone who seemed to be claiming to have followed the relevant science for roughly the same time?

      Re the “Hockey Stick Illusion,” that’s rather too polemical of a title for the book to be taken seriously, don’t you think? (I similarly avoided Fuller and Mosher’s “The CRUTape Letters.”) Also, call me picky, but I prefer my science from non-accountants. But anyway, having followed all of those events closely in real time, I wouldn’t expect to learn anything, and I’m entirely uninterested in Montford’s political spin.

      Re “Bish” and his blog, well, some old saw about dogs and fleas comes to mind. OTOH I doubt it will take more than a year or so for you to satisfy your curiosity about this stuff and move on to greener pastures. Please do keep a running count of any “skeptics” you convince to pay attention to the science. To borrow Calvin Trillin’s phrase, I’d be surprised if the total even reached a high single figure since (repeating myself) for most of them their dislike of the science is all about disliking the implied need for change. I’m sure they’ll continue to treat you politely so long as they think there’s some hope of ideological capture a la Curry (of course Judy was pre-captured, but that’s another discussion).

      • Tamsin Edwards

        Hi Steve,

        I’m afraid you’ve broken policies (a), (d), (g), (h) and possibly others. I don’t want people repeating old arguments and assumptions here.

        I’m sorry you don’t think I’m going to stick it out. It’s been more than a year since I started conversing with Andrew et al., so I don’t know whether that means I’ve already done it or if it only counts from now. Either way, I’d rather you didn’t patronise the host or her guests.

        Please do stay: but please do be a bit more respectful to all and we will return the favour. “Benefit of the doubt” is a guiding principle in my life and in this forum.

        • Steve Bloom

          So hard to keep track, but it does remind me of one of the best cartoons ever.(one of Gary Larson’s lamented “Far Side” panels; not available on line) in which anthropomorphized female and male consonants are having a fight, and she says, “Yes, it’s true! I’ve been seeing all the vowels, A, E, I , O, and U! Oh yes, and sometimes Y.” (For his many wonderful science-related cartoons a biting louse was named after Gary, the sort of honor of which the likes of me can only dream.)

          I didn’t mean to imply you weren’t willing to stick it out with the blog, but rather that after a suitable period of beating your head against the “skeptic” wall you would come to the obvious conclusion and either shift focus or move on entirely. But we shall see. I will go have a look at the BH blog to see what progress you’ve made so far. Out of curiosity, can you point to anything specific as an accomplishment in your time there, other than the obvious one of maintaining polite contact?

          Re my participation, I have my doubts as to whether there’ll be much of interest to me given what you’ve said so far and the demographic to which you’re appealing, but that too we’ll have to see.

      • Richard Betts

        Steve Bloom

        I think you’re getting too hung up on the issue of associating with a certain group of people that you have identified in your mind. My understanding is that Tamsin just wants to have a discussion on a particular aspect of climate science that she knows about, and is happy for anyone to join in as long as they are well-behaved. You say you are not a scientist, which may explain your lack of enthusiasm for the concept – did you know that scientists actually like arguing and explaining their interpretation of the evidence to people who disagree? It’s a way of challenging yourself that you really understand stuff. Have you ever been to a scientific conference and seen the presenters get a right grilling after their talk – it’s all perfectly normal. If this is going to be a blog where everyone just sits around agreeing with each other then it will be pretty boring! 🙂

        Shame you choose not to read some books because of their title or reputation. Again, without wishing to sound patronising myself, that is not the behaviour of a scientist. Someone who is really interested in getting to the bottom of things will read a range of views on the subject and form their own opinion.

        Have you read Fred Pearce’s book “The Climate Files”? That may be a gentle and less threatening entry into the Climategate controversies for you – and it’s very good. After that I would recommend you read Mosher and Fuller (nice and short, and more balanced than you might think) before diving into the Hockey Stick Illusion. It’s well written even if you don’t agree with everything in it, or find it uncomfortable. Mike Mann’s book has just come out as the Kindle edition if you want the other side of that particular story, and I believe the hardcopy is out next month (I’ve not read it yet, but will do soon).

        Another nice, short, amusingly-written book is Garth Paltridge’s “The Climate Caper”. This is perhaps more relevant to this blog as it’s about uncertainties in models and future climate projections – it’s grumpy but engaging, and quite funny in places because of the grumpy style, although you do have to grit your teeth to get past Christopher Monckton’s triumphalist introduction first. Nigel Lawson’s book covers largely the same ground but is much more dull. For the other side you could read Jim Hansen’s “Storms of my Grandchildren”.

        I tried to read Svensmark and Calder on the galactic cosmic ray hypothesis, but got bored – they don’t actually present any evidence, it’s all just narrative.

        Obviously it’s not for me to say what you should read and which blogs you should go on, but if you want to experience science in action then IMHO you be prepared to have (or at least watch) a good old argument (one which respects the rules of the blog, of course!)

        Ah, just noticed the brand new preview button, thanks Tamsin!

        • Anteros

          Richard Betts –

          I shouldn’t be surprised that you’ve been open-minded enough to read books with such ‘diverse’ perspectives. I agree entirely about Garth Paltridge’s ‘Caper’ – grumpy but quite funny. Interesting too, and you’re right about the painful Moncton introduction..

          Steve Bloom – if you want no-holds-barred hard-core, I recommend Michael Mann’s ‘Dire Predictions’. A good test of objectivity is to read the whole of AR4 and then Mann’s book straight after, as it claims to be a representation of the science. Interesting…

          For my money Pat Michaels’ books are a surprise. Political, sure, but much more reasonable than many partisans would expect. If Nigel Lawson’s book was dull [it was], David Archer’s is duller, but makes the interesting claim that most people would prefer a colder climate [LIA, anyone?]

          Seeing as tribalism is the biggest impediment to debate, I’d recommend Mike Hulme’s ‘Why we Disagree about Climate Change’.

        • Steve Bloom

          More like their writings and expressed views, Richard. I’d say shame on you for misrepresenting what I said, but that’s OK. Tell you what, though: Post your reviews of each and I’ll reconsider if it seems like they might contain interesting material with which I’m not already familiar.

          BTW, I’ve been around this “debate” and actively studying the science for about ten years now, so bear that in mind when making suggestions to me. Re “Climategate” in particular, of course I followed that in real time. Why assume I hadn’t?

          Re my lack of enthusiasm for this blog conversation and how unscientific my attitude is, maybe you should start by explaining why most scientists don’t even have a blog, let alone one like this. The numbers would seem to indicate that you’re in a tiny minority for thinking it’s something worth doing, especially considering the firm pushback Judy Curry’s similarly-conceived blog has gotten (although to emphasize I have no reason to expect that Tamsin will go that way). But FYI I am a long-time participant on a number of scientist-run climate blogs (I even named a couple above), including RealClimate from its inception, so clearly I have no beef with climate science blogs in general.

          • Richard Betts

            Hi Steve

            Please feel free to read, or not read, whatever you like, I really don’t care – you brought the subject up. I’m afraid I have better things to do than write reviews in order persuade someone to read something – I read all the books I mentioned simply because I was intrigued, although for three of them it was also made easy for me because someone directly gave or lent me copies – I read Paltridge’s because Jim Lovelock (a mutual friend of Paltridge and myself) gave me his copy, Montford’s because my Met Office colleague Vicky Pope lent me her copy (given to her by Montford – although I since obtained by own copy) and Mosher & Fuller’s because Barry Woods arranged for a free copy to be sent to me. I borrowed Lawson’s from the Met Office library.

            But like I say, you should read or ignore whatever you want!

            I do find your argument that “most scientists don’t have a blog” rather odd. Most scientists don’t write books or present popular science shows on TV, but that doesn’t mean that those few who do are wasting their time.

      • Faustino

        “is all about disliking the implied need for change” Au contraire – change is inevitable, constant, and something humans have handled well for millennia. Changes in policy and behaviour, which you appear to allude to, are best done with good understanding of the likely costs and benefits and the opportunity cost.

  49. Dusty

    Hi Tamsin,

    I’ll have a little say if I may, then get back in my box.

    What persuades people to take the stances you have described in your little red dots and what would persuade them to change? Setting aside Climategates I & II and the politics of climate science, I would like to think that the answer is the extent to which people believe Cause and Effect has been demonstrated.

    I have an engineering/aviation background and am firmly of the ‘sceptic’ persuasion. Originally I was ambivalent about global warming until I saw the hockeystick graph; that fired up every bell, whistle and claxon alarm I’m fitted with. So I started reading and it rapidly became clear to me that none of the evidence on offer was derived from the tried and tested ‘scientific method’. The hypothesis that the CO2 output by mankind is causing the planet to enter a catastrophic warming mode has not, as far as I’m aware, been demonstrated by directly measured, empirical evidence. More problematic, it seems to me, is that Cause and Effect for this particular hypothesis probably cannot be shown by direct measurement.

    However, if Cause and Effect cannot be shown directly then an experimenter might consider decomposing the problem into logically related segments that run from cause to effect with specific ‘measures of effectiveness’ (MoE) or some such term. The segments to be tested and the MoEs to be measured would have to be agreed by all concerned beforehand and be based on a coherent, multidisciplinary plan. Trials success and failure need to be defined beforehand. If needed models and their specific use, design, and verification and validation plans need to be defined. Model limitations must be identified and agreed. This process is already used when necessary, to measure the ability of complex systems to meet their overall requirements in trials. Success in each of the segment trials then infers empirical evidence of (Initial) Cause and (Final) Effect, albeit indirectly. I have not read a single paper where this, or any similar, approach has been taken, advocated or has even been discussed.

    Melting glaciers, unvalidated model outputs, and hard to find polar bears do not provide rigorous scientific evidence of the Cause and Effect required by scientific standards. Currently the observations made in pursuit of the Thermogeddon hypothesis seem to be collected by different universities in an uncoordinated, scattergun approach and the observations that are made simply used to generate emotive statements and doom-laden graphs to win over hearts and minds. “It’s science, Jim, but not as we know it.” as Spock might have said.

    Until such time as the experimenters demonstrate the answer to the exam question to be: Cause (man-made CO2) and Effect (thermogeddon) based on logical, coherent, objective trials with transparent plans and data then I’m afraid that I will remain firmly in my little red ‘sceptic’ dot on your graph.

    I’m back in my box now.


  50. Tamsin Edwards

    BBD, thanks for self-organising. I haven’t responded or chided today as often I could have.

    Please, remain friendly and on-topic. Once we start on the science it will help keep focus, I think.

  51. lolwot

    Given the radiative forcing from a doubling of CO2 is so massive compared to anything else known, and given that human increase in CO2 level is on a scale unknown in Earth’s history am firmly in the convinced section just on that.

    To show that it’s safe I require at the very least the majority of models to be showing low climate sensitivity. I also require more though, like studies on ocean acidification to be yielding widely positive results. The science currently is completely the opposite from what it would take to convince me that human CO2 emissions at the projected rates are safe.

  52. geronimo

    Tamsin: “Once we start on the science it will help keep focus, I think.”
    You have a pretty civilised bunch on here, a lot from the Bishop Hill blog, and I do hope you can keep it on the science once you’ve started. I am in the unconvinced camp, there are lots of things I don’t know about the science, but the reasons I’m unconvinced are simple, the AGW meme has been an opportunity for the the environmental movement to foist their views about how we should live our lives on the population at large. Don’t misunderstand me, many of their views are admirable and I’m all for them, just as I find many of the views of Christianity, Islam and Judaism admirable, but then I look at the results of their views and how they disregard human health and happiness for the sake of the environment and I shy away from them as religious fanatics.

    Still no reason to deny the science you may say, well I fall into the not so numerate category of engineer with a life’s experience of scares and scarers so when someone tells me that the most of the rise in temperature in the 20th century is very likely (90-100)% due to human emissions I look for rock solid evidence, there is none. For this statement to be true then this one must be:

    It is very likely that we know of and understand forcings and the interactions between them in our climate system.

    Would you sign up to that?

    Finally my scepticism about our ability to do anything about it. As the IPCC reports, and indeed the Met Office reports, tend to gloss over timescales I’m not sure whether adaptation is even necessary as a conscious activity, evolution might be the best course of action.

    What I do know with a confidence greater than “very likely” is that the plan to reduce CO2 emissions so we can get back to 350ppm of CO2 is doomed to failure and cannot get my head round why anyone in their right mind would propose such a solution given the global political and economic .environment of the world’s most populace countries springing out of poverty.

    • Anteros

      geronimo –

      I’m not sure whether adaptation is even necessary as a conscious activity

      Sometimes I’m surprised that this excellent observation isn’t more widely made. Just consider the adaptation that we have undergone since our grandparents were our age. Would we have made a better job of it if we’d tried to do it ‘consciously?’ Methinks not. Attempts at adaptation end up with feeding corn oil to motor cars. Especially as it’s exceptionally hard to identify anything that currently needs to be adapted to – let alone consciously.

      • lolwot

        “Would we have made a better job of it if we’d tried to do it ‘consciously?’ Methinks not.”

        Which is why mitigation is a superior solution than adaptation

      • Anteros

        lolwot –

        Not true [And cannot logically follow from my comment]
        What is wrong with natural adaptation?

  53. geronimo

    Sorry:”It is very likely that we know of and understand forcings and the interactions between them in our climate system.”

    Should read:

    It is very likely that we know of and understand all of the forcings and the interactions between them in our climate system.

    Good luck with keeping it focused on the science.

  54. lolwot

    The compass reflects the popular but ill-founded thinking on the climate change issue that the burden of proof is to show “intense and rapid impacts”. The IPCC do this, the skeptics do it, everyone does.

    The climate issue is framed as some kind of single “scientific theory of man-made dangerous climate change” so people demand and expect evidence for that theory and assume a “null hypothesis” of “no impact”. That would be fine if we were starting from a blank slate of knowledge, but we are not. Important underlying facts are known. Among other things we know the CO2 rise is human caused and we know that the projected rate of CO2 increase will produce both a significant radiative forcing and impact on ocean pH. Because of this the burden of proof should be altered to proving the change will cause “no impact”.

    I can’t think of any analogous situation in society where such a significant impacting factor gets given the “prove it’s dangerous” demand rather than a demand to “prove it’s safe” .

    For example drugs companies design a drug for a specific purpose such that when a new drug is produced everyone knows is it something that has an inherent strong influence. It’s not simply a new type of bean. It’s a powerful drug. Much like we know the ongoing CO2 rise has a strong radiative influence. But here’s the difference with the framing: we are expected to believe CO2 rise is safe unless and until “those scientists” can prove it’s dangerous. On the other-hand we are expected to believe a new drug is *dangerous* unless or until scientists prove it’s safe.

    Take another example: geo-engineering. People recognize these schemes to seed the oceans with iron, or pump aerosols into the stratosphere are fundamentally dangerous and people demand scientists prove them safe first. I even see “climate skeptics” make such demands and playing up the fears of something going wrong. Funny that, because on the subject of the actual great CO2 experiment everyone takes the rather opposite view that it’s fine unless someone can prove it isn’t.

    The IPCC report should be framed as a safety test. It should systematically try to prove that the ongoing CO2 rise is safe. What happens then is that uncertainty in the science correctly reflects danger. That is where uncertainty *should* lie. It’s balls backwards at the moment with the “ignorance is bliss” setup.

    In the areas where there is too much uncertainty the report should conclude: Fail. In the areas where there are doubts about safety (eg if a number of GCMs are showing large impacts) the report should conclude: Fail.

    • Maurizio Morabito (omnologos)

      lolwot – your analogy is flawed. A diffuse interaction with the environment (such as the emission of CO2) cannot be handled the same way as a discrete and specific one (such as the use of drugs, or geoengineering).

      Humanity might be falling down a skyscraper still believing everything is ok when passing by the fifth floor because nothing has happened yet…however it is up to the “convinced” to show the risk ahead, if there is any: because there are many ways in which we interact with the environment in a diffuse manner, and if there’s anything special about CO2 it has to be shown as such.

      • MikeO

        I suspect there is little we can do avert the supposed threat. Let us say we actually could pull back CO2 “safe” levels. Seems to me that the outcome of any such action is as little understood and we might regret doing it. I would ask lolwot though what drove Climate Change before humans were on Earth? Or wasn’t there any? I thought this blog was about Computer Models not AGW!

      • BBD


        lolwot – your analogy is flawed. A diffuse interaction with the environment (such as the emission of CO2) cannot be handled the same way as a discrete and specific one (such as the use of drugs, or geoengineering).

        You’ve lost me there. Why is geoengineering – say stratospheric aerosol injection – not “a diffuse interaction with the environment” in exactly the same way as increasing RF from CO2? After all, stratospheric sulphate aerosols modulate RF, just like GHGs, only with the opposite sign.

        I only ask because if you’ve got this wrong then your objection falls to pieces.

        if there’s anything special about CO2 it has to be shown as such.

        Although given that you say this, it rather has, really.

  55. Vaughan Pratt

    I’m with John Russell: far north-east on your compass (though unlike him I’m not troubled by the skeptical-critical distinction since I haven’t found unskeptical criticism terribly effective — you have to become a skeptic to be an effective critic).

    For the past two years I’ve been staring with puzzlement at the smooth rise of CO2 and the very wobbly rise in temperature. If CO2 were a big factor in raising the temperature, shouldn’t the latter be rising more smoothly? It took me a lot of “staring” at the data before I could understand how CO2 could be “the big control knob” as Richard Alley puts it despite the big wobbles.

    On the point about the last decade being flat, I suspect the oceans have a lot to do with this. If you look at Richard Muller’s BEST data for the last decade, which is only for land, it’s obviously extremely flat.

    But some warmistas persist in grumbling that “flat” is relative, and that judging the decade in isolation tells us nothing. They say you should back away a little so you can see just how much flatter the past decade has been than say the previous three decades. Or at least the past 17 years if you go with the Santer minimum.

    Oh come on now, how picky can you get? The past decade is obviously completely flat, no need to compare it with anything.

    But ok, let’s humor these warmistas anyway by backing away and looking at the last four decades of the same dataset. That’s a lot more datapoints so let’s even up the comparison a bit by using annual rather than monthly data. One then gets this plot.

    Wait, what? That’s just steadily climbing. Can’t be, it was completely flat when we looked at the last decade in isolation. What in the world is going on?

    Talking about the world, Thomas L. Friedman claims it’s gotten flat in the last decade. That may be, but I’m not sure I believe the decade itself is flat any more now that I see its temperature in context.

    Where do the oceans enter into this? Simply that this very interesting effect isn’t as pronounced with the combined land-sea data, from any source. There may well be cooling this decade, but if so then it seems to be coming more from the oceans than the land.

    On Roger Pielke’s focus on land use, the authority I’ve been going with for that is the Carbon Dioxide Information Analysis Center’s estimates of how much CO2 is coming from the various sources including fossil fuel and changes in land use. Land use change is indeed a substantial contributor to the anthropogenic component of atmospheric CO2. However when you use the CDIAC’s data to add up all the contributors, the result does not in any way undermine the seriousness of the contribution of fossil fuels. Drawing attention away from the latter by focusing on land use strikes me as a tad disingenuous if that’s not a banned d-word for this blog.

    So I’m up there in Maine on your map. Or Newfoundland if we take the whole Western Hemisphere.

    One thing I’m not clear on is where on your map you are putting those who think this AGW thing is a crisis. Like the flat decade, one needs to look at this in context. No one is debating whether there’s an economic crisis, from which I infer that it’s currently a bigger crisis than global warming, which is heavily debated. CO2 is pushing us into increasing disequilibrium, causing more intense storms, but I have no guesses whatsoever as to whether that effect will overtake the economic crisis in its impact on our well-being before we succeed in weaning ourselves off carbon based fuels. I do expect that by 2020 we’ll have a lot better idea about that.

    For the longer term (crisis or not) I would put my money on inertial confinement fusion, which seems to me a lot more workable than magnetic confinement (tokamak).

      • Vaughan Pratt

        My apologies if so, but I did read policies (a)-(h) (both before and after your warning 🙂 ) and didn’t see one that would disallow reasons why the upper right of your sceptical compass should not be pulled in further than the upper left. Which one did you have in mind?

        True to form I may not have separated the wood from the trees sufficiently clearly in my argument, which at root (or canopy) was that there are excellent reasons to have a red dot well beyond your most north-easterly one. Your distribution of dots suggests that while the faithful are entitled to the conviction of their faith, a convinced skeptic is an oxymoron. I would call the sentiment that the logical are not entitled to the conviction of their logic a hallmark of amateur science. (Disclaimer: though my training was in science my career has been in logic.)

        The climate debate is illogical enough as it is, but I would consider ruling as off-topic the reasoning behind issues with your compass as antilogical. (Actually I’m not sure which is worse. 🙂 )

        BBD’s reply to my comment nicely illustrates an elementary statistical fallacy commonly encountered on both sides of the climate debate. However I’ll respect your judgment as to whether it’s off-topic– your blog, your call. 🙂

        • BBD


          BBD’s reply to my comment nicely illustrates an elementary statistical fallacy commonly encountered on both sides of the climate debate. However I’ll respect your judgment as to whether it’s off-topic– your blog, your call. 🙂

          If Tamsin objects, then let the fault be mine, not yours. Your example was problematic. I simply pointed this out. Elementary statistical fallacies have nothing to do with it.

          • Vaughan Pratt

            In the interests of staying on topic for at least part of this reply, let me start out by restating my position in Bayesian language (assuming we’re looking at this the same way). My position is that the prior could be just as narrow at the right as at the left when the latter is narrow (the convinced sceptic), and moreover just as wide at the left as at the right when the latter is wide (the uncritical believer constantly beset by doubts).

            A bimodal posterior at the left (the dots above and below) would be normal for controversial topics such as whether second-hand cigarette smoke is hazardous, vaccination causes autism, nanotechnology is a threat, etc., just as it is normal for religious faiths to develop denominations with different attitudes and beliefs. A theory of how the uncritical mind arrives at a position about a topic tells us little about the topic, whether it be God or the climate.

            If present at the right however it becomes a more topic-relevant question as to how the critical mind moves vertically in the sceptical compass. For example if the professionals in the field were evenly divided above and below the line one would be inclined to view this as a crisis of some kind in the subject itself. But if the professionals were almost all above the line, one might call it more of a public relations crisis.

            I should pause there in case I’m wildly off track, unclear, or whatever.

            Regarding BBD’s claim,

            Your example was problematic.

            I have to protest that this is a straw man, that is, you are putting words in my mouth. You brought up the two data points at the right, I was mentally blocking them out since they were evident outliers and the graph was obviously flat even without them.

            Fitting a trend line doesn’t change this because the trend line you showed for 2000-2010.2 has an R2 of 0.0548, backing up numerically the evident flatness of the curve.

            Moreover even that R2 depends heavily on the presence of the two years at the left, as this graph shows. It too trends up, but with an R2 of 0.0015, meaning that 99.85% of the variance in the BEST data from 2002 to 2010 is left unexplained by the trend line. Whether judged by eye or by statistics, that decade is to all intents and purposes flat.

            In striking contrast the evident pattern in this graph for 1994 to 2010 does not look at all flat. This appearance is borne out by analysis. There is an evident trend that can be estimated at around .34 °C/decade, on which is superimposed an evident cycle of period around 4 years and amplitude .14 °C, with a phase that puts its middle trough at around mid-2000.

            When one takes this as the model of what’s going on there, with the assumption that the cycle is sinusoidal, the R2 is a much healthier 77.3% leaving only 22.7% of unexplained variance

            While humans may not be able to estimate R2 to two decimal places, they seem pretty reliable at distinguishing between low and high R2 with nothing more than a quick glance. The only ones who are fooled are those who put their faith in trend lines without taking the R2 into account.

    • BBD


      You have used the BEST data without removing the problematic April and May 2010 values. The graph should look like this and does in fact show a warming trend of 0.26C/decade.

      Tamsin – apologies for OT, but I really could not let this pass unremarked.

  56. MikeO

    Okay I have a bachelor of computer science and 30 years of expertise. My view is that where I sit in the compass is of little importance. It reminds me of the useless Myers Briggs test. Tamsin you wish to convince us of the worth of Climate Models? Please look at this link the problems given are very similar to what is presented in a climate models. You really need to understand and prove why any model is different from what Freedman writes about. Such models are not wrong their output is worthless other than for research. There are of course computer models that are used in say engineering but they are testable! A final point is that you must convince us and your colleagues that computer models are correct otherwise you are out of a job!

    • Tamsin Edwards

      Hi MikeO,

      That article is a great starting point for a post about the differences between earth system and financial models, and about how climate scientists have moved away from single calibrated values of parameters to multiple (ensembles) to explore the uncertainties described.

      The vast majority of climate modellers are physicists and mathematicians. I think we’d be ok, careerwise.

      • John Costigane

        I hope there is no further need for Tamsin to comment on commenters. Convinced and Unconvinced alike should refrain from the usual back and forth and allow science to be discussed dispassionately.

        • Tamsin Edwards

          Thanks John.

          With 323 comments (including my own) & 3700 unique visitors in my first 5 days blogging – a big thank you to you all for coming, and sorry I can’t reply to each individual comment or moderate as efficiently as I’d like.

          I hope to post something on science within a few days.

      • MikeO

        Your blog says all models are you really focusing on climate models? I fear you are going to jump into the middle of the science and work from there. As yet I have not found a clear explanation of how any particular climate model was produced. Clearly at some time there has been an hypothesis that enough is known to create a virtual reality in a computer program of the Earth’s climate. On the WUWT site a list has been compiled of those factors affecting climate So first it must be believed that using such a list a worthwhile model can be produced. Starting with that a design specification must be drawn up that explains how it is to be done. From there hardware and software environment must be chosen. Then the development organisation how many people what is the required expertise and how is it to be tested. To me that is just a glimmer of what should be required. I suggest you look at the production of large software systems to understand. The space shuttle software would be a good start, it was very successfull took 400 people 5 years. Surely to model the earths climate much more is needed! Publication of process not science would be a good start. I have seen no evidence of such until I do I will assume climate models are produced by a relative few computer Cowboys using Fortran on super computers and arguing the detail is useless.

  57. Jeff Q

    Hi all,
    Bearing in mind Tamsin’s quest to bring us some clarity on climate modelling and my previous confession that as someone with a scientific background (Physics) the issues around the language and unclear methodologies being used in some climate modelling had led me to a vaguely sceptical point of view.

    I’ll absolutely understand if this comment is moderated out (or even perhaps used as a basis for a post) but, I’ve just finished “The Hockey Stick Illusion”, and, in the spirit of scientific enquiry, I’d like someone to walk me through the opposite view. As it stands at the moment, and with my admitted lack of deep understanding, some of the things described in there in terms of what McIntrye alleges are just, palpably, wrong from a scientific point of view.
    I’m well aware that there is a “consensus” view that McIntyre is wrong but, last time I checked, there is a consensus that man has never walked on the moon (if you pick the right group to ask).
    So, as I said, happy for this to be dragged into a corner and ignored if it will just cause trouble, but if not I’d welcome some more light bed time reading on the matter!

    • Vaughan Pratt

      in the spirit of scientific enquiry, I’d like someone to walk me through the opposite view.

      My sympathies: having read only (Montford’s account of) McIntyre’s view you’re now stuck with the additional task of reading accounts of the opposite view.

      Had you read a random sample of the 42 favorable Amazon reviews of the book first (out of a total of 47) you would not have had this problem. You’d learn that this has been “the biggest scientific fraud since Piltdown Man” and that “as a result of the duplicity of the Hockey Team and the IPCC, billions have been spent and we have no idea whether or how human beings may affect climate change. We are at square one.” After a while all the reviews blur together.

      The opposite view when expressed in the same style might be summarized as “the biggest hatchet job since the Salem witch trials.” (Substitute your favourite scapegoat for the alleged witches.) So where to find support for this view?

      Speaking for myself I found it in the reviews themselves, and felt no need to read further. It seemed to me that they weren’t so much reviewing the book as passing sentence on the work of Mann and the IPCC, while asserting that climate science is a giant conspiracy taking its cues from a newly minted Ph.D. like Mike Mann. One gathers that had Mann studied medieval art instead of medieval climate the case for AGW would be weak to nonexistent.

      With favorable reviews like those, why read the unfavorable ones?

      • Jeff Q

        Thanks for the reply Vaughan, and I’m well aware that what I’ve read is, undeniably, one side of the story.
        What bothers me isn’t the true believers stuff on either side, it’s my (admittedly theoretical / ivory towered) view on how science works. In particular, and I’m happy to be shown how wrong I am, that scientists publish papers that promote a hypothesis, in support of this they provide their data and analysis steps and basically say “prove me wrong” to the world.
        Like it or not, that’s not what appeared to happen in relation to Michael Mann’s initial nor subsequent papers.
        I did enough physics to know that scientists live and die by their published papers (a friend of mine has 125 abstracts / cites in a “hard” science) and, as I mentioned in my very first comment on this blog, the lack of such engagement is what has pushed me across the sceptic axis.
        If you have anything that takes Steve McIntyres work and shows how his analysis is wrong (and by extension Michael Mann’s correct) then I’m all ears.
        I’m a scientist, convince me don’t just tell me I’m stupid or wrong 😀

        • Vaughan Pratt

          scientists publish papers that promote a hypothesis, in support of this they provide their data and analysis steps and basically say “prove me wrong” to the world

          I fully agree, Jeff, with the addition of two words: “of science.” As long as those two words are respected, that sort of engagement is normally productive. However the years Linus Pauling spent ridiculing Dan Shechtman is just one of many counterexamples to that Garden of Eden model of how science works.

          When you substitute businessman Steve McIntyre for scientist Linus Pauling, and increase the level of harassment by two orders of magnitude, you’re no longer in the world of scientific challenge but of what the science side of the debate considers political skullduggery. For some reason this seems to get the backs up of the scientists being harassed. If you’d asked me to predict how they’d react to McIntyre I wouldn’t have known beforehand, but I do now.

          Clearly McIntyre needs to be replaced by a serious academic with outstanding credentials in statistics. Enter very distinguished statistician Edward Wegman, with some 30 years of key administrative positions in statistics: director of ONR’s statistics and probability program 1978-83 and head of ONR’s math sci division 1982-86, then director of the center for computational statistics at GMU 1986-2006, founding chair of GMU’s dept. of applied and engineering statistics 1992-1999, and chair of the data sciences program 2004-2006.

          We now have a more comparable situation, with Mann and Wegman in place of Shechtman and Pauling.

          Except in tone, which has replaced (somewhat) collegial challenges with what the Mann side insists is very hostile politics.

          If you have to ask how politics could possibly have found its way into climate science, I’d say your “I’m a scientist, convince me don’t just tell me I’m stupid or wrong” is deeply disingenuous. Those scientists feel like they’ve been thrust into a war, with the Attorney General of Virginia trying to get Mann thrown in jail, and you’re depicting it as a collegial challenge by fellow scientists. That’s ridiculous. For one thing it’s not as though half the scientific community is under attack by the other half, rather the whole of the scientific community is under attack by a very tiny set of scientists made to look much larger by being backed up by half the non-scientific world, for whom global warming is a bitter pill.

          So how was the Wegman report received? That depends very strongly on who you ask. According to this section of the Wikipedia article on Wegman there is an ongoing investigation into Wegman’s scientific conduct in his report. For the polar opposite you’d be hard pressed to find a skeptic blog that doesn’t believe Wikipedia is under the control of a global conspiracy.

          If there’s a middle ground, a DMZ, it’s about the size of a postage stamp. Basically you get to decide whether Wikipedia or Wegman is lying.

          Is there anyone of Wegman’s stature among academic statisticians willing to back him up? Good question, but as this is getting a bit long let me resist the temptation to address this important question and post this now. In the meantime you might consider doing a Google search to identify such a statistician. I’d be very interested in what you’re able to turn up. Even one strong candidate would help McIntyre’s case a lot. Mann has no shortage of very distinguished scientists on his side.

  58. Michel Crucifix

    Hi all (and greeting Richard B., your post made my day!)

    In the previous post, Doug Cotton challenged me with the following argument:
    ” How and why should one scientist have “authority” over another with an opposing view?
    What matters is which one is applying correct physics, mathematics or whatever, Nothing else matters.”

    This is an important question and source of misunderstanding, so here is my answer:

    It would be just so simple if we had to apply ‘laws’ from a cooking receipt book to solve puzzles…. and scientists would be jobless as computers would do the job. Science does not work this way. Perhaps as a caricature, even solving the movement of a mass attached to a spring requires some judgement, about which approximations to make, which paradigm to use (Newton’s in this case), how to make measurements etc. Not to speak about a complex system, where different paradigms and modelling strategies coexist.
    This is with experience and confrontation of his / her ideas with observations and other scientists that a scientist gradually gains authority. For example, when I read “it is incorrect physics to apply Stefan-Boltzmann calculations to the surface as is done in the development of the AGW hypothesis. ” I am left with the feeling that the person writing this mis appreciates the role that is actually given the the Stefan – Boltzman law in my and colleagues’ conclusions about climate change. I can’t blaim the person who wrote this: developing climate models, thinking about the role they should take in the advance of knowledge about its dynamics, accommodating available information available at a whole range of time scale takes time but has a value that goes beyond what can be written in a single text book. Climate scientists claim authority in the same way that an eminent physician claims authority : this authority surely finds its source in having studied theory, but above all it is justified by having been confronted with the practice for many years. Physicians, just as climate scientists, run sometimes the risk of being locked into a dead end paradigm for a while and err in their judgement, but the constant confrontation to the facts, the demand of the public, their own retrospective thinking and the diversity of the community alleviates this risk.

  59. Jeff Norman

    Enough about us. I would like to read about you, your work and your modelling experiences please.

  60. Theo Goodwin

    I propose that we allow Tamsin to be Tamsin. She is willing to share her vision of modeling from her research. I am jumping at that offer. The debates on modeling that are found on the internet have become stale or maybe stagnant. Even my brilliant criticisms of the uses of models by climate scientists have become rather stale. We need a new beginning. Tamsin is it.

    • Tamsin Edwards

      Jeff and Norman – thank you very much! *grins*

      I’m also very pleased that my colleague Michel is contributing here. And welcome to other newcomers today.

  61. Di Verse

    The astounding quality of comments on this new blog (and their quantity) is intimidating. Do I dare to make a fool of myself in public? Perhaps, just this once …

    I think the differences have all come about because people in all walks of professional life are guided by common sense. Very competent engineers I have known would say something like: “If it looks right, it is right”, which of course comes from experience with comparable projects. Now, common sense is not always a reliable guide. Modern science has delved into many dark corners where what has been found appears to be explicable only in highly counter-intuitive ways, witness quantum mechanics.

    However, there is no argument, even when a theory which almost nobody really understands, that is, one that cannot be explained in terms of familiar day-to-day experience, nonetheless allows calculation and experiment to coincide reliably and accurately. The proof of the pudding is in the eating.

    Many, many aspects of climate science seem very far fetched indeed and the theories most decidedly do not lead to numerically accurate reproduction of climate events, whatever the practitioners try to tell us. The pudding has rather an unpleasant taste about it.

    Climatologists also seem always to hold the view that only approved climatologists may legitimately comment on climate matters. If you do not belong to the inner circle, no matter who you are or what you know, your scepticism is dismissed out of hand. In other sciences scepticism is welcomed and treated seriously.

    Climate science is a human endeavour. It is not a private hobby. The protective attitude of a large number of prominent climate scientists and fellow travellers is utterly ridiculous and strengthens the suspicion that not all is as it should be.

    I believe that absolutely anyone who is sufficiently interested is perfectly capable of understanding all aspects of climate theories. Moreover, if such theories cannot be explained in a reasonably understandable fashion, it must mean that the climatologist does not really understand them either. I am sure this is the case at present.

    Perhaps I am showing my ignorance, but I feel that a meteorologist in former times would not need to have the theories at his finger tips. He probably worked much more from pragmatic experience than numerical analysis. The modern approach takes forecasting to a wholly different level of ambition and the necessary set of equations which have tentatively put in place for solution are not really up to the job, or so it seems. The method developers perhaps do not need the same grounding, as long as the equations are right, but has a lot of useful experience been sacrificed? Are advances in reliability and accuracy really being made?

    Whatever the subject, putting together the observations, the theories and the successful calculations implementing the theories, most of us are able to understand what is going on. Climate science is alone in not permitting this process.

    That is why it suffers so much sceptical opposition – and rightly so.

    • Michel Crucifix

      Dear Di Verse. You write : “I believe that absolutely anyone who is sufficiently interested is perfectly capable of understanding all aspects of climate theories. ”

      Would you say : “I believe that absolutely anyone who is sufficiently interested is perfectly capable of understanding all aspects of medicine. ” ?

      Do not take me wrong. Being an expert does not entitle to the right of not being contradicted, challenged or asked for explanation. Experience simply tells us a involving an expert in a decision increases your chance of making a successful decision…. unless the expert has a hidden agenda against you. This is this latest reason that sometimes makes us a bit wary when the garage owner tell us our car needs expensive repairing. As regards climate scientists, I would argue you are on the safe side because at least my judgement is that there is more immediate benefit for an individual in claiming there is no AGW than in claiming there is, put aside considerations of morality.

      • Jeremy Harvey


        It is nice to read your contribution on this nice new blog (I wish you well with it, btw, Tamsin) especially as I was educated at UCL (Chemistry & Philosophy – did you too study Philosophy – I don’t think we ever met?). Expertise is a difficult area. You make an excellent point that all things being equal, one should defer to experts. I nevertheless place myself squarely on the ‘lukewarmer’ point on Tamsin’s graph. Why? Well, experts are not always right. A dishonest person in a garage can of course consciously invent a case for doing work that does not need to be done. In keeping with Tamsin’s comment policy, but also with my own beliefs, I do not think that any significant number of scientists behave like this in the climate field – I would even grant that there might be (slightly) more on the sceptic side. That being said, subconscious motivations, beliefs and inauthenticity offer a rich variety of ways in which scientific claims can be swayed in a field with significant uncertainty.

    • Steve Bloom

      Or try quantum physics. IIRC Feynman once said something to the effect that if you think you understand it, you’re wrong. And yet quantum physics is very, very useful.

    • Steve Bloom

      “Are advances in (weather forecasting) reliability and accuracy really being made?”

      Yes, thoroughly documented and vast, such that (IIRC) a four day forecast today is about as reliable as a one day forecast 40 years ago. And yes, although theory and observations have of course advanced greatly, it’s because of the models.

      I expect the Wikipedia article on weather forecasting would be thoroughly informative on this subject.

      • MikeO

        I did an experiment I recorded the predicted maximum temperature for next day from th Bureau Of Metrology in Australia. At the end of each I also recorded what it was from their AWS. This was just one at Canberra airport. I did it for 3 months. I thought it should be close but was astounded just how much error there was. Up to 5 degrees C! If the weather is calm okay but as to when a front would arrive and what temperature change they have little idea. Broadly their prediction is seeing a front from eye in the sky and knowing it will be at point x with a margin of error in days and up to 5 degrees.

  62. WebHubTelescope

    This is in regards to the naming of your blog.

    A pet peeve of mine is the use of quotes without having the right context in place. George Box’s quote of “All models are wrong, but some are useful” is a good case in point. The original context is from the book “Empirical Model-Building and Response Surfaces” by Box and Draper:

    “The fact that the polynomial is an approximation does not necessarily detract from its usefulness because all models are approximations. Essentially, all models are wrong but some are useful. However, the approximate nature of the model must always be borne in mind.”
    So the context is to be careful in numerical computations so that the model numbers match the observations.

    On the rest of the page, Box and Draper present a concise description of the differences between epistemic and aleatory uncertainty, which is really the scope of Edwards (and Judith Curry’s) efforts. Epistemic uncertainties are the systematic errors that one can introduce in a statistical model while aleatory errors are those that are fundamental in the natural behavior itself, be it noise or some other random effect.

    So Box is essentially describing why we should be careful in numerical errors in statistical modeling, while everyone else has re-interpreted the quote to question the validity of using models in the first place. That is just plain ridiculous — all one has to consider is that all of mathematical engineering is based on models, and look at how far that has gotten us!

    That’s what happens when people take the quote and never read the original context.

    • Theo Goodwin

      “That is just plain ridiculous — all one has to consider is that all of mathematical engineering is based on models, and look at how far that has gotten us!”

      Models are excellent analytic tools for engineers. But models cannot substitute for well confirmed physical theory. Climate scientists claim that models can substitute for well confirmed physical theory.

        • Theo Goodwin

          The purpose of models is to reproduce reality. The purpose of physical theory is to describe the natural regularities that make up reality. The big difference is between reproduction and description.

          Models produce simulations. See that word ‘simulation’. Models simulate reality. They are to be judged on the degree of completeness in the simulation and on other factors. Simulations are neither true nor false.

          Physical theories, when combined with statements of initial conditions, imply observation statements that describe reality. When these implied statements are true we say that the physical theory has produced true predictions. The physical hypotheses that describe natural regularities offer explanations of true predictions and the events that they describe. Because physical theories imply observation statements that are true or false then the theories can be true or false.

          Some people claim that models can be used for prediction. That is obviously false. If I had a perfect simulation of Earth I would have just another Earth. How does an additional Earth enable prediction that the original Earth did not? What these people are thinking about is all the rejiggering that they do to their models. Anything that might count as prediction is found in that rejiggering and not in the simulation that it yields.

          • Michel Crucifix

            Interesting post, thank you. I might oppose the first statement, stating that the purpose of simulators is to produce a description of a phenomenon that can be linked to the real phenomenon. But this semantic subtlety is not crucial to your argument I believe. More interesting is the last paragraph.

            What you call ‘rejiggering’ is what philosopher E. Winsberg would call ‘ad hoc hypotheses’. They are unavoidable in a simulator of a complex system, and they rightly cause concern about model predictions. This is the reason the climate scientist community has embarked on two approaches : one is to invest in reducing these ad hoc hypotheses as much as possible and improve the statistical description of turbulent phenomena with appropriate stochastic formulations, which really requires massive, military standard, computer resources for a limited number of experiments; the other approach is to invest in the statistical description of the discrepancy between the simulation and reality. Then of course you need to model how this discrepancy behaves with climate change, and this is one of the reason d’etre of palaeoclimate modelling. In practice, this second approach requires numerous experiments, with different simulators and different parameter combinations, in order to identify those features which are independent on the ad hoc hypotheses. The epistemology of ensemble of simulations is admittedly in its infancy, but both ‘on the field’ scientists and philosophers of science are working hard on it.

            In conclusion, your post is right to the point, but you should not be left with the impression that climate scientists (if this is those you refer to when you speak about ‘these people’) are unaware of it. Different approaches to the problem have been proposed, which is the source of a healthy and a lively debate.

            The mathematical, physical and statistical concepts behind this are quite involved. This is expected when you touch the state of the art of a discipline (same in quantum mechanics, cosmology, cell biology, etc. ) When a particle physicist speaks about the Higgs boson on TV he is not immediately contradicted. Given what the urgency and importance and what is at stake (Funtowitz and Ravetz’s post-normal science) climate scientists have a stronger burden than their colleagues when they communicate and answer the concerns of the public.

            Ruminating about model discrepancy and communicating about it to the general public are two quite different exercises. The difficulty with communicating with the general public is not that this public is ignorant or obtuse. The difficulty lays in the fact that it is diverse: some are positivist, some are religious, others are instinctive; there is no argument that will resonate to everyone at once.

            When I read ‘some people claim’ this makes me think that you are actually referring to the communication exercise. On that specific point, as a climate scientist, I am aware that I have much to learn from the response of the general public.

          • Anteros

            Michel –

            I don’t know if Theo feels you’ve addressed his concerns (which are deep), but I thought your comment was remarkable for its clarity. I particularly appreciated the reminder that climate modellers are not unaware of all the issues surrounding what models can/can’t do and the relevance of model simulations. It perhaps should be obvious, but scepticism can sometimes slide into suspicion…. Many modellers probably makes less claims for their models than we think they do. Like you say, communication is difficult but important.

            As someone focused on how our underlying beliefs affect how we see the world and interpret things we think of as ‘facts’, I noticed you saying this –

            Given the urgency and importance and what is at stake…

            I wonder – have models or modelling played a part in leading you to this conclusion? Could you identify a chain of reasoning for it?
            I have a sense that for even the most rational of us, a short period of contemplating the future generally leads to the conjuring of particular kinds of ‘pictures’. Given depth by emotion and painted by imagination, it is profoundly easy for us to convince ourselves that we have reasoned our way to our projections about the future.
            As Richard Betts has said, whether we think of a future scenario as ‘dangerous’ [or ‘interesting’, ‘benign’, or ‘urgent’ ] depends on a value judgement. It has very little to do with science or evidence.

            I’d be interested in your thoughts about how you arrived at the conclusion that the situation was important and urgent.
            Obviously, a great many people think [or rather feel] the same thing, but I don’t encounter many who display such perceptive clarity.

            I do have a suspicion that my interests are slightly tangential to Tamsin’s blog, but hope that because models (and their uncertainty) are significant contributors to our beliefs about the future [which, to my mind is the bottom line] asking questions about how we arrive at our ‘pictures’ is not too OT.

          • Michel Crucifix


            Good point. I borrowed the words from Funtowitz and Ravetz, who describe post-normal science as ” one where facts are uncertain, values in dispute, stakes high and decisions urgent. ” but this does not exempt me from the duty of justifying it why it applies to climate.

            One way to answer is just by acknowledging the present situation. The very fact that head of states, parliaments, head of large corporate companies justify taking measures that potentially have an impact on the world economy by the ‘importance of climate change’ makes climate change important, and gives climate scientists a special responsibility. And it is urgent, because if indeed climate change is caused by CO2 (which I believe it is but that is not the question), the fact that CO2 emissions get CO2 to increase impressively fast by reference to its natural variability makes the issue urgent.

            Now a different question would be whether climate models have played a role in the global awareness about climate change. Indisputably they have, but we should be careful not overstate this role. When alarm bells were first ran by Calandar and then Keeling, climate models where nowhere near what they are today. And Manabe’s first 2xCO2 experiments were done with a single column model that does not consider the general circulation, so that I wouldn’t call it a climate model by modern standards. Yet these are these and other pioneering efforts that have triggered in the UK and in the US the government push to develop numerical modelling of climate, in the late 80s (the creation of the Hadley Centre was Margareth Thatcher’s initiative).

            Finally, this is off topic but important, the debate on climate change should not obscure two points

            (a) climate change is only one consequence of CO2 rise. Other ones are the acidifcation of the oceans and disruption of ecosystems by altering the relative competitiveness of different sorts of plants, of which photosynthesis is more or less limited by CO2.

            (b) CO2 rise is only one of the many problems associated with the over-use of natural resources, others being damages inflicted to the nitrogen cycles and biodiversity.

            I believe reminding this is important because even if the climate change story was completely dull, there would still be many, many good reasons to control our use of natural resources.

          • John Costigane


            Climate change has been blamed for many things but not always found guilty. “No More Snow” was one UK example, only, winters 2009/10 and 2010/11 had excess snow. I expect other things to be found not guilty as well.

            The various Jetstream positions over the UK seem the likelier cause of severe winter weather, and recently solar UV has been implicated as setting its location. Natural variability is guilty as charged.

          • Theo Goodwin


            What a wonderful response. I am really stoked for this blog. My initial response is based on the following:

            “What you call ‘rejiggering’ is what philosopher E. Winsberg would call ‘ad hoc hypotheses’. They are unavoidable in a simulator of a complex system, and they rightly cause concern about model predictions.”

            The problem is that they are using physical hypotheses as they rejigger but they are treating them as part of the model rather than treating them as physical hypotheses. This substitution of standards cannot stand. Models reproduce reality and physical hypotheses describe the natural regularities that make up reality. This point is not semantic. No amount of new epistemology will bring the twain, reproduction and description, together.

            As regards your claims about statistics, no science has consisted of data points and statistics alone and none ever will. The people who best understand the use of statistics in empirical work are the Population Geneticists. Notice the name. Their statistical work is always tied to an existing, real world population. By contrast, climate scientists are entirely unwilling to discuss their “event space” for any given claim. That is, they are quite willing to take temperature readings from Trinidad, Nome, the 19th century, the 21st century, you name it, and treat them as comparable. It never occurs to them that valid comparisons require physical hypotheses about the real world which explain why those disparate temperature readings are comparable. The Population Geneticists point to their real populations and to the growing number of physical hypotheses which describe them.

          • Theo Goodwin

            Anteros writes:

            “Michel –

            I don’t know if Theo feels you’ve addressed his concerns (which are deep), but I thought your comment was remarkable for its clarity. I particularly appreciated the reminder that climate modellers are not unaware of all the issues surrounding what models can/can’t do and the relevance of model simulations. It perhaps should be obvious, but scepticism can sometimes slide into suspicion…. Many modellers probably makes less claims for their models than we think they do. Like you say, communication is difficult but important.”

            How many times have you heard a climate scientist argue as follows: “Without our assumptions about the effects of rising concentrations of manmade CO2 we could not reproduce the temperature rise of the late 20th century so we conclude that the CO2 caused the temperature rise.”

            Is it not obvious that they are switching from talk about models to talk about physical theory, where causal connections can be expressed, and they are blithely unaware of what they are doing – or worse?

          • Sashka

            @ Theo,

            > The purpose of models is to reproduce reality. T

            Not necessarily. The purpose of many models is to reproduce certain aspects of reality., those that the modeler wants to study in detail.

            > Some people claim that models can be used for prediction.

            Would you not agree that models are used for weather prediction with some success?

          • Faustino

            Michel, your 11.35 am post gives an indication of the vast amount of work yet to be done, an indication of the uncertainties to be resolved. Yet many climate scientists have for years claimed “The science is settled, no debate is needed,” something contradicted by your post and Judith Curry’s work on uncertainty in climate science. The ill-founded claims of uncertainty have been a major driver of scepticism and suspicion.

            And I echo Anteros’ point, that “Given the urgency and importance and what is at stake” is an indication of your socio-political-economic view, not a statement re science or modelling. This suggests that you might view the science through a particular non-science prism.

            You say (4.36 pm) that “The very fact that head of states, parliaments, head of large corporate companies justify taking measures that potentially have an impact on the world economy by the ‘importance of climate change’ makes climate change important.” But it does not tell us that it is important. I have advised prime ministers of the UK and Australia and State premiers in Oz on economic policy matters, I know well the fact that they give importance to climate change is not an indication that the CAGW is valid or even worth serious attention. Stick to scientific arguments, please.

            Similarly, re “over-use of natural resources.” This is a value judgement. Is the present rate of use optimal for the well-being of humans, current and prospective? This is not a science question, although science can contribute knowledge which helps answer it. I would not disagree that not only the well-being of humans is at issue, but again, this is a human-made value judgement, rather than anything inherent in science. You are way off Tamsin’s topic here.

          • Theo Goodwin

            Sashka writes:
            “@ Theo,
            > The purpose of models is to reproduce reality. T
            Not necessarily. The purpose of many models is to reproduce certain aspects of reality., those that the modeler wants to study in detail.
            > Some people claim that models can be used for prediction.
            Would you not agree that models are used for weather prediction with some success?”

            Of course you are correct. Models are designed to reproduce the features of reality that their designers take to be the salient features.

            I call it weather forecasting and it is not quite as good as extrapolation from old graphs. Unfortunately for mankind, some people think that extrapolation from old graphs is prediction. Prediction and explanation are symmetric. So, how do old graphs or any graphs explain something?

          • steven mosher

            physical theory is nothing but a model
            simulations and physical theory are not epistemically different
            neither can be true. both are more or less useful for defined purposes.

          • Theo Goodwin

            steven mosher writes:
            “physical theory is nothing but a model
            simulations and physical theory are not epistemically different
            neither can be true. both are more or less useful for defined purposes.”

            If you explicate these claims here then I would greatly enjoy commenting upon them. If you offer no explication then why make the claims?

          • Jeff Norman

            This exchange between Michel and Anteros is exactly the kind od civil intellectual discussion I was hoping to read here. Thank you both and please continue.

          • Steven mosher

            to get you to think theo.

            take any physical ‘theory” . Instantiate it and you have a simulation.
            There is no fundamental epsitemic difference between a model and a physical theory.

            Sit down and write a simulation of a bullet’s flight. That’s easy enough. Then we can use that to actually see and test if you know what your talking about when your try to draw distinctions between physical theory and simulations. we will see if these ‘observations” of yours have any merit. that is, we will test your account empirically.

          • Anteros

            Theo/SM –

            I realise part of my confusion is that I don’t think everybody means the same things when they say ‘model’. I’d like a really basic homing in on what you guys mean – as simply and as clearly as possible.
            Are there important generalities that can be said of all kinds of models? Economic, psychlogical [I’m recalling ‘models’ of memory], demographic, mechanical, etc etc.

            Theo – do you always have the same thing in mind when you refer to a ‘model’?

            Are all climate models themselves the same kind of thing?

            Is it true that climate models can ‘say’, ‘suggest’, ‘indicate’ or ‘project’ things? I mean, specifically is it the models themselves that can do these things? Or is that a linguistic sleight of hand?

            I feel in need of a “Model 101 – what do we mean by climate model?”

            That would make much more interesting [for me] questions such as “In what way are climate models useful”. And I think I’d also be in a better position to consider Theo’s criticisms of models.

          • John Costigane


            Your bullet could be an analogy for ENSO. You have a machine gun, firing, and not firing, ( el Nino and la Nina respectively). The heat of the chamber propels El Nino on firing, further bullets are fired, until over-heating occurs. This is the la Nina phase, where bullets misfire and are ejected. This may repeat until heat subsides.

            Questions: What is the source of heat (Solar with albedo or geothermal)?
            Does the varying level of heat govern the strength of el Nino?

          • Theo Goodwin

            Steven mosher
            to get you to think theo.

            “take any physical ‘theory” . Instantiate it and you have a simulation.
            There is no fundamental epsitemic difference between a model and a physical theory.”

            I am pleased that you know the word “instantiation.” Look up its definition. You will find that it is a relational term: X instantiates Y. X and Y are different sorts of things. The objects that instantiate a physical theory make the statements of the physical theory true. The objects are the model and the physical theory is the set of statements rendered true. There can be many models for a given set of statements. That is one reason that it is so important to keep in mind that only one of those models is the real world. Accurate prediction from the physical theory identifies that model.

            “Sit down and write a simulation of a bullet’s flight. That’s easy enough. Then we can use that to actually see and test if you know what your talking about when your try to draw distinctions between physical theory and simulations. we will see if these ‘observations” of yours have any merit. that is, we will test your account empirically.”

            Galileo did the leg work on projectile motion. Newton invented the calculus and made Galileo’s work rigorous. The Laws of Projectile Motion that are taught in college physics-with-calculus are Newton’s Laws. Did you want me to write them out? The path of a projectile under the influence of gravity is a model of those laws. The Laws are statements that are true or false. The motion of the projectile is a model of those Laws and is neither true nor false.

            The problem for climate science is that they are attempting to construct the model before they have the Laws, they are failing to distinguish between the hypotheses and the model, and they are making inferences from the model rather than the hypotheses.

    • Steve Bloom

      Just to add, a better summary of that passage would be “All models are approximations, and if done correctly are useful.” That’s a little different from the direct implication of the blog title. OTOH Tamsin has a specific audience in mind, and the title as is certainly appeals to them.

      As far as I know if Peter didn’t raise this specific point, but it’s illuminating to consider his remarks in its context

      Hmm, come to think of it, this sort of out-of-context quotation was rather at the root of a certain “scandal” of a couple years ago. Now there’s a truckload of irony.

      • Michel Crucifix


        Thank you for your answer. I would like to answer to your two comments:

        (a) “But it does not tell us that it is important. I have advised prime ministers of the UK and Australia and State premiers in Oz on economic policy matters, I know well the fact that they give importance to climate change is not an indication that the CAGW is valid or even worth serious attention. Stick to scientific arguments, please.”

        one important word indeed slept out of my keyboard and this will clarify my point : The very fact that everyone is concerned about climate change makes climate change an important *topic*. Whether climate change occurs and whether it is caused by human is not the question here. As a scientist, I cannot ignore the fact that the importance of public debate exerts a pressure on my daily activity. So, if you wish, my claim is not a ‘climate science’ claim; it is a retrospective observation that I need, as a climate scientist, to conscientise, precisely to guard against undesired interferences between personal, ideologic convictions and professional practice.

        (b) you consider that my phrase ‘over-use of natural resources’ is a deviation from scientific language. That obviously depends on the sense given to over-use. One could be morale, in which case of course the statement would not be scientific, but that was not my intention. The other is the statement that the human use of natural resources exceeds the natural flows and as such disturbs very significantly the way these natural cycles have operated until the pre-industrial revolution. That is a statement that I feel I have authority to pronounce quietly because I can draw a line of scientific reasoning that leads me to that statement. Now, the point that is off-topic of Tamsin’s I accept.

        By the way, for 15 years I have worked in climate science, I never heard a colleague telling me that the science is settled or that there was no uncertainty. So it is interesting, and slightly worrying I have to say, that a prime-minister adviser has come to this conclusion.

        • Anteros

          Michel Crucifix –

          You make some interesting points. I find myself agreeing with Faustino that many of your statements express socio-economic beliefs, and an acceptance of ‘the ways things are’ that emanate from environmentalist perspectives [for instance], and not from science. I have no problem with this, but like Faustino I think it is critical to delineate that which is a product of scientific investigation, and that which is a belief that can be supported in some way by ‘facts’ which in themselves exist partially as a product of our perspective.

          I agree that climate science is urgent because some people are saying it is. Governments and others want answers. But this doesn’t necessarily mean climate itself needs our attention urgently – this requires us to make assumptions that are profoundly non scientific in nature. Especially about our values, beliefs and feelings about the future – our feelings about change and risk.

          BTW Those of us interested in climate often over-stress the ubiquity of the public interest. You say –

          The very fact that everyone is concerned about climate change makes climate change an important *topic*

          Perhaps it would be more appropriate to say ‘one person in a thousand is concerned about climate change’…. {or something} This also obscures a very fundamental point – there are many of us who are interested in climate who are not concerned about climate change at all. What we are concerned with is the fact that many other people are concerned about climate change and as we are not, we’d rather not see a concern turn into a full-blown hysteria. We see a social and psychological phenomenon more prominently than a climatic one.

          As a very pertinent example of something you say that I feel is based on a world-view very different to mine – with neither being supportable by ‘science’ or ‘scientific reasoning’ [whatever that is – I never heard of that when I studied reasoning and philosophical logic] – you talk about the ‘disruption of ecosytems’. I think I would start from a point where I would not use an expression like that. Ecosystems are changing one moment to the next. Always. Sometimes faster, sometimes slower, and they rarely have a boundary or a beginining or an edge. I think it mischarachterises them as something that can be ‘disrupted’ or ‘damaged’ or ‘hurt’. This implies they are somehow a) fixed entities or processes, and b) fragile.

          I believe neither are appropriate, so the ‘disruption of ecosystems’ to me is a very strange conception. It implies human agency is non-natural which I also find strange. I can ask, what was the Great Barrier Reef ecosystem at the time when people first started cutting down oak trees in England and sitting by fires? For a large part it was a Eucalyptus forest. Over 15,000 years ‘it’ has continuously changed. But it wasn’t ‘fixed’ in the first view, and neither is it ‘fixed’ now. Always changing, slower or faster. And I think of a Eucalyptus forest turning into a large reef as a big change, even over 15,000 years, but ‘disrupted’?

          Would you say the ecosystem on any day in those 15,000 years was disrupted?

          Each day it was (and will be) different according to the many and various influences.

          You see, here we have different views of the world of ecosystems of change, of natural cycles and natural resources. My point is to ask whether you at any point can show that one of these views has been produced by something scientific, or whether scientific results can be used/are used to confirm our biases and re-enforce our world views.

          Finally [phew..] do we construct models with utter neutrality or are our beliefs and expectations a big part of what we build – and therefore what it is possible for us to find?

          • John Costigane


            I think we can cut Michel a bit of slack in his reasoning. After all, global warming/climate change propaganda has been drummed into the young for many years and will not suddenly disappear. The problem has been a corrupting political/scientific interface which has formed the damaging group think among senior climatologists, not to mention eager policy types, and politicians.

            What is more hopeful is Tamsin’s direct approach to the sceptical crowd, avoiding previous negativity. This is a good starting point. We should keep an open mind and allow the climate scientists to present their case.

          • Michel Crucifix

            I am ready to concede that ‘disturbance’ was more appropriate than ‘disruption’. Disturbance has a more identifiable meaning. Disruption could be the phenomenological interpretation of the mathematical phenomenon of bifurcation. The lesson is that we, scientists, have had to become aware that many words that we use in our professional practice may, for the general public, cary a morale load. This causes a risk of being misunderstood in the better case, or accused of manipulation or propaganda in the worst case.

            Regarding my socio-political economical views, I have been careful so far not to unveil them outside the private sphere. I did not conclude on what ‘should’ be done or how one ‘should’ act. I do not have authority to do so.

            You claim that 1 on a thousand is concerned by climate change; John speaks about propaganda. Yes, definitely, we have left the domain of science, so I’d better drop off.

        • Anteros

          John Costigane –

          Fair point, although I did think I was being quite gentle! Not casting aspersions etc..

          I think you’re right about Tamsin’s approach – it is refreshingly open. Perhaps my one reservation is the placing in opposition of ‘scientists’ and ‘sceptics’. I struggle to understand that. I am fully 100% behind the science. And I am very, very sceptical.
          Where do I go?

          • John Costigane


            Scientists and sceptics are both sceptical: scientists need to be self-critical to improve and we have been demonized generally. Judith Curry gave us a forum to discuss climate matters, for which she has been criticised. Tamsin is giving us direct access to climate models which are the basis for projections, which will improve. She too will be criticised. Have you not noticed the negative posts? Our role is to support this move to proper science, while having the chance to give our perspectives.

    • Tamsin Edwards


      I haven’t seen that page before. He may have been talking about statistical rather than physically-based models, but I don’t think it’s unreasonable to apply it to simulators (or indeed conceptual models). For example, you could interpret the vector discrepancy as covering numerical approximations, code errors, scientific approximations, and missing processes: this would correspond to what we do indeed call [in climate] the model discrepancy, or structural uncertainty.

      My underlying reason for choosing the blog name was to highlight the importance of trying to estimate this discrepancy term (which is rather difficult for such high dimensional, correlated outputs).

      • Steve Bloom

        So let’s try again:

        “All models have discrepancies, but with proper care they can be minimized to allow useful results.”

        The flavor of that is somehow different.

          • Sashka

            When you are trying to put too much in one sentence the end product doesn’t necessarily improve.

            Your formulation implies that discrepancies always make the models less useful. Quite often the opposite is true. Think of Lorenz for example. To me, the simplest models are the most useful.

            Personally, I don’t like “proper care” in this context. To me, it implies that there is a protocol, a set of rules, like doing oil change quarterly, and you’ll get useful results. Maybe,sometimes. Certainly not always.

            In climate modeling people are trying to use “proper care” long and hard. Maybe you can do a post to illustrate how it helped minimize and quantify the discrepancies.

  63. Jim S

    Libertarians = Indeterminist

    Authoritarians = Determinist

    This is the true divide. And where you stand on this issue determines your approach to science, economics, politics, religion, etc.

  64. AMac

    “All models have discrepancies, but with proper care they can be minimized and quantified to allow useful results.”

    An unwieldy blog title. Alas!

    • Theo Goodwin

      “All models have blemishes but ugly men with Photoshop can make them perfect.” Is the occasional attempt at humor OK?

  65. Gordon Lehman

    Models will improve by iteration. The short term forecact models seem to be peeking at their rival’s tests and gravitating to consensus. Maybe a good thing for Bulean learning, but where have I heard consensus berore?…

  66. TGSG

    Just logging on so future comments are easier. Looking forward to the discussions. Welcome back to blogland Tamsin.

  67. Pharos

    I have flagged the new paper by Abdussamatov, just published in Vol. 4, No. 1 of the APR journal (February 2012), on Bishop Hill unthreaded.
    He predicts continuation of current temps for the next two years then gradual cooling from 2014 to another Little Ice Age max at 2042 to 2055

      • Pharos

        (Tamsin- You may recall I had mentioned Abdussamatov a few weeks ago on a BH blog thread you and Richard were posting on when I was trying to make a case for doing studies on climate impacts for a cooling scenario.)

  68. steven mosher

    the fact that bloom glieck and others have any issue with tamsim Is the issue. there is a thin green line that u are not allowed to cross. books you can’t read. people u can’t speak to.

    • Theo Goodwin

      Amen, brother. If Gleick’s MacArthur runs dry he can always be employed as a Diversity Dean.

      • steven mosher

        Lest you forget my skeptical friends give me shit for associating with zeke nick and muller. weird. I cannot think of another feild where sides try to inflict speach codes.

        • Theo Goodwin

          I bless your association with zeke, nick, and muller. This blog should be a new beginning for all of us.

        • DC

          I don’t think you get shit for associating with them. I think you get shit when you quote them and only them as gospel, and leave out any relevant counters to them.

    • DC

      This is absolutely a very large problem in climate science and the application of climate science: ref. Steve Bloom etc.

  69. David Wojick

    Regarding the compass, since the horizontal axis seems to refer to something like how much study one has put into the issue, or worse, I do not see it’s utility. Perhaps it might be used in some sort of demographic study, but the data would be difficult to get. Unfortunately it lends itself to ad hominems, but it is amusing that left is bad.

    • John Costigane


      There is a togetherness (closeness) on the sceptical side which the extremes cannot match. This visual cue describes the positive body of posters Tamsin can gather. This can be the new ‘open’ consensus necessary for scientific progress. Others may reconsider their extreme stance in the light of experience.

  70. Faustino

    Re the Political Compass test, I scored 3.25, -4.15, which sounds about right. I believe that personal and economic freedom best serve human well-being; but I’m not an extremist. In terms of emissions-reduction policies, they are contrary to my view, and I tend to oppose them unless convinced of net benefits greater than from any alternative feasible action.

    But this blog is about modelling: the issue is does my view bias my assessment of scientific evidence, techniques and argument? Of course, we are all driven by our sub-conscious, but in my career as an economic policy adviser I believe that I have been able to look at issues and evidence fairly dispassionately rather than being driven by confirmation bias (though I sometimes see it at work in me on CAGW issues). I’ve been following the issues since the 1980s, and was briefed by Sir John Houghton around 1989. I’m not a modeller, but have worked on, directed and interpreted many economic models since 1966. So when this blog gets down to it’s hard-core issues, I’ll read and maybe contribute unless it is beyond my technical competence, in which case I would say, enjoyed the intro, can’t keep up with the nitty-gritty, adios.

  71. Theo Goodwin

    Michel Crucifix writes:

    “The mathematical, physical and statistical concepts behind this are quite involved. This is expected when you touch the state of the art of a discipline (same in quantum mechanics, cosmology, cell biology, etc. ) When a particle physicist speaks about the Higgs boson on TV he is not immediately contradicted. Given what the urgency and importance and what is at stake (Funtowitz and Ravetz’s post-normal science) climate scientists have a stronger burden than their colleagues when they communicate and answer the concerns of the public.”

    The post in which this quotation is found is an excellent post but I must state my strong objection to this paragraph. The key word in my explanaton of my objection is “ramified.” The theories that you list, especially the physical theory that guides the CERN scientists in their search for physical evidence of the Higgs Boson, are highly ramified. To say that they are highly ramified is to say that years of challenging study and work in the field are necessary before one could hope to contribute to some small part of the theory. Climate science is not highly ramified. The basic theory of climate science consists of Arrhenius’ physical hypotheses about CO2 in laboratory atmosphere and claims about paleoclimate reconstructions of temperature using proxies such as tree rings. This basic theory does employ statistics but nothing that would challenge the yeoman statistician. This basic theory employs black body radiation theory but it is yeoman stuff. Nothing in climate science deserves to be called ramified much less highly ramified.

    In addition to physical theory, climate science employs computer models. However, as I have explained rather carefully here, computer models and the simulations that they produce cannot substitute for physical theory and the climate scientists have been mistaken every time they have implied that the substitution is legitimate. The claim that “we know that manmade CO2 causes warming because we have to add such an effect if we are to model the late 20th century rise in temperatures” clearly states that a causal hypothesis is known to be true because of what is in a model. No one who has produced models for a living, except maybe some Dow modelers, has ever made such a claim. Let me put this in starkly simple terms. If you had a perfect model of Earth you could not infer anything from that model beyond what you could infer beforehand from Earth itself. The causal hypotheses, the well confirmed physical hypotheses, are necessarily independent of any model.

  72. Alexander Harvey

    We have an opportunity to characterise the ways and degrees to which simulators can be rendered informative as to the nature of the world in ways that perhaps the world itself may not. That is an important and significant challenge niether straightforward nor clear cut.

    If people with appropriate expertise listen to what the simulators say with sufficient patience and dilligence we may all be a little the wiser. Perhaps most importantly this process needs to inform us as to the degree of confidence that we should assign to this information.

    I like many wonder how close a representation of earthlike weather and climate the simulators produce. I suspect I may, perhaps with some effort, be able to spot simulated output, but not as easily as would have been the case 10 or 20 years ago. Even if that is so, it does not deny that the simulations may be informative. The trick is in the characterisation. Knowing in which ways and to what degree they are informative. Confidence may grow from the quantifying and understanding of limitations.

    There is a view to the effect that exposure to a the discrepancy and short-comings of the simulators would result in the public loosing trust or faith in the modelling process. There is no harm if it results in due confidence replacing trust. There is something rather patronising the view given by Peter Gleick that I find both unhelpful and franky alarming. I fail to see how we can establish confidence without transparency. I mean confidence in its neutral sense of the range between none and certainty. Lacking transparency one might be tempted to ignore the range and plump for none, and some might go for certainty.

    There is an alternative view that the simulations cannot possibly be right for various reasons. That is the essential starting point without which no progress can be made and from which quantification, understanding and information flows. My question is: are we in a mood to accept both the wrongness of the simulations and yet the possibility that they be informative? I think that is going to be a tough bargain. Stating that their being inherently wrong must bar the simulators from being informative is problematic if just that same wrongness is a core assumption of the information analysis. If we are going to see the simulators dealt with in a more transparent fashion it will be interesting to see if we are all up to that challenge for that cuts both ways.


    • Theo Goodwin

      Spot on. Simulations can be immensely valuable. But if we are to trust their users then we must have the simulations explicated for us. And those explications will be constructively criticized. Let us get on with it.

      • Alexander Harvey

        Perhaps not quite so spon on.

        The information analysis can be as detailed and transparent as one wants but the actual behaviour of the simulator may remain opaque not for want of trying but just because it is. That may be regretable and although the analysis may shed light on its workings and illustrate problems with it, the true natural of the simulator may be effectively ineffable. A simulator must be based on the physics of the problem and a description of the world it is simulating but if it is accepted that it is imperfect i.e. wrong, making it informative can proceed by comparison to the real world and by exploring its parameter space. Criticizing it is not really part of that process but a criticism or critique may be an emerging result of that process.

        I shall be interested to see how these things are dealt with here but a disinterested e.g. essentially impartial, unbiased and to that extent of meaning uncritical approach to making a simulator informative is possible and may be preferable. It is either informative or it isn’t and that can be determined by analysis in a statistical sense. Not perfect but better than not trying.

        When I mentioned that the process may be challenging this is part of what I meant. I cannot be sure how Tamsin will present the subject but I have other presentations that have included her work and others and in those cases the simulator discrepancies have to be embraced and taken into consideration but they are what they are and criticising them not particularly fruitful. Obviously knowledge of were the majority of the bodies were buried is a boon but ultimately the job in hand is to render the simulator outputs in combination with reference data informative taking account of the known and making allowance for the unknown issues. If this is how she does present it, many may find that unsatisfying, some may find it doesn’t allow sufficient scope for a good argument.

        I sha’n’t second guess her but I have taken a lead from the blog title and would suggest that arguments about simulators being wrong etc. are by the bye, that is the understood state of affairs from which you do all the clever stuff, (hint I find some of it bewilderingly complex) that combines simulated and real world data to make both more informative than they would be separately.

        To argue against this one must make the case that the simulators are uniformative, the analysis can make that case too, so it doesn’t have to be argued from knowledge of the nature of the simulator, views on its construction, or opinions on the competence and morals of its designers, builders, and operators. If it is a heap of cow poo, that is something for analysis to determine. I am informed that there has been at least one extensive and expensive suite of simulator runs junked because just such analysis detected cow poo.


        • Theo Goodwin

          “When I mentioned that the process may be challenging this is part of what I meant. I cannot be sure how Tamsin will present the subject but I have other presentations that have included her work and others and in those cases the simulator discrepancies have to be embraced and taken into consideration but they are what they are and criticising them not particularly fruitful. Obviously knowledge of were the majority of the bodies were buried is a boon but ultimately the job in hand is to render the simulator outputs in combination with reference data informative taking account of the known and making allowance for the unknown issues. If this is how she does present it, many may find that unsatisfying, some may find it doesn’t allow sufficient scope for a good argument.”

          This is a very helpful cautionary note. I did not expect that these presentations and discussions would be easy. I remain eager to let Tamsin be Tamsin.

  73. Vaughan Pratt

    My hardworking assistant Robbie, who has been looking over my shoulder as I read, has observed that at 218 contributions (original post plus 217 comments) there are five contributors to this thread with double-digit-many posts, namely Tamsin with 28, Theo Goodwin 17, Anteros 15, Steve Bloom 12, and BBD 10. Of those with 7, 6, 5, 4, 3, 2, and 1 posts there are respectively 1, 1, 2, 8, 9, 7, and 40.

    While I haven’t checked Robbie’s math personally, he tends to be pretty reliable so I’m willing to take his word for it. 😉

    The slight failure of monotonicity at 2 posts is interesting.

  74. Jeff Norman

    It might be interesting/useful/fun if each person were allowed to self declare where they believe they fit in on the Skeptical Compass by using a small representation of it as their posting icon. This might be as good as a theatrical program that helps the audience keep track of who is who.

    • tonyb

      Jeff Norman

      That is an interesting idea concerning ‘posting icons’. However, there are many people with closed minds and I can’t help feeling that a number of them who won’t read a comment if they know in advance it comes from a viewpoint they disagree with. What do others think?

      I note the coment above yours re the number of comments made by various people. That Tamsin is hogging this site isn’t she? Perhaps she ought to be given a time out 🙂

      • DC

        Note to Tamsin: I would suggest, when you get around to it, a recommend button, as I would recommend everyone consider this post to which I am replying, but to which I have nothing to add.

        [Yes, I did try to install a plug-in but it didn’t work. If any one can recommend a robust one let me know. — Tamsin ]

  75. David Young

    Tamsin, Just wondering if you had an opinion on the work of Paul Williams. I personally was struck by the issue of numerical error control in the models and its relationship to the subgrid models. In fluid dynamics, subgrid models such as turbulence models are generally calibrated, i.e., the constants determined, using analytical solutions to simple problems, e.g., a flat plate turbulent boundary layer with no pressure gradient. At least this has the virtue that numerical discretization plays no role. Other methods use tons of test data to set constants and correlations. One concern of mine is that the constants could be set in climate models based on matching test data, in which case the dependence on grid size, time step, etc. must be carefully analyzed. One way to do this is the modern finite element method and its rigorous error control via optimal control methods. Could you comment?

  76. Jay Currie

    First, welcome to the interesting world of climate blogs. You have attracted lots of attention and some solid people on both sides….keep going!

    Second, in your profile you state the following:

    “I am working on the EU Framework-7 programme ice2sea, which aims to improve projections of the contribution of ice to future sea-level rise. For this I am using the GLIMMER-CISM ice sheet model to make predictions of the contribution of Antarctica to sea level riseover the next 200 years. I am also co-ordinating a model intercomparison exercise for predictions for Antarctica and Greenland from several ice sheet models, to assess the overall uncertainty in predictions of future sea level rise.”

    Isn’t it interesting that the conclusion seems to be sea level rise no matter what and without qualification. Rather obviously there has been sea level rise – at a relatively minimal level – for hundreds if not thousands of years. But from a CAGW perspective the .5mm/y or decade is really not going to make it.

    So what is your view on this rise you are trying to model. Are poor Jim Hansen and Gavin going to have to invest in water wings or a kayak to get to work? If so, when? What’s the deal with the increase in Antarctic ice? And is sea level rising?

    As I said, welcome to the climate blogosphere.

  77. Tamsin Edwards

    A few quick responses:

    Steven, Anteros – I absolutely agree we need to clarify what we mean by the very over-burdened word “model”. Note that I am not only planning to talk about GCMs but also other types of model I encounter/use (e.g. one dimensional models of glaciers). Anyway, this is definitely the topic of my next post.

    theo – “There is no fundamental epsitemic difference between a model and a physical theory.”

    This is an interesting point. My first instinct is to agree with this: a model is (in the context of this blog) a coded representation of physical theory, with the necessary numerical and scientific approximations etc. It may also have coding errors and therefore not exactly represent the theory. A simulation is an instantiation of the theory, given certain initial and boundary conditions. But I’m open to having a debate about whether this is the case 🙂 (perhaps better after I write the post though).

    David Young – Paul Williams is a superb scientist and has done lots of very interesting work. Improving sub-grid parameterisations is one focus of his work, while propagating the uncertainty they induce is a focus of mine. One very important point that this blog will revolve around is that we have been moving steadily away from fixed parameters (i.e. the constants you mention) in GCMs to (a) sampling many different values with different simulations (my research) and (b) randomly varying the parameters within a single simulation (Paul Williams research, and what they do in weather prediction). Optimal control methods – well, this is definitely the kind of stuff I’m also planning to mention, though perhaps at first in quite general terms. It fits into the frequentist versus Bayesian stuff that is one of the arguments we have at climate science conferences 🙂

    Jay Currie – thanks for the welcome and encouragement. You raise an important point about language. Whenever I use the phrase sea level rise (SLR) I do have your concern in mind – that people might think we are assuming a particular direction (positive). But this is not the case. We just talk about negative sea level rise instead: for example, Antarctica’s negative SLR contribution due to increasing accumulation. It may seem perverse to say negative SLR rather than negative sea level change (though we do also use the latter), but I suppose the phrase has stuck and physicists are comfortable with this kind of thing – i.e. that the plus or minus sign tells you the direction. Hope that helps.

    I really want to write a proper post soon but have to get some teaching, paper writing, and plot-making out of the way first. (that’s plots as in figures… 😉 )

    • HaroldW

      @Tamsin Edwards (February 6, 2012 – 10:21 am)
      This is perhaps wandering off topic, but I’m confused about your mention of “Antarctica’s negative SLR contribution due to increasing accumulation.” I was under the impression that Antarctica was losing mass (thereby contributing to positive SLR); e.g. Velicogna (GRL,2009) gives a recent loss rate of 246 Gtonnes/yr. Is that information outdated?

    • Sashka

      (a) sampling many different values with different simulations (my research) and (b) randomly varying the parameters within a single simulation (Paul Williams research, and what they do in weather prediction).

      Interesting. Looking forward for details.

    • David Young

      Tamsid, What I mean by optimal control is the rigorous theory of the adjoint and its use to control numerical error. It also offers a much faster to do parameter studies than just re-running the model. Hope to see more on model numerics and subgrid models.

  78. Barry Woods

    surely the phrase ‘negative sea level rise’ comes from the discipline of Media managment, rather than science. It is a shame that this has been picked up by the scientists..

    ie why for the preceding few decades have we not been saying ‘positive sea level fall’

    Dare not say ‘sea level fell’. lest to quote someone ‘sceptics misuse’ it…

    or the Presidents of the Maldive , Tuvalu, etc get upset.. (or to make Nils axel Morner’s day)

    Lots of sensible theories why it fell of course, we now need to work out which ones are correct (multiple reasons also possible)

    I heard the other day that rate of warming is ‘accelerating’ some even have graphs with carefully drawn lines on it to show it..(whilst omitting other lines, that would complicate it)

    Now do I have to scratch my head, ro work out if they mean -ve or +ve ‘acceleration’.. or deceleration, or the rate of rise is dropping, or that it has actually dropped a bit..

    Science is about clarity of thought, expressing it clearly and unambiguoulsy is vital.

  79. Barry Woods

    Try saying ‘negative sea level rise’, to the public..

    Imagine a question from the audience – do you mean it actually fell ?

    The public has been sold a story (media/greenpeace version of climate science) of rising ‘dangerous’ sea level and sinking islands… Yet Even Tuvalu and Kiribai are growing,( according to the science)

    BBC: A new geological study has shown that many low-lying Pacific islands are growing, not sinking.
    (for the exact reasons Charles Darwin formulated, over 150 years ago, On the Formation of Coral Reef)

    So what will the public do, when they work out what a ‘negative sea level rise’ is, will the public shake their heads, and distrust climate scientists (or the communication of climate change science) just a little more…

    Thoughts about communicating to the public (that includes me)

    • Tamsin Edwards

      Hi Barry

      “Try saying ‘negative sea level rise’, to the public..”

      I wouldn’t in public talks – I would say sea level fall. I accept my webpage is also part of public communication, not just for academics, so I’ll change that.

      Discussing communication is on-topic. Growing islands are off-topic… cheers Barry

  80. Barry Woods

    sorry tamsin 🙁

    just thought it relevant in a comment about communicating sea level rise…
    ie the public have been communicated to, that the islands are sinking because of it, especially those islands mentioned) we would not be concerened about communicating sea level rise otherwise. (ie the public)

    Bring on your first computer model post 😉 ! of course

  81. Joshua

    tamsin –

    Quite an impressive launch. It’s interesting that your blog has attracted such interest.

    I’m a bit curious if you might speak to the kind of rhetoric that is represented by this post:

    “February 2, 2012 – 5:07 pm Jonathan Jones

    In other words, BBD, the “subscribers to the mainstream view” are lying to avoid being misunderstood. Tempting, but foolish.

    In addition to violating your rules for commenting, this comment is of a kind that isn’t unusual to find among “skeptics.” Do you think it is important for “skeptics,” themselves, to distinguish between this type of viewpoint and serious skepticism in the same way that it is important for “realists” to distinguish between advocacy and science?

    • Tamsin Edwards

      Thanks Joshua. It did breach the rules, as BBD pointed out (I’m all for self-policing…).

      Do you think it is important for “skeptics,” themselves, to distinguish between this type of viewpoint and serious skepticism in the same way that it is important for “realists” to distinguish between advocacy and science?

      That was the point I was making with the horizontal axis. We all aim to be at the right-hand side, but some end up on the left, at least for some issues: when it comes to scientific topics, the left-hand side of the axis does nobody good, whether they are in the positive or negative region of the vertical axis.

      I’m not trying to label Jonathan with coordinates with that statement. I do agree with him that misleading simplification – lying by omission – can harm both the speaker and the listener. I didn’t agree with his implied generalisation that all subscribers to the mainstream view do this.

      Hope that clarifies.

      • Joshua

        Thanks tansim –

        It clarifies a bit, yes.

        And of course, lying by omission is an important issue to be addressed.

        But my question wasn’t really w/r/t labeling the debaters (or commenters). Please note, I asked about “this type of viewpoint,” not “this type of person.”

        The point, IMO, is to work together to openly analyze statements that people make, so that people can share dialog to clarify meaning, and clarify the certainty implied by statements (or misstatements, or misleading statements), or fail to do so.

        There is an almost inevitable error, in my view (notice I said almost) when people try to generalize from a statement to labeling the commenter (particularly when someone doesn’t know the commenter personally). I might suggest that might be like extrapolating from a model without properly quantifying certainty.

        What I find particularly interesting is when people who self-identify as “skeptics” – (i.e. someone globally concerned with underestimations of uncertainty in contrast to how they view “realists” ) – make statements that at least at a surface level (without added clarification to address possible misinterpretation on the part of the reader or a lack of comprehensiveness in communication on the part of the commenter) seem facile. What does it mean that “skeptics” make comments that display a lack of skepticism? Is it similarly instructive as when “realists” draw facile conclusions, say, about extreme weather events and attribution?

        In my view, it means that we are all subject to the same kinds of influences in how we approach these issues. Perhaps trying to find places to locate individuals on a matrix, and assigning them a label, misses a larger frame of reference that would be more informative in the long run?

        • Sashka


          If you said some “skeptics” make comments that display a lack of skepticism it would be a good observation. But you didn’t. An interesting omission, no?

          You are certainly free to give your own definition of “skeptic” or “realist” but don’t be surprised when people disagree. Some people have more beef with climate science than just uncertainty.

          Personally, I wouldn’t call people who don’t understand uncertainty “realists”. Not sure what would be an inoffensive description but surely not that.

          • Sashka

            The HTML doesn’t quite work for me. I meant some in bold (that didn’t work) and your quote in italic. The rest in italic is probably my error.

    • DC

      1) As Tamsin said, the comment was already noted as in violation of the quite civilized comment policy.
      2) There is no dot on the graph for “realists”, so your analysis fails entirely. Initial and all subsequent comments that use the term “realist” in this thread are irrelevant without a definition in context of the post.
      3) Joshua has been around the blogosphere, I’ll refrain from ad-homs, You can make your own decision.
      Sorry if only point 2 was on topic, but I tried to keep it short.

      [I’m approving all these comments this time, but should probably close the thread for comments… — Tamsin]

  82. Theo Goodwin

    Tamsin writes:

    “theo – “There is no fundamental epsitemic difference between a model and a physical theory.”

    This is an interesting point. My first instinct is to agree with this: a model is (in the context of this blog) a coded representation of physical theory, with the necessary numerical and scientific approximations etc. It may also have coding errors and therefore not exactly represent the theory. A simulation is an instantiation of the theory, given certain initial and boundary conditions. But I’m open to having a debate about whether this is the case 🙂 (perhaps better after I write the post though).”

    I thank you for your response to me. I remain interested in seeing Tamsin be Tamsin. For that reason, I suggest that we put my topic on the back burner. I will explain my point very briefly here. However, I will address the larger context of my comments.

    A physical theory is a set of universally quantified conditional statements about physical reality. As statements about physical reality they will prove to be true or false depending on our future experiences. A good physical theory consists of well confirmed physical hypotheses. A good physical theory has a track record of all true predictions and no false predictions. A prediction is implied by a combination of some physical hypotheses and some set of observation statements that describe the initial conditions, the known observable facts at the time of the prediction. The statements that make up a good physical theory describe the natural regularities that make up reality. A future or past event can be given a scientific explanation by showing that it could have been predicted from some of our physical hypotheses. The set of all such true predictions and retrodictions (predictions into the past) constitute all of observable reality past and future. A good physical theory occupies the very special position of implying all of observable reality. (Of course, we are speaking of a logical ideal. Actual physical theories imply the features that we take to be the salient features of reality.)

    A model of a good physical theory is the set of all events implied by the theory. In commonsense terms, reality is the model of our physical theory. In formal terms, a model of our physical theory is any set of objects that renders true all the hypotheses in our theory. A good illustration of these points is the famous example of Kepler’s Three Laws. Kepler’s theory is about the motions of the visible planets in our solar system. After Newton invented calculus, Kepler’s theory was rigorously formulated by Newton and shown to be deducible from Newton’s Theory of Gravitation.

    After Galileo built himself a telescope, he created quite a fuss by observing phenomena such as Jupiter’s moons and Venus’ phases. He used Kepler’s hypotheses to predict systematically the phases of Venus. For some decades now, colleges have been able to purchase a computer system and a hemispherical dome that shows the sky at night as seen from Earth. This system is based on Newton’s formulation of Kepler’s Laws. The points of light that appear on the ceiling can be taken as a model of Kepler’s Laws. You can “fast forward” and watch Mercury as it travels across the sky, stops, backs up, and continues its journey. This “fast forwarding” is based on prediction from Kepler’s Laws. Of course one of Kepler’s great achievements was to explain that the apparent “backing up” by Mercury resulted from the varying speeds of Mars and Earth in their orbits and the positions of the orbits relative to one another.

    Let’s review the main terminology. The computer driven system is a model of our solar system and the sky at night as seen from Earth. This model is specified by Newton’s formulation of Kepler’s hypotheses. That is, Newton’s hypotheses are used in construction of the computer code that generates the show on the hemispherical ceiling. You can dial up any event you would like to see. The dial up is done by predicting the event from Kepler’s Laws. Any event that is “dialed up” can be scientifically explained by reference to Kepler’s Laws.

    It is very important that we understand the relationships of logic and evidence that exist between our physical theory, Kepler’s Laws, and its models. Kepler’s Laws imply both what we observe from Earth when we view the planets in the night sky and the points of light in our computer driven observatory. However, the evidence for Kepler’s Laws must come entirely from our observations of the night time sky, our observations of the solar system. Clearly, we cannot use our observations from the toy observatory as evidence for Kepler’s Laws. Kepler’s hypotheses must be evaluated on the basis of their ability to make true predictions in our experience of the world and that evaluation must take place independently of any model that we might construct. The evidence for statements that embody our understanding of reality are necessarily independent of any model of that reality that we might construct in a computer. And no less important, any model that we might construct independently of our physical theory cannot be known to rest on a genuine understanding of reality until there is a physical theory which consists only of well confirmed physical hypotheses and which implies that model. In epistemic terms, well confirmed physical theory must precede any model that we might construct.

    Does this mean that computers models are entirely useless? Of course not. It means simply that we cannot make inferences from constructed models to physical theory. We cannot create a model and claim to have learned from it that there exists a particular causal connection in reality. To claim that “we believe that manmade CO2 causes warming because only the addition of the effects of manmade CO2 enables our models to reproduce the increasing temperatures of the late 20th century” is to make exactly this error.

    It seems to me that when climate scientists refer to their models they are actually referring to something that is part well confirmed physical theory, though an incomplete theory, and part simulation. We know that the models and the simulations that they generate are incomplete. I suggest that climate scientists need to learn to strictly separate the two items, theory and model, along the lines I have explained above. Until that task is accomplished, neither climate scientists nor their critics will know what they are claiming or what evidence they offer. One thing is for sure. The claim that existing models can be used as evidence, or as any sort of reason, for believing that CO2 causes warming is misguided.

      • Theo Goodwin

        Tamsin, take it easy on this stuff. I still want Tamsin to be Tamsin. You are toooo kind.

    • Joshua

      So – in one comment first we have this:

      A physical theory is a set of universally quantified conditional statements about physical reality. As statements about physical reality they will prove to be true or false depending on our future experiences. A good physical theory consists of well confirmed physical hypotheses. A good physical theory has a track record of all true predictions and no false predictions.

      And then we have this:

      The claim that existing models can be used as evidence, or as any sort of reason,for believing that CO2 causes warming is misguided.

    • Joshua

      Theo –

      “we believe that manmade CO2 causes warming because only the addition of the effects of manmade CO2 enables our models to reproduce the increasing temperatures of the late 20th century”

      Now those quotation marks in the statement above came from your comment. Does that mean that someone actually said what you placed between quotation marks, or is that a paraphrase of yours that you added the quotation marks to?

      As I understand it, the paraphrase you use there would not accurately summarize the view of many frequently considered to be among the “consensus.”

      Take, for example, the IPCC perspective of some 90% certainty that more than 50% of recent (anomalous) warming is caused by ACO2. There is a difference, there, between the “certainty” represented by your paraphrase and my summary of the IPCC perspective. Your statement suggest an absolute belief – as opposed to:

      “we believe that it is very likely thatmanmade CO2 causes warming , and most of recent anamolous warming because only the addition of the effects of manmade CO2 enables our models to reproduce the increasing temperatures of the late 20th century.”

      That isn’t to say that some involved in the debate might not make statements without any representation of uncertainty as suggested by your paraphrase. But don’t you think it is important to take care to represent the variety of viewpoints accurately, and to carefully attribute viewpoints?

      I think that the conclusion of your post is highly dependent on the accuracy of your paraphrase. If someone says that physical models absolutely confirm a conclusion it is different than saying that physical models are evidence that represent probabilities in support of a conclusion. Argue as you will about the probabilities, and their validity – but not recognizing the differences there seems problematic, IMO.

      • Joshua

        I should add…. “not recognizing the differences there seems problematic, IMO – no matter who it is that is failing to understand the differences between absolute certainty and an estimation of probabilities.

      • Theo Goodwin

        Good question. Mine is a paraphrase. Changing it to a probabilistic claim makes no difference. They are still using a model to make an inference that a causal connection exists in reality (though only probably so). That point is inescapable. It is also unforgivable in genuine science. I cannot make an inference from a model that I could not have made beforehand from the thing modeled. To do so is to create something out of nothing. I do not do theology.

        • Joshua

          They are still using a model to make an inference that a causal connection exists in reality (though only probably so)….I cannot make an inference from a model that I could not have made beforehand from the thing modeled.

          I need some help with that. Perhaps the answer lies in the distinction, or lack thereof, between a physical theory and a model that flies over my head, but I fail to see the problem to make an inference about a causal connection with acknowledgement of probabilities.

          I drop an apple and it falls to the ground. Now maybe that happens because some supernatural being is always on the watch and whenever I drop an apple, for some unexplained reason, sees fit to make that apple drop to the ground – so I have to recognize probabilities there. But I also offer a model (or is it a physical theory?) that gravity is a casual connection between two observed phenomena. My model provides a different estimation of probabilities in the cause-and-effect relationship than I would have had absent the model (without the model, I would be likely to presume that a supernatural being was the link between cause-and-effect).

          As I understand your reasoning, such a causal connection supplied by a model (physical theory) of gravity would be logically and scientifically invalid. How am I failing to understand your point (you’re going to have to dumb it down for me – I need simple and non-scientific terms).

          • Theo Goodwin

            I am reading from the bottom up on this blog so I answered Tamsin below. Check that answer.

            Apparently you learned probability from climate scientists who persist in their mistaken belief that there can be science of data and statistics alone without the benefit of either natural regularities or the physical hypotheses that describe them. Read the Population Geneticists on statistics and probability. Among scientists, they are the best on the topic.

            In your Newton example, what you call a model is actually Newton’s Law of Gravitation which is a collection of well confirmed physical hypotheses. (We ignore Einstein for the moment and doing so changes nothing.) So, you were cribbing Newton. The falling apple is the model.

            A model that is a simulation produced by a computer is a reproduction of reality not a description of it. Just look at the results of a computer run which produces a simulation of something. What you have are numbers. Those are magnitudes or some such that are assigned to salient features of reality. See, the simulation reproduces reality but does not describe it.

          • Joshua

            Theo –

            In your Newton example, what you call a model is actually Newton’s Law of Gravitation which is a collection of well confirmed physical hypotheses….The falling apple is the model.

            We all build models in one form or another to understand cause-and-effect relationships between phenomena we observe. In fact, gravity is a theory of understanding that is built upon a model that infers much from what we can’t understand, nor observe.

            A model that is a simulation produced by a computer is a reproduction of reality not a description of it.

            The theory of gravity is a mental simulation produced by our brains and is a reproduction of reality not a description of it.

            And Theo – my discussion here has nothing, at all, in any way, to do with “learn[ing” probability from climate scientists.”

            You have built a mental model that is easily falsified – and the reason why it is so easily falsified is because you assumed certainty without sufficiently validating your conclusions about cause-and-effect relationships.

        • Tamsin Edwards

          Interesting thread. I don’t understand this sentence Theo (too many negatives…) – could you expand? Are you saying that inferences can only be made from observations, and never from simulations?

          I cannot make an inference from a model that I could not have made beforehand from the thing modeled.

          • Theo Goodwin

            Let us start with Earth. I create a model of Earth. This model is the most perfect model possible. It reproduces through computer simulation all salient features of Earth. Looking at my model, I find something, some connection, that I have not yet found on Earth. Let us say that I find a connection between increasing manmade CO2 concentrations and warming. I conclude that Earth (not the model) is warming because of increasing CO2 concentrations. .

            See, I just inferred something about Earth from a model when I was unable to infer that thing from Earth itself. Surely, that is creating something from nothing. Surely that cannot be outside of theology. Hence my prohibition: You must not infer some real causal connection from your model when you could not infer it from reality itself.

            The key word in all this is ‘inference’. I cannot infer from my model what I cannot infer from reality. Because all scientific evidence is a matter of inference (my physical hypotheses are good ones because I can infer from them true observation sentences (predictions)), my prohibition stated above means that my model offers no scientific evidence or reason for believing that there is a real causal connection.

            Now to finish the story. If my model suggests to me (suggests) a hypothesis about a connection between CO2 and warming, I should attempt to formulate that suggestion as a physical hypotheses and attempt to confirm it. Creation and confirmation of hypotheses exists entirely in reality and in science. So, models can be immensely useful, and they are far more useful as analytic tools than as sources of suggestions for hypotheses.

            Hope that helps.

          • Theo Goodwin

            Something that might be helpful just occurred to me. Let’s start with my prohibition:

            “Hence my prohibition: You must not infer some real causal connection from your model when you could not infer it from reality itself.”

            Another way to say this is the following: “Anything that you infer from your model that is about reality but that cannot be inferred from reality itself is necessarily an artefact of the model.”

            Keep in mind that reality is what is described by science. Reality is what is described by our complete set of well confirmed hypotheses. Our model is a reproduction of the salient features of reality. Our science is a set of conditional statements. Our model is a set of objects.

          • Joshua

            Theo –

            Let us start with Earth. I create a model of Earth. This model is the most perfect model possible. It reproduces through computer simulation all salient features of Earth.

            The problem there is an assumption of a “the most perfect model” that produces “all salient features of Earth.” If someone were to make such an assumptions, and from that infer absolute conclusions, they we’d have a problem.

            If, on the other hand, if someone says that they have built a model with an estimated probability of representing an estimated percentage of salient features of the Earth, they might infer estimates of probabilities of certain outcomes. We might argue about their probability estimates – but what they are doing is not the same thing as in the hypothetical you provided. To reject AGW on the basis of the logic of your hypoethetical is rejecting AGW on the basis of a straw man. It is of little use, IMO.

          • Vaughan Pratt

            I cannot make an inference from a model that I could not have made beforehand from the thing modeled.

            Assuming I’ve understood this the way Theo intended (which could well not be the case), I would disagree with it as follows.

            Suppose some phenomenon happens to be following the shape of a gaussian or bell curve, but you can only see the top, between the two points of inflexion. It looks like the top of a gaussian to you. To be sure you do a least-squares fit and obtain an R2 of 0.98. This is sufficiently solid evidence for a gaussian so you infer from the phenomenon that it is a gaussian, and hence will decay to zero in due course.

            But then someone asks how you can distinguish that shape from a sine wave, pointing out that in that case it would not die out but keep oscillating.

            Interesting thought, so you try fitting a sine wave. This time you get an R2 of .9999. This would be amazing except that the phenomenon is so free of noise that such an R2 is not impossible.

            You furthermore realize that with the gaussian model there is not enough noise to account for the unexplained variance of 0.02. This casts doubt on the gaussian model, prompting you to switch to the sinusoidal model.

            Now was this inference made from the phenomenon? Well, you’d already inferred from the phenomenon that it was a gaussian. It wasn’t until you had a second model, and could compare the quality of fit of the two models, that you were in a position to infer that it is more likely to be a sinusoid than a gaussian, even though the gaussian at first seemed a perfectly reasonable inference.

            So I would judge the inference that the phenomenon was more likely to be sinusoidal than gaussian to have been drawn from the model, not from the phenomenon itself.

            Theo may have his reasons for arguing that ultimately the inference was drawn from the phenomenon. But even if he could persuade others of his reasons, they would not show that modeling served no purpose, as it clearly did in order to decide whether it was more likely that the phenomenon would die out or oscillate.

    • Stefan Thiesen

      [ Stefan, sorry this long and thoughtful post got stuck in moderation. I’m back now! — Tamsin ]

      @Theo: I also fail to understand the fundamental difference between a Theory/Hypothesis and a model. Much of theoretical Physics is modeling and long has been, and deducing a conclusion from a theory/hypothesis/model leads to a new hypothesis/theory/model. A plain example is Pauli’s postulation of the Neutrino. As you probably know, he was embarrassed to conclude the existence of a particle that could not be detected at the time. I consider this good scientific intuition (a term Tamsin seems to really abhor, if I get it right). The underlying fundamental concept that led to the conclusion basically was energy conservation. In a sense the Neutrino was the dark matter of it’s time. Quantum Electro Dynamics partly came about in similar ways – Feynmans’s intuition and Tomonaga’s, Schwinger’s and Dysons solid number work. The result was – a model, if I am concerned, that made predictions which later were confirmed. The process was a guided random walk of sorts, as theoretical physics always is.

      You mention Kepler’s laws as an example and their exact formulation by Newton. We could also see it the other way round: Kepler made observations which he fitted into a model. Kepler’s laws are neither a theory nor hypothesis – they are purely descriptive. Hence they are called “Kepler’s laws” and not “Kepler’s theory”. Newton then, with his (seemingly) exact formulation, built the hypothesis, which observation over time confirmed into a theory. Yet still only an incomplete theory, because it turned out that Newton’s analysis also was a simplification of reality – and I am not even speaking about relativistic effects. In his book “The Transition to Chaos in Conservative Classical Systems” Reichl wrote: “In a sense Newton (and Western Science) were fortunate, because the solar system has amazingly regular behavior considering its complexity, and one can predict its short term behavior with fairly good accuracy”. For the long term behavior of the solar system the toy models are an illusion. N-body problems are not only mind blowingly complex – they also cannot be solved analytically. Even whether the solar system will remain stable in the long run is unclear. There is a certain sensitivity aspect involved – the parameters constantly change, even if ever so slightly (mass change of the sun, impacts etc., etc.)… Ultimately Newton’s theory IS the model. It is the raw software. The actual computer models that employ various numerical methods to SIMULATE the solar system merely are formulations of Newton’s theory/model. Whether I calculate on paper or by means of computers makes no difference, epistemologically.

      Where does that leave us with climate models? Interestingly simple 1 D climate models
      that were based upon radiative fluxes and radiative forcings by atmospheric compounds in a very simple manner describe the temperatures of planets in the solar system fairly well. The close correlation between the forcings and the measured planetary temperatures hinted that the earlier proposed CO2 induced radiative forcing (Fourier, Arrhenius, Tyndall etc.) had a physical foundation. So what we have was physical intuition and hypothesizing (Fourier) –> early experimentation (Tyndall) –> early theoretical formulations, including the first atmospheric model (Arrhenius). This cycle since then continued and will continue into the future. The outputs of climate models constantly are matched with observation to see whether they make right predictions. And when they don’t, that may have one of two reasons: either the model is wrong (and can be fixed), or certain parameters principally cannot be properly predicted. Now Michiu Kaku once told me that Chaos Theory didn’t give us a thing, because it has no predictive power. Physicists hate that. And I totally disagree. If we have a correct model of nature that still fails to predict the future, we still do learn something about the nature of nature. And since for the most part modern climate science came about to assist policy making, we are entering the muddy realms of decision making processes at this point. Often the models tell us a lot about the general behavior of complex systems. In many cases the probability of certain outcomes can be estimated. So what is inferred from complex models often is insight into the actual behavior of the underlying physical systems and improved mathematical methodologies for their description. Like with Newton and Kepler it is a guided random walk. An approximation of reality. It is a mess – but it is all that we have.

    • John Costigane


      I am saddened to see the behaviour of both extremes transfer to your website, which I see as a positive development. Sceptics have been invited to participate in this process, with an open mind, and should act accordingly.

      Have you considered requiring posters to provide their full names? There is no place here for disruptive trolls.

        • John Costigane


          The behaviour of the unidentified is destructive on many sites. My personal approach is to avoid them, as a rule. I accept the need for anonymity. There is too much heat and less light in the other sites, while this site should be all light and no heat.

          I accept the need for anonymity

          • Sashka

            Depending on how you define “unidentified”. Just because someone calls himself Jack Smith doesn’t really mean that it’s a real name. You’ll have to go further to positively id the person via, I don’t know, Google+ or Facebook, as NY Times has recently done.

            In my experience, both the worst trolls and the most interesting people are anonymous.

      • Barry Woods

        Theo does have some good points about models.

        Can’t wait for Tamsin’s actuall first post about models – Given the name of the blog!

      • Anteros

        John –

        Funnily enough I’ve had the opposite impression. My take is that compared to almost anywhere else in the blogosphere, people here are falling over themselves to be polite and follow the comment policy. Theo’s transgression was in my mind trivial, Tamsin’s strictness good-natured…

        I hope Tamsin keeps up with the effort of reminding us and re-drawing the line frequently enough to prevent a slide into common blogospheric trolling.. It might get to be a chore but I think worth the hassle..

        You certainly have a point about anonymity – for me it’s just a habit. I think you’re right that it feeds the trolls within us, but I don’t think insisting on real names is going to happen – although I’d be happy to conform. Anteros would still remain anonymous elsewhere!

        • John Costigane


          Open minds make no assumptions or demands. It is better to follow the reasoning in Tamsin’s posts and make relevant comment there. In the meantime, the tone should be civil and we should all be patient.

          On the trolls issue, good behaviour is essential from everyone. Scientific discourse needs cool heads.

      • tonyb

        John Costigane

        I tend to agree that the posting of full names is desirable (for the record my full name is Tony Brown)

        I call myself tonyb to differentiate myself from the hundreds of other Tony Brown’s out there, not for the purpose of trying to create a shield behind whch I could troll. In other blogs I have seen a variety of people explain why they don’t want to be named, of which possible complications at work are a factor (I know of two people in this category)

        I know Anthony Watts and other sceptics have felt threatened and I’m sure it cuts both ways. I suppose it might be useful to determine what the meaing of a troll is and I think that is likely to vary according to the blog and subject matter. There are a couple posting here who are considered trolls elsewhere but they are without malice and just tend to like to talk about whatever subject is being discussed-whether or not they have any knowledge. I don’t think that asking them to post their full name would deter them.

        So perhaps the keyword is the blog here needs to be moderated for comment relevance from the start and those who will argue about anything for the sake of it will quickly get the idea and move on to more fertile pastures. This will allow those who want to be anonymous to do so and allow them to be judged on the quality of their comment and not their name.

  83. Tamsin Edwards

    My take is that compared to almost anywhere else in the blogosphere, people here are falling over themselves to be polite and follow the comment policy.

    – Anteros

    I agree, I think people have (mostly) been keeping to the policy well – thank you all. It makes a huge difference.

    I’m aiming to post tomorrow…

    • Anteros

      Tamsin –

      I think people so far have been keeping to the comments policy for many reasons., and I agree it’s great for those of who want to discuss and learn and so on.
      I do think though there are trolls within many of us and the continuing civility here will correlate with your willingness to do the nagging and boundary-defining that will keep those trolls at bay.
      It might be tiresome but will be appreciated by the majority – as well as those of us with poor self-discipline!

    • Tamsin Edwards

      Hi Steve,

      Thanks for dropping in. Lots of useful explanation in that paper, though I confess I didn’t understand the meaning of the diagram. I absolutely agree there is a risk of scientists using statistical tools because they exist and are convenient, without thinking too much about their power or suitability.

  84. Pharos

    With apologies in advance is this is considered too elementary, but one aspect of climate data and climate models I would appreciate being clarified in some detail at some future point, is what the conventional practice in climate science is regarding land surface datum. Is the continental topographic surface used as a conventional datum as it is, or are altitudinal corrections, for example lapse rate/pressure adjustments to mean sea level datum applied, or indeed both? In the Greenland and Antarctic Ice sheets, from what survey data is the ice base surface defined, ie seismic, gravity-magnetic? Clearly topography has a major influence on weather patterns and must be represented somehow in the weather forecasting models, but what about the climate models? What map projections to represent the global surface are preferred? What grid densities are used? Are climate ‘regions’ defined by polygons?

  85. John Costigane


    Great to have your input here, as the most respected on the sceptical side. What is your view on Tamsin’s Sceptical Compass?

  86. Anteros

    Tamsin –

    I’ve been thinking some more about your sceptical compass. Your labelling of the vertical axis has stumped me from the beginning – it’s been a bit of a head-scratcher!. You say –

    The sceptic/scientist shorthand I use corresponds to this axis.

    From what I know, your definitions make Al Gore and George Monbiot much higher up the ‘scientist’ scale than Richard Betts or Richard Tol. How can this be?

    The other problem is that you inadvertently characterise scientists and sceptics as in opposition to each other – the more sceptical one is, the less one is a scientist, and vice versa, which is also odd. John Christy would fail to appear as a scientist on your compass at all!

    I can sort of understand that from a consensus point of view, the sceptical ‘voices’ that you hear may be predominantly ‘against’ the (scientific) consensus, but I think this is misleading [apart from a few vocal American extremists]. Much of the scepticism in the climate debate is about either the imagined (negative) consequences of a certain amount of warming or about the wisdom of de-carbonisation as a policy option. The GWPF are an example of both these. And other non-consensus positions [Lindzen, Spencer etc] are non-consenus for reasons of science. They are not anti-scientific, but they are also very much towards the no impact end of your vertical scale.

    My gripe (if I can call it that) is that your ‘scientist’ position is defined as one that believes in ‘intense and rapid’ impacts. So, the more you believe there will be intense impacts of climate change, the more scientific you are. That seems very bizarre!

    There aren’t, as far as I can see perfect alternatives, but I do think having scepticism and science in opposition is misrepresenting one or the other. In a way I would drop the critical variable simply because most people would want to be at the right hand end – it is a meritorious quality. In it’s place one could ask how much warming [say, over the next century] do you think there will be? It’s a simple variable but gets to the heart of the debate. And, if you like, allows for an expression of scepticism towards the (scientific) consensus. How much warming? 1 degree?, 2 degrees?, 3?, More? Then the vertical axis could be used for how alarmed people are by the warming they believe is likely. This also is the big motivating factor in the debate – the worry or fear concerning the future.

    The above is admittedly moving away from your work and your blogs aims, but it would be illuminating. For instance the degree of alarm may well not correlate with the imagined degree of warming. And we know that James Hansen is convinced two degrees of warming threatens catastrophe for life on earth whereas Richard Betts doesn’t subscribe to the meme that it is dangerous..

    The point of all this, in a way, is for me to ask the question where lies the edge of scientific authority? In agreement with Richard Betts, I don’t think science has anything to say about danger – it is a value judgement. Whereas Michel Cricifix and Stefan Rahmstorf believe they can say as scientists that x, y, or z is dangerous for humanity or ecosystems or whatever.

    You made a comment a day or two ago about hoping to restore trust in climate science. I’m sure there can be improvements on that score post-climategate, but I think there is a bigger and deeper problem, which is that many perceive climate scientists as speaking beyond their authority. Not just minimising uncertainties in the science, but claiming authority to speak about the meaning of climatic changes and the degree of ‘impact’ certain things will have.

    From what I can gather, you quite sensibly stick to science in your public statements, but many scientists clearly don’t and my suspicion is that a great deal of the groundswell of scepticism is not against pure, value-free science, but against scientist-activist advocacy. Perhaps I’m arguing for greater clarity in the demarcation between something that is scientific evidence [for a physical process in nature] and the motivations that this evidence provokes in the human beings who use the results.

  87. Vince Whirlwind

    ” I have never seen a “sceptic” graph which is so patently idiotic as his example.”

    But that’s precisely what the claims of a recent “pause” in warming are: equivalent to one of the “pauses” on the graph.
    There is no “pause” and you’re right – it is indeed patently idiotic to call a cherry-picked sequence of variability, a “pause”, especially when including all the other data which is available (instead of ignoring it to manufacture a false narrative) shows you very clearly what the trend is that underlies that variability.

    It’s actually beyond comprehension that people even fall for such obvious statistical tricks.

  88. Ian H

    Your compass is helpful as far as it goes. However I would find it only of very limited use in classifying opinions on this subject. It is too general and doesn’t get to the heart of the matter.

    The problem I think is that your compass approach treats global warming is a single discrete idea about which one can be – believing or disbelieving, critical or uncritical – as the case may be. It does not recognise the fact that the theory as presented to us is not a single simple idea. It is instead a rather large and complicated story, a complex chain of reasoning. Those who are sceptical about any single step in that chain will most likely view the conclusion with doubt and will be classified by society as sceptics.

    I find the most useful way to classify sceptics is to first determine first the aspects of the story that they are sceptical about. Only then would your compass be useful in analysing the nature of their scepticism at that point.

    There are some places where the chain is weak and scepticism is a quite rational response. For example the standard story dismisses the possibility that the observed warming might be natural far too readily. The assumption of positive feedback from clouds is also very dubious as all evidence that I have seen points in the reverse direction and without such feedback the warming from CO_2 alone is likely to be minimal. One may also rationally doubt the downplaying of the correlation with the solar cycle; the bogus statistics behind the construction of the hockeystick; the massaging of the historical instrument record; the overstating of the risk of sea level rise; the imputation of catastrophic effects to small temperature changes. In my opinion it is entirely rational to be sceptical about the story at these links in the chain. I am not sceptical at all of these places myself, but I recognise that scepticism is rational at these points. Identifying which of these things a sceptic is sceptical about helps me understand where they are coming from and what type of sceptic they are.

    And of course there are some sceptics who are sceptical for reasons that I find irrational. Knowing which aspects of the story they are sceptical about is useful to for identifying them as well.

  89. Greg, from Spokane

    “This post is already too long…”

    Nope, not true. Make it as long as it needs to be in order to say what you need to say.

  90. Greg Goodknight

    Tamsin, I think another valid subtitle might be “Some models are more wrong than others”.

    A mere BS Physics, MSEE, I became convinced in 2007 that all of the GCM were wrong due to the simplistic cloud models, a statement by Lindzen (iirc) that the IPCC view was that more moisture evaporated from surface waters would not increase cloud cover, Svensmark and Friis-Christensen’s (and other) thoughts on the effects of GCR and solar magnetic field on climate and the incredible alignment of Nir Shaviv’s plot of the solar system’s orbit around the galaxy, its relationship with carbon 14, and the near match of carbon 14 over 500+ million years with Jan Veizer’s oxygen isotope ratio proxy for ocean temperatures.

    One postdoc friend of mine in a paleomagnetism lab, when I laid my early 2007 revelations on them, told me the paleos had noted some time ago that there seemed to be a relationship to carbon 14 dating data and global temps, but the fossil hunters didn’t have a clue what to make of it so they applied standard human behavior… it pretty much got ignored.

    I couldn’t see how a mechanism that obviously drove climate to 7C temperature spreads peak to peak, averaged over millions of solar cycles, could not be the cause of a fraction of a degree over a half century when the sun was in perhaps an 8000 year maximum (by the Solanki letter to Nature). What I heard from my purely social NCAR connections was that Shaviv & Veizer was geologic time, this isn’t, so we can ignore it. Madness.

    Forgive my hardness, but the intense personal defamations I’ve endured since 2007 have left their mark. Until the GCM have included solar-magnetic effects, with a better model for clouds, and have something other than failed predictions to show for past published scenarios, they are worse than wrong, at least for the purposes of politicians deciding on winners and losers.

    Adding onto what Ian writes, I believe there are as many irrational Believers and the Convinced, swayed purely by authority, as there are irrational disbelievers and unconvinced.

  91. Stefan Thiesen

    What I don’t like about the “skeptical compass” is that it is not merely descriptive – it also contains values. Skepticism is built into the very methodology of science itself. It is at the core of science. But being “uncritical” is akin to saying mindless, ignorant or, in short, “stupid”. Unfortunately most of us are stupid in this respect, when it comes to climate science (or any other major scientific endeavor). In my case – although I did some graduate work in atmospheric sciences (many years ago), I fail to understand much of the contemporary work and cannot properly evaluate it, lest it is based upon complete and obvious crackpot bogus. Also keeping abreast with all the developments and discussions is difficult enough when working in the field full time. As a casual sideline it is impossible even for some specialized sectors. So – as a result I am more or less ignorant as well. And at least 99.999% of the world population are even more ignorant in the field than I am. It is essentially pointless to be skeptical in a field one is entirely clueless about. On top of it – there is a large percentage of the human population that is not only not numerate, but that entirely lacks the ability to properly learn even basic calculus. There are people, who otherwise function entirely normally in life, who simply cannot grasp higher math of any sort. Anyway – what I think I CAN evaluate to a certain extend is the reliability of certain authors and institutions, so I do tend to value results from scientists working at publicly financed institutions higher than those coming from private entities of any sort. So – for the most part, I am guilty of trusting sources I “consider authoritative”. Because I am a firm believer – in the peer review process. But actually – don’t we all trust sources we consider authoritative? Obviously we don’t trust sources we don’t consider authoritative. And equally obvious: we cannot repeat every experiment and derive every equation in the history of science ourselves. I am afraid, as with all models, your compass has a few weaknesses. For slight improvement I would add a third dimension: Level of education, with “50 years experience in climate modeling” at the far positive end of the Z and “Dyscalculia” on its far negative side. But – it might cause new logical and value inconsistencies… admittedly the idea is not really thought out…

    [ Stefan, thanks again for this. I don’t think ‘uncritical’ is as judgemental as you imply: we are all (at times) guilty of it, because we are human. This sentence made me smile: “There are people, who otherwise function entirely normally in life, who simply cannot grasp higher math of any sort.” I do think that people without training in climate science or any other STEM subject can ask very useful and challenging questions. This is partly where intuition comes in! I don’t abhor the word at all – it’s been very important in making great scientific leaps – but also I wouldn’t necessarily trust a prediction based on intuition over one based on physical theory. — Tamsin ]

    • Stefan Thiesen

      Hi Tamsin: Well – looks like I am replying to myself here now… of course my “otherwise they are functioning normally” was pretty pointed. I know people who are language geniuses while being mathematically challenged. But intuition: I think (i.e. my intuition tells me) that a person who mastered a subject and worked in and on and about it for decades (like Jim Hansen) does develop a scientific intuition that incorporates all the facts and experiences from his or her professional life. Physics is a notoriously un-intuitive field for the uninitiated, because often the abyss between physical reality and personal experience is so wide that no intuition short of true genius can bridge it. So – when Jim Hansen vaguely feels that the system state of Earth’s atmosphere is hovering somewhere near a point of no return to runaway warming that strikes me as more relevant than someone intuitively saying “there is no global warming – last night frost killed the flowers in my front yard.” Still: non-experts can ask useful questions. (In economics a useful question regularly asked by non economists: how can something grow forever…). Alfred Wegener also was a non-expert in the field of his most important discovery. Anyway – my big issue is: how to decide when the facts are and remain necessarily vague? And it comes down to trusting people, to accepting their “educated guesses”. And that is why systematically dismantling their public image (as it happened and still happens to James Hansen, Michael Mann and others) strikes me as particularly sinister. Also on another level, because the onlooking layperson cannot distinguish anymore between the relevance of the claims made by different opponents in a discussion. Ultimately: I also wouldn’t trust the predictions of an elder scientists intuition more than that of established theory. But it can help in cases when theory gets all muddy and unclear and we have to make decisions in realms of uncertainty. Carl Sagan was once pressed to express opinion about the probability of Alien life. He said he wouldn’t know. He was asked for his gut feeling and answered, he’d try to not think with his guts. But occasionally our gut feelings can save our lives (and I suddenly have a gut feeling that this was a somewhat messy posting).

  92. Simon Bowyer

    This is a welcome intervention indeed in the never ending AGW debate. I would say that 5 years ago I was uncritical convinced, I am now sceptical unconvinced. Climategate was one factor, but the sheer obnoxiousness of RealClimate and the antics of the team convinced me that they had something to hide. The Charles Monnet affair last year really pushed me over the edge.

    Perhaps, a little gentle and reasoned debate as promised here, could tempt me back to the sceptical lukewarmer position. Genuinely excited about discussing science, models and data without the logical fallacies and name calling usually associated with AGW debates.