I’m back!

This is my first blog post in almost exactly 10 years. As an experiment I have asked a Generative AI questions about a social science climate related topic – The 97% Consensus.

I first wrote about my concerns that “97% of climate scientists” was being used as a soundbite, over a decade ago at Watts Up With That (Woods, 2012) , evaluating how both the Doran and Anderegg papers had been publicised.

What else did the 97% of Scientists say? – WUWT

The questions I have put to a Gen AI are in bold, and the results are complete and unedited. and I have added some blogger commentary at the end of this post.

Question – What is the object of the 97% consensus on climate change. What do scientists agreed about. [I even had a typo! agree not agreed]

The 97% consensus on climate change refers to the agreement among climate scientists that global warming is occurring and is primarily caused by human activities, particularly the emission of greenhouse gases like carbon dioxide from burning fossil fuels, deforestation, and industrial processes. This consensus is based on extensive peer-reviewed research, including studies like Cook et al. (2013), which analyzed thousands of scientific papers and found that 97% of those expressing a position on climate change endorsed human-caused global warming.

Key points of agreement:

1. Global Warming is Happening: Earth’s average surface temperature is increasing, as evidenced by data from NASA, NOAA, and the IPCC.
2. Human Activities are the Primary Cause: Greenhouse gas emissions from human activities, such as burning fossil fuels and land-use changes, are the dominant drivers of observed warming since the mid-20th century.
3. Scientific Evidence is Robust: Multiple lines of evidence, including temperature records, ice core data, sea level rise, and glacial retreat, support the conclusion of anthropogenic climate change.

The consensus does not imply agreement on every detail, such as the exact rate of warming, specific regional impacts, or policy solutions, but it reflects a strong scientific agreement on the core issue of human-driven climate change.Do DeepSearch

Question – Do the 97% consensus papers Cook, Doran, Anderegg, Oreskes, and others report or evaluate whether there is a consensus that climate change is dangerous.

The 97% consensus papers, including Cook et al. (2013), Doran and Zimmerman (2009), Anderegg et al. (2010), and Oreskes (2004), primarily focus on establishing scientific agreement that global warming is occurring and is primarily caused by human activities, particularly greenhouse gas emissions. These studies do not explicitly evaluate or report a consensus on whether climate change is “dangerous” as a specific claim. Here’s a breakdown:

• Cook et al. (2013): Analyzed abstracts of peer-reviewed papers and found 97% of those expressing a position endorsed human-caused global warming. The study categorized papers based on whether they attributed warming to human activity, not on assessments of the severity or “danger” of impacts.
• Doran and Zimmerman (2009): Surveyed scientists and found 97% of climate experts agreed that human activities are a significant contributor to global temperature increases. The survey questions focused on warming and its causes, not on whether the consequences are dangerous.
• Anderegg et al. (2010): Examined the expertise and publications of climate researchers, finding 97–98% agreement among actively publishing climate scientists on anthropogenic climate change. The study did not assess consensus on the dangerousness of climate change.
• Oreskes (2004): Reviewed 928 abstracts and found no disagreement with the consensus that human activities are causing global warming. The focus was on the existence and cause of warming, not on evaluating its risks or dangers.

These studies aim to quantify agreement on the reality and anthropogenic cause of climate change, not on subjective or normative judgments like “dangerous.” However, related IPCC reports (e.g., AR5, 2014) discuss risks and impacts, such as sea level rise, extreme weather, and ecosystem disruption, which imply significant threats. The “dangerous” label often appears in policy discussions (e.g., the UNFCCC’s aim to prevent “dangerous anthropogenic interference”) but is not the focus of these consensus studies.

If you’re interested in whether scientists agree on the severity of climate impacts, the IPCC reports or surveys like Bray and von Storch (2016) provide more insight into perceived risks, though they still show varied opinions on specific impacts and timelines. Do DeepSearch

Question: What are the criticisms of the Cook 97 Consensus paper, both academical and on social media and mainstream media (including the BBC) For example Professor Richard Betts tweets that Cook’s paper endorsed by Barack Obama did not support a dangerous claim. Andrew Neil on the BBC. Prof Mike Hulme at the Making Science Public blog.

The Cook et al. (2013) paper, which reported a 97% consensus among climate scientists that global warming is occurring and primarily human-caused, has faced various criticisms from academic, social media, and mainstream media sources. These critiques range from methodological concerns to debates about the paper’s implications and relevance. Below, I outline the criticisms, addressing specific references to Professor Richard Betts, Andrew Neil, and Professor Mike Hulme, as well as broader academic and media commentary.

Academic Criticisms

1. Methodological Issues:
   • Paper Misclassification: Critics like David Legates et al. (2013) argued that the Cook paper inflated the consensus by misclassifying papers. They claimed only 0.3% of the analyzed abstracts explicitly endorsed the position that humans are responsible for most global warming, suggesting the 97% figure was misleading due to broad categorization.
   • Subjective Categorization: Richard Tol (2014) criticized the paper’s methodology, particularly the subjective nature of categorizing abstracts. He argued that the process of determining whether a paper endorsed the consensus was prone to bias and lacked transparency, as raters’ interpretations could vary.
   • Inclusion of Irrelevant Papers: Some academics noted that the study included papers not directly related to climate science (e.g., on spam or heating systems), which could skew the consensus figure. This was highlighted in discussions on blogs and by critics like Mike Hulme, who called the study’s design flawed.
2. Overemphasis on Consensus:
   • Mike Hulme’s Critique: Professor Mike Hulme, in a comment on the University of Nottingham’s Making Science Public blog, described the Cook paper as “poorly conceived, poorly designed, and poorly executed,” arguing it obscures the complexities of climate issues. He suggested that focusing on a numerical consensus (97%) is irrelevant to policy debates, as it does not address what actions should be taken. Hulme later clarified that his critique was not solely about the Cook study but about the broader use of consensus in public discourse.
   • Philosophical Concerns: Hulme and others, like Ben Pile, argued that the paper oversimplifies climate science by framing it as a binary (believer vs. denier), ignoring nuances in scientific uncertainty and policy implications. This was seen as stifling debate rather than advancing it.
3. Limited Scope: Critics noted that the paper focused on whether humans cause warming but did not assess consensus on the severity or “danger” of climate change impacts. This was a key point in academic discussions, as the paper’s findings were sometimes misrepresented in public discourse as implying catastrophic outcomes, which it did not explicitly address.

Social Media Criticisms

1. Professor Richard Betts’ Commentary:
   • Richard Betts, a climate scientist, tweeted in response to the Cook paper’s endorsement by Barack Obama, clarifying that the 97% consensus does not specifically support claims of “dangerous” climate change. He emphasized that the consensus is about human causation of warming, not the extent of its impacts. This reflects a broader social media concern that the paper’s findings were overstated in public narratives to imply catastrophic consequences.
   • Betts’ comments align with discussions on platforms like X, where users have debated the paper’s methodology and its use in political arguments. For example, posts on X cited Legates et al., claiming the consensus was inflated due to flawed classifications.
2. Skeptical Narratives: On social media, particularly X, critics like @elizatino have echoed academic concerns, pointing to misclassification and the inclusion of irrelevant studies. These posts often frame the paper as part of a broader narrative to push climate alarmism, though such claims are inconclusive without further evidence.

Mainstream Media Criticisms (Including BBC)

1. Andrew Neil on BBC Sunday Politics:
   • Andrew Neil, during a 2013 BBC Sunday Politics interview with Energy Minister Ed Davey, challenged the Cook paper’s findings, questioning the 97% consensus figure. Neil argued that even skeptics who acknowledge some human influence on climate might be included in the consensus, suggesting the figure was too broad. He also made scientific errors in the discussion (e.g., about temperature trends), which were later critiqued by Dana Nuccitelli in The Guardian. Neil’s critique reflected a mainstream media tendency to question the paper’s robustness and its use in policy advocacy.
   • The BBC’s coverage, including Neil’s segment, drew attention for amplifying skepticism, with some arguing it gave undue weight to contrarian views. This sparked debates about BBC’s balance in climate reporting.
2. Broader Media Scrutiny:
   • Outlets like The Register (2009) quoted Mike Hulme criticizing the overemphasis on scientific consensus in climate debates, indirectly relevant to the Cook paper’s later prominence. Hulme argued that relying on consensus numbers expects too much from science’s ability to resolve political disputes.
   • Mainstream media often framed the Cook paper as a political tool, with critics like Ben Pile (in Spiked Online and blogs) arguing it was used to silence dissent rather than foster scientific discussion. This narrative was echoed in BBC-related discussions, where the paper was seen as part of a polarized debate.

Specific Responses to Criticisms

• Cook’s Defense: Dana Nuccitelli, a co-author, responded to critics like Neil and Pile, arguing that misinterpretations of the study stemmed from a lack of nuance. He clarified that the 97% figure applied to papers explicitly or implicitly endorsing human-caused warming, not necessarily catastrophic impacts. Nuccitelli also addressed methodological critiques, asserting the study’s robustness and alignment with other consensus studies (e.g., Doran, Anderegg).
• Hulme’s Clarification: Hulme later distanced himself from the “infamous” label for the Cook paper, noting his critique was about its use in policy debates, not a rejection of its findings. He emphasized the need for a broader discussion beyond scientific consensus.
• Betts’ Nuance: Betts’ social media posts aimed to correct public misperceptions, emphasizing that the consensus is about causation, not danger, aligning with academic calls for clearer communication.

Broader Context

• Academic Divide: While the Cook paper is widely cited, critics like Tol and Legates highlight the challenge of quantifying consensus in a field with diverse views on impacts and solutions. These debates reflect broader tensions between scientific rigor and public communication.
• Media Polarization: Mainstream media, including the BBC, often amplify these critiques to present “balance,” sometimes giving disproportionate attention to skeptical voices, as seen in Neil’s interview.
• Social Media Echo Chambers: Platforms like X amplify both defenses and critiques, with users citing studies like Legates et al. to challenge the 97% figure, though these claims often lack the nuance of academic discourse.

Conclusion

The Cook et al. (2013) paper faced academic criticism for methodological issues (misclassification, subjectivity) and its limited focus on causation rather than impacts. Social media, including Betts’ tweets, highlighted misinterpretations of the paper as implying “dangerous” climate change. Mainstream media, notably Andrew Neil’s BBC segment, questioned the consensus figure’s breadth, while Mike Hulme’s critique on the Making Science Public blog emphasized its irrelevance to policy complexity. Despite these criticisms, the paper remains a cornerstone of climate communication, though its use in polarized debates underscores the challenge of conveying scientific consensus accurately.

Note: Some criticisms, especially on social media, may reflect ideological biases rather than rigorous analysis, and claims like those in Legates et al. are debated within the scientific community. Always critically evaluate sources for accuracy and context.

End of “Unsettled Climate AI“

Blogger Note – I have been quite (very) cynical about Gen AI – So this was an experiment of about 15 minutes work (and made be added to in the next few days) And I am really impressed. It took minutes, HOWEVER it of course produces a result, of many hours of work, research by real people that know the subject, and there is quality data for Gen Ai to analyse. It did not research anything new.

I have not corrected or edited this this at all. It misses a few things, and nuance could be added, and I could do more deeper searching questions, and even potentially add a set of question to defend the papers, ‘for ‘balance’. With Gen AI – in this context (or any other, or say law) ultimately, an author has to know the subject to fact/sanity check it. It is potentially a very useful tool and will not replace anybody (it might replace what they do drudge work wise, but, and I think make people more productive)

Inspiration for this post and experiment (credit to John Cook/CARDS/Tortoise media) This post was written after a look into CARDS AI by John Cook, as I found myself in the output of Hot Air tool by Tortoise media which is based on CARDS and partnered with the University of Exeter. Where to my amusement, my tweets collated by HOT Air, gave me classifications under John Cook model as a misinformation category. But then when I put in climate scientist tweets saying exactly the same thing. (not about this topic) they duly received the exact same label… More on this another day.

So it can’t tell the ‘good guys’ from the ‘bad guys’ – whoever “they” are, and of course everyone thinks they are the “goodies”, not the “baddies”. I for one have always said that I believed I was on the side of science and scientists, but sceptical (cynical now) of climate polices, climate activists and lobbyists and the small percentage of activist academics

I may run a comparison, with other AIs for “research” purposes – Google Ai was pretty similar at first glance.

The beauty here, is that ANY reader can do there own research after reading this, and check AI themselves and explore further, and hopefully onto the original material, written by real people (who put the hours in)

Will this be the first so called ‘Climate sceptic’ AI post? (or has someone else got there first)

The label I would prefer, if we must all be labelled is ‘member of the Public’ – like everyone.

[I may come back to this later, and add more]

[Addition: I first wrote about my concerns that “97% of climate scientists” was being used as a soundbite, over a decade ago at Watts Up With That (Woods, 2012) , evaluating how both the Doran and Anderegg papers had been publicised.

What else did the '97% of scientists' say?

In part, Cook’s 97% paper, maybe a response to that WUWT post, especially as it was posted at Watts Up With that. (and very many other people questions other 97% consensus papers) The motivation of the Skeptical Science ‘researchers’/’activists’. The Consensus Project (leaked SkS forum and public SkS page) – and the leaked SkS forum resources was clearly designed to get a better 97% consensus paper and how to market it. SkS Ari Jokimaki thought talk of marketing the 97 consensus results, before they had done the research and got the results, a bit odd. (shall we say)

And to be a definitive response to criticisms of earlier papers, with the objective.

[So clearly I should be considered an opponent of Skeptical Science/John Cook (and vice versa) So please read and evaluate everything on its merits of the argument.]

A final quote……

“The Consensus Project is an example of taking a longer view – developing a long-term campaign of seeping the notion of a “strengthening consensus” into the public consciousness” – John Cook – Skeptical Science 18th January 2011