In 2013, a psychology professor reviewing Malcolm Gladwell’s latest best-selling book was critical of the author’s modus operandi:
He excels at telling just-so stories and cherry-picking science to back them.
That charge had been percolating for a while, but people were suddenly paying more attention to it, including science journalists. After the WSJ review triggered a larger debate on Gladwell, longtime science journalist Paul Raeburn weighed in at MIT’s Science Tracker (recently shuttered), a journalism watchdog site that had monitored how science was covered in the media. Raeburn picked up on the mounting criticisms of Gladwell to make some important points, such as this one:
It’s the power of narrative that makes it so dangerous: Seductive storytelling robs us of our critical skills.
I’ve periodically discussed in this space how seductive qualities have helped certain climate change narratives take hold. For example, I’ve tracked how nearly every major severe weather event has in the media become linked to or associated with climate change. That narrative gave rise to a meme called “the new normal.”
The severe weather disaster = climate change is a seductive storyline for those who want to increase attention to the very real threats posed by global warming. Whether the media has conveniently embraced that framing is a question I’ll leave for others to debate.
Another similar example of a seductive narrative for those concerned about a warming planet is the “climate wars” story, which I’ve also tracked. Here is a recent unwitting demonstration from one climate change writer on how that storyline has solidified (my emphasis):
For the last couple years, Middle East experts have pointed to the ongoing civil war in Syria as a prime example of how climate change can contribute to violent conflict. The country’s worst drought on record arrived just as widespread outrage with President Bashar al-Assad’s dictatorial regime was reaching critical mass; as crops failed, an estimated 1.5 million people were driven off rural farms and into cities. While grievances with the Assad regime are many, from economic stagnation to violent crackdowns on protesters, the impacts of the drought were likely the final straw.
The narrative in Syria fit perfectly with what many top military leaders, including at the Pentagon, were beginning to project: In parts of the world where tensions are already high, the impacts of natural disasters and competition for resources are increasingly likely to ignite violence.
Yes, and when a study recently published in the Proceedings of the National Academy of Sciences (PNAS) linked climate change to the civil war now tearing apart Syria, many in mainstream media (and in climate messaging circles) jumped all over it. The narrative fit perfectly. Global warming had “hastened,” “fueled,” or “helped trigger” the Syrian conflict, and even, according to one prominent climate communicator, “spawned” the rise of ISIS or ISIL, the group of savage Islamic militants that have taken control of and terrorized parts of Iraq.
I discussed the largely uncritical press coverage of the PNAS study in this post. It’s worth mentioning that most of the media stories did a good job emphasizing the multi-causal nature of the Syrian conflict. The study itself laid out the socio-political factors responsible for the country’s uprising. And the press coverage reflected this nuanced aspect of the study. But the authors of the study also fingered climate change-fueled drought as a “catalytic” trigger. Only a few reporters bothered to closely examine the merits of that claim.
In a follow-up post, I reviewed what a number of experts in the multi-disciplinary environmental security field were saying about the study. I also solicited responses from several of them. In short, they did not think highly of the study. Since then, additional reactions have filtered out, including this essay by Halvard Buhaug, Research Director and Research Professor at the Peace Research Institute, Oslo (PRIO), who noted:
Recent research has reported a strong effect of climate extremes on violent conflict, yet many researchers question the robustness of such a link. Some even argue the relationship between climate and conflict is so complex that it can never fully be captured and understood.
Freelance journalist Amy Westervelt, writing in the Guardian several days ago, discusses this quandary:
Any attempt by scholars over the past several years to link climate change with conflict has been hotly contested, and not just by climate deniers. Many respected conflict researchers believe that climate change is happening, that humans are contributing to it, and that it’s a big problem, but that focusing on it as a cause of war may be wrongheaded.
The problem is both scientific and social. “If you want to show that climate change has contributed to an increase in civil violence, then you need to control for other factors,” explains Andrew Solow, senior scientist at the Woods Hole Oceanographic Institute in Massachusetts. “This is a fundamental scientific principle. But it is difficult to do.”
He also expressed other misgivings about the PNAS study in an article published in Nature:
Solow worries that focusing on the effects of climate change in the hopes of preventing conflict may be a distraction. “You don’t have to cut CO2 emissions by 80% to provide clean water to poor people living in Africa or [implement] better agricultural practices,” he says. “So to me the clearest policy issue is how do we strengthen civil institutions and bring people up in their standard of living so they’re not living on the edge.”
When I was soliciting scholarly responses to the PNAS study, one came in after I published my last post on the topic. It’s from Thomas Bernauer, a Switzerland-based political scientist with an expertise in climate and environmental policy issues. Here are seven bullet points he emailed to me:
1. This PNAS paper echoes earlier claims by some policy-analysts that climate change may have been a factor in the outbreak of the Syrian civil war.
2. This claim is, therefore, not new. But is it scientifically more convincing than earlier, similar claims? A careful reading of the paper tells me: NO!
3. I trust that the geophysical (climate science) part of the paper is fine. It shows that climatic changes are likely to have contributed to the severe drought in Syria, with obvious implications for agriculture. But what I take issue with is the second part of the paper’s argument, that the drought has contributed to the outbreak of war in Syria, and that there is thus a causal effect of climate change on armed conflict in that particular case. The fact that the paper makes this climate-conflict claim very prominently up front and motivates the entire paper in those terms makes the whole paper very problematic, and it is actually a bad service to good climate science.
4. What’s the problem with this climate-conflict claim? In contrast to what the authors say studies comparing many countries over many years have NOT been able to show that there is a systematic (meaning statistically significant and substantively strong) relationship between climatic changes or climatic variability and armed conflict. Kelley et al. essentially side with findings of one single research group in this regard (Hsiang et al.), and those findings differ strongly from what many other research groups have found and have been strongly questioned on methodological grounds. Most conflict researchers worldwide agree, and this is also reflected in the IPCC reports, that any effect of climate change on armed conflict will not be direct, but will, IF IT EXISTS AT ALL, only operate via negative effects on economic performance and via migration. But I have not seen any scientific study that has come up with robust scientific evidence for such an effect.
5. Statistical comparisons of many countries over many years tell us something about average effects. So even if we DO NOT find a direct or indirect effect of climate change on armed conflict on average, could it be that Syria is an important exception and that climate change, in that single and perhaps highly unusual case, has in fact contributed to civil war there?
6. The Kelley et al [PNAS] study makes that point quite prominently up front in the paper, but when you read the rest of the paper carefully it becomes clear – judging from the authors’ own words – that this part of the argument is highly speculative. My reading of the existing evidence, and also of the authors’ evidence – if one does not get blinded by their bold claims at some points in the paper, but pays attention to the empirical evidence they present – is that the Syrian civil war was caused by a whole range of factors that have nothing to do with climate change: decades long political repression, ethnic and religious fragmentation, the worst possible agricultural and economic policy, huge influx of refugees from war-torn neighbor countries, and contagion effects of the Arab spring.
7. To be clear: There are many very good environmental and economic reasons to cut greenhouse gas emissions and move towards a carbon-free energy system. We don’t need scientifically very weak claims about climate change causing wars to motivate ambitious climate policies.
I sent this detailed critique from Bernaur, along with the other afore-mentioned criticisms, to the authors of the PNAS study. I told them that I would like to publish their response in full. Here is the email I received from Richard Seager, a climate scientist at Columbia University’s Earth Institute, and one of the paper’s coauthors:
We read the critiques and comments. For those that are reasonable and substantive, the answers may be found in the PNAS paper itself. For those that are unreasonable, the answers are to be found in a modicum of common sense.
We respect Dr. Solow’s opinion, but we are not aware of any evidence that this paper or others about climate impacts diminish awareness of the importance of poverty and poor governance as causes of civil conflict. Our argument calls these out as essential elements in the Syrian conflict.
As for the suggestion that the paper has no policy relevance [made here by one researcher I contacted], we are already aware of the paper being discussed in policy circles.
If true, this discussion of their paper in policy circles doesn’t happen in a vacuum. In other words, never underestimate the power of a narrative.