Our perceptual relationship with the world works because we trust prior stories. We could not fully perceive a tree if we did not know (because others have told us) that it is the product of a long growth process and that it does not grow overnight. This certainty is part of our “understanding” that a tree is a tree, and not a flower. We accept a story that our ancestors have handed down to us as being true, even though today we call these ancestors scientists.
No one lives in the immediate present; we link things and events thanks to the adhesive function of memory, both personal and collective (history and myth). We rely upon a previous tale when, in saying “I,” we do not question that we are the natural continuation of an individual who (according to our parents or the registry office) was born at that precise time, on that precise day, in that precise year, and in that precise place. Living with two memories (our individual memory, which enables us to relate what we did yesterday, and the collective memory, which tells us when and where our mother was born), we often tend to confuse them, as if we had witnessed the birth of our mother (and also Julius Caesar’s) in the same way we “witnessed” the scenes of our own past experiences.
This tangle of individual and collective memory prolongs our life, by extending it back through time, and appears to us as a promise of immortality. When we partake of this collective memory (through the tales of our elders or through books), we are like Borges gazing at the magical Aleph—the point that contains the entire universe: in the course of our lifetime we can, in a way, shiver along with Napoleon as a sudden gust of cold wind sweeps over Saint Helena, rejoice with Henry V over the victory at Agincourt, and suffer with Caesar as a result of Brutus’ betrayal.
And so it is easy to understand why fiction fascinates us so. It offers us the opportunity to employ limitlessly our faculties for perceiving the world and reconstructing the past. Fiction has the same function that games have. In playing, children learn to live, because they simulate situations in which they may find themselves as adults. And it is through fiction that we adults train our ability to structure our past and present experience.
From Umberto Eco’s lecture “Fictional Protocols,” part of Six Walks in the Fictional Woods.
A New Zealand woman who is being held at a US immigration centre with her six-year-old son after they were detained crossing the Canada-US border, is being wrongly “treated like a criminal”, according to her friend and advocate.
Sarah Shaw, 33, a New Zealander who has lived in Washington state for just over three years, dropped her two eldest children to Vancouver airport on 24 July, so they could take a direct flight back to New Zealand for a holiday with their grandparents.
When Shaw attempted to re-enter the US, Immigration and Customs Enforcement (Ice) detained her and her youngest son, in what was a “terrifying” ordeal, said Victoria Besancon, Shaw’s friend who is helping to raise money for her legal fight.
“Sarah thought she was being kidnapped,” she said. “They didn’t really explain anything to her at first, they just kind of quietly took her and her son and immediately put them in like an unmarked white van.”
Ice confiscated Shaw’s phone and transported the mother and son to the Dilley immigration processing center in south Texas, many states away from her home, Besancon said. Foreign nationals caught up in the Trump administration’s immigration crackdown have similarly been transported to centres far from their homes, support networks and legal representation.
“It has been absolutely horrible,” Besancon said, adding that aside from the staff, Shaw and her son are the only English speakers, are locked in their shared bedroom from 8pm to 8am and are not allowed to wear their own clothes.
“It really is kind of like being in jail ... it has been absolutely devastating and it’s kind of barbaric.”
Shaw is on what is known as a “combo card” visa – an employment visa, which she obtained through her employment at a maximum security juvenile facility, and an I-360 visa, which can grant immigration status to domestic violence survivors.
Shaw had recently received a letter confirming her visa renewal, not realising that the I-360 element of her visa was still pending approval.
“It wasn’t until she tried to come back across the border that she realised only half of the combination card – because it’s only one physical card – had been fully approved,” said Besancon.
Border officials did not need to detain Shaw and could have filed for humanitarian parole, Besancon said. Meanwhile, all three of Shaw’s children have had their I-360 visas approved, and Besancon alleged her youngest son was therefore being detained “illegally”.
Besancon, a retired US Navy officer, said her country’s treatment of Shaw and other immigrants was appalling.
“It’s so heartbreaking now to see people who, like Sarah, are not only legal, but who are contributing to American society,” she said, adding that the situation is taking a huge financial and emotional toll on Shaw and her son.
“She gives therapy and counselling to some of our most at risk youth … and to be treated like a criminal herself has just been absolutely devastating.”
“The trauma this has already caused for her and her son may never be healed,” said Mike Yestramski, the union’s president and a psychiatric social worker at Western State hospital.
The union “vehemently opposes Ice practices” and the broader immigration policies that enable them as they contradict American values and human rights, Yestramski said.
New Zealand’s foreign affairs ministry said it is in contact with Shaw, but cannot comment further on the case due to privacy issues.
The Guardian has contacted Ice and the US embassy in New Zealand for comment.
This is going to get more attention because it happened to a wire woman, and we need to keep in mind how many people are getting worse treatment but lack the allies who can get it in the news.
I’ve been traveling with my family on vacation in Iceland for the past week, so the timing of recent events has been less than ideal. Given that I am cited in both the recent DOE report and EPA proposed rule, I thought it would be useful to address how my work was used and misused in more detail. Expect more on the broader topics covered in the report in coming weeks.
Moving away from current policy
In their proposed rule titled “Reconsideration of 2009 Endangerment Finding and Greenhouse Gas Vehicle Standards”, the EPA cites my work (among others) to argue that the original 2009 endangerment finding was too pessimistic. Specifically, they say that:
With respect to projected increases in GHG concentrations and global temperatures, the projections relied upon in the Endangerment Finding appear unduly pessimistic in light of empirical observations made after it was finalized in 2009 through 2024. The Endangerment Finding relied primarily on IPCC AR4 to predict global temperature increases between 1.8 and 4 degrees Celsius by 2100, an extremely wide and variable range that necessarily impacts the existence, extent, and severity of anticipated dangers to public health and welfare. However, as previously noted, IPCC scenarios depicting worst-case, “business as usual” assessments have been criticized as misleading (2025 CWG Draft Report; Hausfather and Peters, 2020; Burgess et al., 2021; Pielke and Ritchie, 2020), and empirical data suggest that actual GHG emission concentration increase and corresponding warming trends through 2025 have tracked the IPCC’s more optimistic scenarios.
When I initially read this, I assumed that the range of 1.8C and 4C referred to in the document referred to warming since preindustrial, as is the convention for such numbers today. However, it cites the EPA 2009 endangerment finding, which itself references warming relative to a more recent 1990 baseline rather than 2100 (as Roger Pielke Jr rather assertively pointed out).
Warming between the preindustrial baseline of 1850-1900 and 1990 was approximately 0.7C, which would make the EPA’s range approximately equal to 2.5C to 4.7C. The high end of the EPA’s range thus approximately tracks the median warming in the SSP5-8.5 scenario in 2100.
In our 2020 Nature paper critiquing the use of worst case emissions scenarios like RCP8.5/SSP5-8.5 as the most likely outcome, we said:
Finally, we suggest that climate-impact studies using models developed for AR6 should include scenarios that reflect more-plausible outcomes, such as SSP2-4.5, SSP4-6.0 and SSP3-7.0. When RCP8.5 or its successor SSP5-8.5 are deployed, they should be clearly labelled as unlikely worst cases rather than as business as usual.
More recently, I’ve suggested that SSP2-4.5 is the scenario most consistent with current policy trajectories (Hausfather 2024).
But emissions scenarios only tell us part of the story. For any given emissions scenario – such as SSP2-4.5 – there is a wide range of possible climate outcomes driven by uncertainties in both the sensitivity of the climate to our emissions due to feedbacks like cloud changes, albedo changes, water vapor, etc. (climate sensitivity) and sensitivity of biosphere and ocean carbon sinks to our emissions and their associated warming (carbon cycle feedbacks).
The figure below shows the likelihood of different 2100 global warming outcomes (relative to preindustrial), based on a climate model emulator tuned to the range of climate sensitivity and carbon cycle feedbacks in the recent IPCC AR6 report. While we cannot rule out warming approaching 4C by 2100 under this scenario, it is very unlikely (<0.5%) that we experience warming exceeding 4.5C under this emissions scenario.
But its instructive to take a look at what this emissions scenario actually entails. The figure below shows the annual CO2 emissions in the three emissions scenarios we identified as “more plausible” outcomes in our Nature piece: SSP2-4.5, SSP4-6.0, and SSP3-7.0.
It is increasingly clear that an emissions scenario like SSP3 is unlikely; emissions have plateaued over the past decade rather than continuing to rise, and the falling cost of clean energy worldwide makes a 21st century dominated by coal (which SSP3 foresees) pretty implausible.
At the same time, we should not take for granted that the world will “spontaneously decarbonize” and cut emissions by 75% by 2100 as we see in SSP2-4.5. When we refer to SSP2-4.5 as a current policy scenario, we mean just that – that its consistent with policies put in place by countries to drive deep decarbonization, as well as continued declines in clean energy costs over the 21st century. SSP2-4.5 is not a baseline scenario where the world takes no climate action or does not care about climate change (and, for that matter, neither is SSP4-6.0 with its ~40% cut in emissions by 2100).
To use current policy scenarios in order to justify the repeal of current policies (as the EPA is attempting to do) rests on a fundamentally flawed premise. If the world actively repeals climate policies – which the US proposes doing – it would potentially push us to a higher emissions scenario. Indeed, the high end uncertainty of a SSP4-6.0 scenario – which still involves large CO2 emissions reductions by 2100 – would be quite close to the upper end of the EPA’s 2009 range.
Similarly, the fact that the world has made some progress in bending down the curve of future emissions should not be used as a justification that climate change is not a problem. As we also noted in our Nature piece:
This admission does not make climate action less urgent. The need to limit warming to 1.5C, as made clear in the IPCC’s 2018 special report, does not depend on having a 5C counterpoint.
The performance of climate models
The DOE’s “A Critical Review of Impacts of Greenhouse Gas Emissions on the U.S. Climate” has an interesting take on my 2019 paper on the performance of old climate models (Hausfather et al., 2019). As readers of The Climate Brink may be aware, that paper evaluated the performance of old climate models in the years after they were published and found that old models had done quite well; 10 of the 17 models evaluated were more or less spot on with the rate of future warming (within its uncertainty range), while this increased to 14 of 17 models when accounting for mismatches between projected and observed forcings (as even a physically perfect model would be off if it misforecast human emissions of CO2 and other greenhouse gases).
However, rather than actually citing anything directly in my paper (or any of its conclusions), the authors pulled out a rather obscure figure in my supplementary materials (Figure S4) to claim that “observed atmospheric CO2 concentrations tracked the low end of the SRES range and also of subsequent IPCC scenario ranges”.
Figure S4 from my 2019 paper on the performance of old climate models.
First of all, their quoted text is incorrect, as this figure does not show CO2 concentrations compared to the 2000 SRES scenarios (indeed, all models that disagree with observed CO2 concentrations well predate the SRES scenarios). The figure they would actually want to use is this one by Glen Peters, which in fact shows historical emissions more or less in the middle of the SRES range: above B1, B2, and A1T but below A1F1, A2, and A1B.
Comparison of total CO2 emissions to scenarios through 2023. Figure from Glen Peters.
More problematically, however, their use of my Figure S4 ignored the rather more prominent Figure 1 in my paper that showed the total model forcings compared to observationally-informed forcings. If anything, older models tended to have too-low rather than too-high forcings, as most of the ones published before the mid-1980s only included CO2 and did not model future increases in other GHGs like CH4, N2O, or halocarbons.
Modeled and observationally informed trends in forcings over the model projection period. Figure 1 in Hausfather et al., 2019.
Of course, the even bigger problem is that they scoured my paper on the performance of climate models to find the one figure (deep in the supplementary materials) to reinforce the point they were trying to make, and never actually referred to the broader conclusion of the paper that old models had by-and-large performed quite well. This is indicative of a deeper problem in the DOE report: it cherrypicks figures and parts of studies to support a preconceived narrative that minimizes the risk of climate change. In this case, the actual content of my paper went counter to the narrative they were trying to present, and thus was ignored.
As Andrew Dessler has noted this is a strategy common in the legal field where the goal is to make a case for one point of view, but more foreign in the science where the search for the truth (wherever it may lie) is the driving motivation.
Deep-sixing the US national climate assessment
But the most egregious part of this whole process is the promotion of one-sided views by a small number of contrarian researchers while simultaneously suppressing the actual congressionally-mandated National Climate Assessment (NCA). The 5th NCA, which I helped author, was a process involving hundreds of scientists over three years. It went through many rounds of public reviews, agency reviews, and a review by the National Academy of Sciences. It was signed off on by 13 different federal agencies – as was the 4th NCA that was released during the first Trump administration.
All NCAs are now gone from government websites. If you go to the website of any of the NCAs you get a “This site can’t be reached” error. Indeed, the DOE’s report is strewn with dead links to the now-unavailable NCA reports.
Thats why I’ve publically called this process a farce. It would be one thing to undertake Steve Koonin’s original proposal of a “red team / blue team” exercise that he pushed for in the first Trump administration. That would give undue weight to somewhat fringe views compared to the broader scientific literature, but it would at least provide a set of counterarguments.
But providing only the red team (and a very rushed red team document written in two months with only five authors at that) while actively removing the official US National Climate Assessment shows what this actually is: a politically motivated pretext to overturn the EPA’s endangerment finding that CO2 is a pollutant, not an actual search for the truth.
There will be more on both the EPA proposed rule and the DOE’s report in the coming weeks; it takes time to assemble a group of experts broad enough to address the veritable Gish Gallop of claims in the 151-page DOE report (and the 302 page EPA proposed rule). But the broader climate science community is working on a number of efforts to provide a rebuttals.
it’s actually wild how terrified of the general public most usamericans are. like you don’t realize it if you’re someone who mostly walks and takes transit and spends a lot of time in populous public spaces but then you talk to one of the thousands of people that seemingly never set foot in any public space besides a parking garage or a starbucks and you suddenly understand why it’s so easy for fascist rhetoric about the dangerous alien to take root. this country’s median voter pretty much never interacts with strangers who aren’t their coworkers or people they met on dating apps
saw a post on instagram that was literally someone citing statistics saying public transit is one of the safest travel options out there and the comments were literally just “ummmmm op this is so ableist and misogynistic of you :) don’t you know the average public transit user is a dangerous violent criminal who wants to set you on fire :)))”
it must be so terrifying and sad to go through life convinced if you set foot outside your car in public or interact with people outside your nuclear family you’ll instantly be raped and robbed by the Evil Poors no wonder so many of these people are reactionary tar pits
Log in
reply
