I have been reading Sarah Wilson’s I Eat the Stars: How to Live Fully and Beautifully in a Collapsing World, and listening to her talk about it, and I find the argument persuasive. Her claim is that we are living through the kind of systemic collapse that every complex civilization before us has undergone, and that the useful response is neither panic nor denial but a clear look at what is actually ending.
What is ending, in my reading, is not humanity. It is a particular arrangement, an economy that has to grow in order to stay upright. So I went back through the sources behind her chapter on hope to see how well that holds up. This is what I found.
The money is not aligned, even judged on its own generous terms
In August 2023 the sustainability data firm Clarity AI, working with the disclosure organization CDP, published an analysis of 23,424 investment funds holding more than $25 trillion in assets, using data to June 2023. It found that 1.5 percent were aligned with a 1.5 degrees Celsius scenario. Once emissions from suppliers and from the products those companies sell were included, none were.
Aggregated, the funds came out at an average implied warming of 2.4 degrees. Counting the fuller emissions picture, 2.6 degrees.
Two things about that number deserve to be said plainly, because they cut in opposite directions.
The first is that Clarity AI is a commercial data provider analyzing its own dataset. This is not a peer-reviewed paper, and it should not be quoted as one.
The second is the more important one, and it makes the finding worse rather than better. The method rests on implied temperature ratings built from companies’ stated emissions reduction targets. As Clarity AI puts it, the measure “does not measure actual alignment but alignment of ambition.” Companies that had disclosed no target at all were assigned a default score. So this is not a scorecard of what the world’s largest funds are doing. It is a scorecard of what the companies they hold say they intend to do, and on that generous test the money is still pointed at somewhere north of two degrees.
The offsets that were supposed to close the gap
If the underlying activity does not add up, the usual answer is offsets. In August 2023, Thales West and colleagues published an assessment in Science of 26 forest conservation offset sites across six countries, using synthetic control methods to build a counterfactual for each. They reported that most projects had not significantly reduced deforestation, and that where reductions did occur they were substantially smaller than the projects had claimed.
That paper is contested, and readers should know it. A rebuttal by Edward Mitchard and colleagues argues that the satellite data was inappropriate for the comparison, that the control areas were poorly selected, and that two calculation errors mean the proportion of credits delivering real benefit should be revised upward by 62 percent. They call for retraction or heavy revision.
Two things are worth holding onto there. The rebuttal is a preprint rather than a peer-reviewed publication, and several of its authors declare commercial interests in the sector, including employment by a company that supplies data for nature-based projects and membership of an advisory group at the standards body that certified the schemes under examination. None of that makes them wrong, and all of it belongs in the open.
And their own corrected figure moves effectiveness from 6.1 percent to 9.8 percent. In their words, that is “still far from 100%.” The most favorable available reading of the dispute is that forest offsets deliver about a tenth of what is sold.
The physical indicators are indifferent to all of it
The best annual accounting of where the climate system actually stands comes from Indicators of Global Climate Change, a project involving fifty researchers led by Piers Forster at the University of Leeds, published in Earth System Science Data to fill the gap between IPCC reports.
Their 2022 update found human-induced warming of 1.14 degrees averaged across 2013 to 2022, and 1.26 degrees in 2022 alone. The rate of that warming they describe as unprecedented, at over 0.2 degrees per decade, driven by greenhouse gas emissions running at an all-time high of around 54 billion tonnes of carbon dioxide equivalent a year across the decade.
There is a detail in that paper that I had previously only seen in opinion columns, and it belongs in the evidence rather than the commentary. Part of the acceleration comes from the weakening of aerosol cooling. We have cleaned up air pollution, which saves a great many lives, and in doing so we have removed a screen that was masking some of the warming we had already caused. Both of those things are good and true at once.
On the remaining budget, the authors are careful in a way that summaries of them often are not. Writing alongside the paper, they put the remaining carbon budget for a 50 percent chance of holding to 1.5 degrees at roughly 250 billion tonnes from the start of 2023, and described it as increasingly small and very uncertain. In the paper itself they note inherent limits to the precision with which such budgets can be calculated at all. That uncertainty is not a loophole. It cuts both ways.
Where the growth model sends the bill
The transition away from fossil fuels is not weightless, and the weight does not fall evenly.
In September 2023 Amnesty International and the Congolese organization Initiative pour la Bonne Gouvernance et les Droits Humains published Powering Change or Business as Usual?, based on interviews with more than 130 people at six industrial mining projects around Kolwezi, in the Democratic Republic of the Congo’s Lualaba province.
They documented forced evictions tied to the expansion of cobalt and copper mining for rechargeable batteries. In one neighborhood of some 39,000 people, residents reported learning their homes were to be demolished when red crosses appeared on the walls. A settlement of around 400 structures, including a school, a health facility and a church, was destroyed; the company operating the nearby project denies involvement. At another site, farmers described their fields being bulldozed under military occupation, and Amnesty documented violence against residents including sexual violence. The company concerned said it had no control over the deployment of soldiers.
This is not the fossil economy. This is the clean one. A system organized around growth does not stop extracting when it changes fuel; it extracts different things from different people.
What I think this evidence actually supports
Not that we are doomed, and not that technology will handle it.
Something narrower and harder. The financial system that requires growth is not on the path its own institutions describe, and falls short even when graded on stated intentions rather than results. Its main mechanism for buying the difference delivers a fraction of what it sells, on the most sympathetic available reading. And the physical warming is accelerating, partly for the uncomfortable reason that we succeeded in cleaning the air.
That is a set of findings about a model, not about a species.
Joseph Tainter argued decades ago in The Collapse of Complex Societies that societies grow complex to solve problems, that complexity has to be paid for in energy, and that the returns on it eventually diminish. What follows in his account is not extinction but simplification. Tainter has endorsed Wilson’s book, which I did not know when I started reading it and which struck me as significant when I found out.
That lineage is not incidental. On her podcast, Wild, Wilson has run a series on collapse, and one of those episodes is an interview with Nate Hagens about what he calls the Great Simplification, the argument that an economy built on a one-time inheritance of cheap fossil energy cannot keep growing once that inheritance thins out. Hagens takes Tainter’s structure and applies it to energy. Wilson takes it and asks what it means for a life actually being lived. The three of them are working the same seam from different ends, and I found it useful to read them together rather than separately.
The numbers above do not tell me humanity is finished. They tell me that an economy which must expand to stay standing has committed itself to something the physics will not permit, and has been marking its own homework while doing it. Those are different propositions. I have stopped hoping for the second one.
