How a 1955 Guess Became the FALSE Foundation of Human-Caused Climate Science — and How One Paper in 2020 Cemented It as Unquestionable ‘Truth’ In 1955, Hans Suess observed that the radioactive carbon-14 in tree rings had been declining over the previous fifty years. He offered a tentative explanation. His exact words were: “The decrease can be attributed to the introduction of a certain amount of C¹⁴-free CO₂ into the atmosphere by artificial coal and oil combustion and to the rate of isotopic exchange between atmospheric CO₂ and the bicarbonate dissolved in the oceans.” Read that sentence carefully. Suess proposed two mechanisms — fossil fuel combustion and natural oceanic exchange — joined by “and,” with no claim about which one mattered more. The phrase “can be attributed to” is the language of a hypothesis, not a finding. The phrase “a certain amount” is deliberately vague. This is a scientist doing what scientists are supposed to do: offering a plausible explanation and leaving the question open. What happened next was not science. It was sociology. The subsequent literature took Suess’s sentence, deleted the second mechanism (oceanic exchange), promoted the first (fossil fuels) from “can be attributed to” to “is caused by,” and named the result “the Suess effect” — as though Suess himself had identified fossil fuels as the sole cause. He had not. He had offered a two-part conjecture with no quantitative commitment to either part. But the simplified version was easier to cite, easier to teach, easier to build models around, and easier to fund. So the simplified version won. By the 1970s, it was in the textbooks. By the 1990s, it was in the IPCC reports. By the 2000s, it was treated as so obvious that stating it without evidence was considered sufficient. Then, in November 2020, Heather Graven (leader of the Carbon Cycle research group at Imperial College London since 2013), Ralph Keeling (son of the Charles David Keeling (1928-2005)) and Joeri Rogelj published what would become the definitive statement of this ‘consensus’: “Changes to Carbon Isotopes in Atmospheric CO₂ Over the Industrial Era and Into the Future,” in Global Biogeochemical Cycles. It was designated a “Grand Challenges” paper — a flagship review intended to define the state of knowledge for an entire field. It has been cited widely, used by the IPCC, referenced by the MIT Climate Portal, and treated as the authoritative source on the isotopic evidence for human-caused CO₂ accumulation. The paper is comprehensive, technically detailed, and meticulously referenced. It is also built, from the first sentence of the abstract to the last projection in Section 8, on the unexamined Suess assumption. And it is structured in a way that makes the assumption invisible. Here is how. The architecture of a circular argument The abstract of Graven et al. (2020) contains this sentence: “The sign and magnitude of the changes are mainly determined by global fossil fuel emissions.” This reads as a finding — a conclusion drawn from evidence. It is not. It is a description of what happens inside the authors’ own model when they vary one input. The model was built with fossil fuel emissions as the dominant exogenous forcing. The model was tuned to reproduce historical isotopic data under the assumption that fossil fuels caused the observed changes. The model was then run forward under six future scenarios in which only fossil fuel emissions vary. The output, unsurprisingly, shows that fossil fuel emissions “mainly determine” the result. This is then reported in the abstract as though it were a discovery about the physical world. The circularity runs in a tight loop: Assume fossil fuels are the dominant driver of isotopic change. Build a model in which fossil fuels are the dominant driver. Tune the model to reproduce the historical record under that assumption. Vary only fossil fuels across future scenarios. Report that fossil fuels dominate the output. At no point in this chain is the assumption tested. At every point, it is reinforced. The paper never asks the one question that could break the circle: has the net isotopic input to the atmosphere actually changed in a way consistent with growing fossil fuel emissions? This question has an answer. Demetris Koutsoyiannis, in a 2024 paper using the same Scripps data that Graven et al. rely on, performed the mass-balance inversion that Graven et al. did not. He solved for the net input signature δ¹³C_I — the isotopic fingerprint of whatever is flowing into the atmosphere from all sources combined, minus all sinks. If fossil fuels were “mainly determining” the isotopic changes, this quantity would have to shift downward as cumulative emissions grew. Cumulative emissions roughly tripled between 1978 and 2022. The input signature did not move. It sat at approximately −13.2 parts per thousand, constant within noise, across four globally distributed observation stations, across 40+ years of instrumental data, and across five centuries of proxy records extending back to the Little Ice Age. Koutsoyiannis then built a two-parameter model driven entirely by natural biospheric seasonality — the annual breathing of the planet's vegetation — with zero anthropogenic term. It reproduced the entire δ¹³C record at all four stations with 98 to 99 percent explained variance. The fossil fuel term is not small. It is superfluous. Adding it does not improve the fit. What Graven et al. should have done but didn't The remarkable thing about Graven et al. (2020) is that the authors are fully aware of the tool that would have tested their premise. They discuss Keeling plots — the very method Koutsoyiannis used — in Section 6 of their paper. They describe how the Keeling plot “quantifies the isotopic signature of a CO₂ source or sink by manipulating the CO₂ and ¹³CO₂ mass balance equations so that the isotopic signature is given by the intercept or slope of a regression fit.” They cite applications of this method at local and regional scales. But they never apply it globally. They never ask what the Keeling plot intercept looks like for the entire atmosphere over the industrial period. They never check whether the net input signature has shifted. The one diagnostic that could confirm or refute their central claim — and it is a simple calculation, requiring only the data they already have — is absent from the paper. Instead, they proceed directly from assumption to model to projection, treating the assumption as established background knowledge too obvious to require verification. The biosphere hiding in plain sight Graven et al. reproduce, as their Figure 1, a diagram showing the typical δ¹³C ranges for each carbon pool that interacts with atmospheric CO₂. This diagram — which Koutsoyiannis also reproduces and analyzes — shows clearly that the terrestrial biosphere (C3 plants at −26 to −28‰, soil organic matter even more depleted) overlaps with and is frequently more isotopically light than fossil fuels (coal at roughly −24‰, oil at −26 to −30‰, natural gas at −44‰). The biosphere's annual CO₂ flux is approximately 96% of total atmospheric turnover. Fossil fuels contribute roughly 4%. The paper presents this diagram and draws the opposite conclusion from what it shows. If 96% of the CO₂ entering the atmosphere comes from natural sources that are as isotopically depleted as or more depleted than fossil fuels, then the qualitative argument — "the air is getting isotopically lighter, therefore fossil fuels must be the cause" — does not follow. The biosphere is a vastly larger source of isotopically light carbon than fossil fuel combustion. The paper's own Figure 1 demonstrates this, and the paper ignores it. How this indoctrinated a generation Graven et al. (2020) is not an ordinary paper. It is a “Grand Challenges” review — the kind of paper that defines a field's self-understanding. It is assigned to graduate students as required reading. It is cited in grant proposals as establishing the framework within which new research will be conducted. It is referenced by the IPCC as the authoritative compilation. It is linked by science communication outlets — MIT, NASA, university websites — as the definitive explanation of why carbon isotopes prove human responsibility for rising CO₂. A graduate student reading this paper in 2021 would come away with the following understanding: the isotopic evidence for anthropogenic CO₂ attribution is settled, comprehensive, and quantitative. The “Suess effect” is an established physical mechanism, not a hypothesis. The models reproduce the data. The future projections follow from the physics. The only remaining questions are about the details of future scenarios and their implications for radiocarbon dating and forensic applications. At no point would that student encounter the information that the net isotopic input signature has not changed. At no point would they learn that a simple biospheric model with zero fossil fuel term reproduces 99% of the variance in the data. At no point would they discover that Suess himself offered two mechanisms and committed to neither. At no point would they be invited to perform the one calculation — the global Keeling plot intercept — that would test the premise on which the entire paper rests. This is how indoctrination works in science. It requires only that a premise be old enough, widely enough shared, and deeply enough embedded in the methodology that it becomes invisible. The Suess attribution entered the literature in 1955 as a tentative conjecture. By 2020, it had become the load-bearing wall of a “Grand Challenges” review paper in one of the field’s most prestigious journals, stated as fact in the abstract, never tested in the body, and transmitted to a new generation of scientists as established knowledge. Koutsoyiannis tested it in 2024. It failed. The input signature is constant. The biosphere explains everything. Fossil fuels are undetectable in the data. The most authoritative paper in the field is built on an assumption that the data, when properly interrogated, do not support. The generation of scientists trained on this paper will have to reckon with that, sooner or later. The mathematics does not care how many times the assumption has been cited or how prestigious the journal that published it. Conservation of mass is not a peer-reviewed opinion. It is an accounting identity. And the account does not balance the way Graven et al. said it does.