What The Science Can’t Say About Climate Change

“The good thing about Science is that it’s true, whether or not you believe in it.” –Neil deGrasse Tyson

It is a comforting idea that Science is true, whether or not someone accepts it. This way of thinking, endorsed by the renowned scientist and science advocate Neil deGrasse Tyson, suggests that in an uncertain world where beliefs often clash, Science can be an arbiter of truth.

Unfortunately, there is no objective arbiter of truth. Although “Science Says” may be accurate for a limited scope of fundamental inquiries, it simply doesn’t apply to most of the complex decisions societies face today. And that is why it’s so alarming that the phrase has become a common prefix in discussions about climate change.

Science as bludgeon

It’s no secret that invoking the special authority of science can be a powerful rhetorical device, creating a contrast between the scientifically minded protagonist and their dogmatically ignorant antagonists. It’s also not surprising that politicians and political movements often exploit this tactic. “I believe in Science,” Hillary Clinton declared to thunderous applause during her 2016 speech accepting her party’s nomination to run for president. “I believe climate change is real,” she continued, “and that we can save our planet while creating millions of good-paying clean energy jobs.” The subsequent “March for Science” became a platform for opposition to the Trump administration, and the popular sign reading “In this house, we believe ... Science is real” remains a fixture in many front yards.

But what does science say about saving the planet with good-paying jobs? Or about whether Trump should be president? Almost nothing. And what does it say about how we should address the climate change problem? Only a little bit more.

First, what Science has taught us: Climate change is a real side effect of humans’ efforts to harness energy—the bedrock of our material well-being—via fossil fuel combustion. We know that there are sufficient fossil fuels around us that, if we were to burn all of them, global temperatures could potentially be raised by more than 10°C above preindustrial levels over thousands of years. We understand that this would be a magnitude and rate of change matched only by catastrophic events like the end-Cretaceous extinction, which caused the demise of the dinosaurs as well as around 75% of all species on the planet.

But we also know that human material well-being is fundamentally tied to the availability and affordability of energy, and over the past several centuries, humanity has found that the combustion of fossil fuels constitutes a particularly effective means of obtaining this energy. In the most extreme hypothetical case of halting all greenhouse gas emissions immediately (i.e., in a matter of weeks or months), we know that global economic production and trade could grind to a halt, and that the basic provision of food, water, and protection from the elements would be out of reach for large swaths of the global population.

Most people would thus agree that we would like to avoid both the consequences of the combustion of all fossil fuels as well as the consequences of eliminating all fossil fuels immediately. There’s a sweet spot, in short, somewhere in the middle.

Climate action advocates often claim that, according to The Science, we have already blown past that sweet spot and that it is therefore necessary to embark on a very rapid transition away from current energy, industrial, and agricultural systems, such that there are net zero greenhouse gas emissions by 2050 and the globe avoids 1.5°C of warming. In the corporate sustainability world, such goals are even referred to as science-based targets.

In this worldview, humanity’s apparent inability to fully adopt science-based targets is an indictment of our systems of governance—and of human nature itself. Slower-than-recommended progress on decarbonization must mean that politicians are bought off and that we as a species are selfish, myopic, and rife with science-denialism. But does The Science actually dictate the optimal speed and pathway to decarbonization? It does not.

There is no single entity named The Science to call upon to make such a prescription, and even if there were, it would not be able to objectively and definitively weigh all the pros and cons of various courses of action across different people, societies, and species over space and time in order to come to such a conclusion. Thus, the best we can hope for is what already exists: a messy process of decision-making in which mostly democratically elected leaders—influenced by a variety of stakeholders—try to determine reasonable policies and actions.

ChatGPT HAS ENTERED THE CONVERSATION...

In environmentalist discussions, the phrase “the science says” is often used as a way of referring to the body of scientific evidence and research that supports a particular environmental argument or viewpoint. Environmentalists often use this phrase to emphasize the importance of relying on credible, peer-reviewed scientific evidence when making decisions about environmental policy and to counter arguments that are not supported by the scientific community.

For example, in the context of climate change, an environmentalist might say “the science says that burning fossil fuels is causing the Earth’s temperature to rise, leading to dangerous and potentially irreversible impacts on the environment.” By using the phrase “the science says,” the speaker is implying that their argument is based on a solid foundation of scientific evidence and research, and is not simply a matter of personal opinion or political ideology.

The phrase “the science says” is used to assert the importance of evidence-based decision-making in the face of conflicting opinions and conflicting political or ideological interests. It is also used to emphasize the role of science in informing and guiding policy decisions that have important environmental consequences, and to highlight the need for transparency, accuracy, and impartiality in the use of scientific evidence in environmental debates.

What is science?

To understand why science cannot make our most important decisions for us, it’s important to recall what science is, and the different types of knowledge that fall under its umbrella. Under the strictest definitions, the scientific method involves making observations, forming hypotheses about the underlying causes of those observations, and then meticulously designing experiments to test the validity of the hypotheses. The strongest hypotheses, and those that get conferred the status of being more-or-less true, are those that have withstood all attempts to falsify them to date.

Unfortunately, the types of questions where it is feasible to design neat tests of clear hypotheses are rather narrow and tend to be limited to fundamental aspects of physics, chemistry, and biology. Outside of those types of questions, what gets called “science” usually resembles something more like the inductive inference conducted by a detective: You have some evidence and clues, and you attempt to assemble the general story that is most consistent with these lines of evidence.

This style of reasoning, however, can be weak if the practitioner is not open to letting the evidence take them to any and all conclusions. Consider the case of Galileo. His meticulous observations seemed to point to a heliocentric model of space over the geocentric model favored by the Catholic Church. Yet without irrefutable hypothesis tests, many astronomers and mathematicians of his era were just as able to defend elaborate systems of “epicycles and deferents”—circles within circles that could account for the apparent motions of the planets and other celestial bodies while keeping the earth at the center of the universe.

As more and more observations were made, it became increasingly clear that the heliocentric model was a more elegant and simpler explanation of the motions of the planets, and it largely won out on those grounds. But the lesson remains: When there are several possible interpretations of the same observations, the detective’s preferences and biases may play a large role in their conclusions. This may be built into our reasoning at a fundamental level. In fact, cognitive scientists Hugo Mercier and Dan Sperber have argued that human reasoning likely did not evolve simply to orient us towards truth but rather to convince others of our own preferred views.

In practice, this means that when we as a species (and scientists are no exception) encounter evidence that conforms to our preferred views, we tend to ask ourselves, “Can I believe this?”—and when we encounter evidence that does not conform to our preferred views, we ask ourselves, “Must I believe this?” The answer to the first question is almost always yes, and the answer to the second question is almost always no; it’s easy to come up with a reason to doubt the quality or credibility of undesirable evidence.

Understanding as much informed the work of the philosopher Carl Popper, who argued that the difference between pseudoscience and science is that pseudoscientific experiments are designed to use evidence to try to confirm hypotheses, while science experiments focus on trying to falsify them. Given our propensity to seek out evidence that confirms our hypotheses, we are all naturally pseudoscientists.

To counter that tendency, formal scientific institutions like scientific societies and journals have historically attempted to pit different perspectives against each other so that biases could be canceled out, and what emerged would be the closest possible approximation to the truth. (This is similar to the criminal justice system, where two presentations of the case are explicitly contrasted to try to arrive at the truth.)

Other checks on the scientific process have been developed over time, including the use of double-blind controlled experiments. At a more macro level, studies are subjected to peer review before publication. Anonymous experts are assigned the task of scrutinizing any given study for any flaws or limitations, and if a study withstands the review process, it is deemed sufficiently credible to be published. This is far from saying that the study has revealed the truth, of course, but it is at least a useful signal that the study has overcome a first hurdle.

So what to make of the thousands of studies that do get published? Do they amount to a truth? Here we run into another problem: The choice of what to study and how to conduct the study is left to individual researchers and research groups, which leads to a diversity of answers to slightly different questions and a general, uncertain fuzzy picture of reality. The task of combining this knowledge into a clearer overall picture is often taken up by researchers who write so-called synthesis reports. These authors are experts who evaluate and highlight what they believe to be the overarching findings of the field. But inevitably, even they must rely on their own subjective judgments to decide what research questions are worthwhile and what published studies to call attention to.

The Science cannot say how to address the climate problem in principle

All the shortcomings of scientific research collide when it comes to climate change. For the climate problem, the most well-known and widely respected assessment reports have been written by the United Nations’ Intergovernmental Panel on Climate Change, and it is the IPCC that is often cited as representing The Science that mandates limiting global warming to 1.5°C and complete global decarbonization by 2050.

However, a reader of the reports will not find any such sentiment explicitly articulated. Indeed, it would be impossible for the IPCC to credibly do so. Here’s why:

The mainstream climate change narrative can be summarized as the understanding that A) the phenomenon is real; B) it is caused by human activities; C) it has negative impacts; and D) global decarbonization over the next few decades without substantial societal disruption is plausible. The IPCC reports are structured around this narrative, with the first part of the reports addressing the reality and causes of climate change, the second part addressing the impacts and consequences, and the third part addressing potential solutions.

Although climate science is heavily constrained by the impossibility of using copies of earth to test any hypotheses about the causes and consequences of global climate change, statements A and B are essentially narrow enough to be addressed by components of the scientific method. We can test, for example, that gases like carbon dioxide absorb and remit radiation at wavelengths that cause them to have a greenhouse effect. We can observe the increase in the greenhouse effect directly using satellites and surface instruments, and we can document the enhanced greenhouse effect’s predicted consequences on temperatures through independent observations and instrumental networks. Finally, we have physics-based mathematical models that allow us to conduct simulated hypothesis tests of what the world would look like with and without increased greenhouse gas concentrations. The oft-quoted “97% scientific consensus on global warming,” applies to statements A and B.

But from there, things become less scientific. Statement C—on the negative impacts of climate change—veers from a descriptive claim of reality to a normative claim of desirability, invoking values outside of the traditional realm of science. Statement D, about the feasibility of decarbonization, involves considerations about engineering, economics, governance, and social behavior that cannot possibly be constrained by the scientific method.

In terms of the negative impacts, Statement C—changes to the climate system that are of concern—include increases in extreme heat, sea level rise, increases in floods and droughts in some regions, enhanced fire weather conditions, and possibly stronger hurricanes. It may be widely agreed—though not scientifically “proven”—that these obvious first-order negative impacts from carbon emissions outweigh the obvious first-order positive impacts (from, for example, reduced deaths from cold spells, carbon dioxide fertilization of plants, or the expansion of agricultural zones in higher latitudes).

Beyond the first-order effects, though, there is also the net effect of industrialization via fossil fuels to consider. And on that score, global warming has coincided with a major increase in human material well-being, including through increased human lifespan, increased food supply, increased clean water supply, decreased poverty, decreased child mortality, decreased occurrence of many diseases, and many other welcome changes in the human condition. Our vulnerability to climate hazards has even decreased dramatically since we first began using fossil fuels.

And then there is the feasibility of decarbonization, Statement D. Alternatives to fossil fuels are, of course, available, but much research, including my own, highlights that there are major costs associated with the transition away from fossil fuels. One of my studies, for example, showed that the net economic effect of meeting the 1.5°C Paris Agreement temperature target would be a substantial loss of global GDP relative to no climate policy this century. The net costs come about because transitioning from energy-dense fossil fuels back to dilute and intermittent renewable sources of energy, like solar and wind, requires more inputs in terms of land, material resources, human time, and machinery to produce the same amount of energy.

Those drawbacks would be difficult enough for wealthy countries to stomach, but consider, too, that decarbonization recommendations also typically include potential brakes on the growth of energy use for low-income countries. Given the close correlation between development and societal resilience, such plans may leave the world’s poorest more exposed to the elements when they do experience extreme weather and climate events. When we acknowledge a trade-off between more efficient poverty reduction and more warming, a strong case can be made for reducing poverty and accepting the additional warming.

A case could be made for the opposite, too. These considerations are all extremely uncertain, but the thought exercise demonstrates the enormous scope of the issue and makes it clear how unrealistic it is to expect something called The Science to be able to deliver us a packaged correct course of action.

And yet many still claim that the IPCC provides just that, particularly through its use of the 1.5°C target. But if that target didn’t spring directly from The Science, where did it even come from?

The economist William Nordhaus first made a target of 2°C prominent in the late 1970s, reasoning that such an increase was near the upper boundary of the range of global temperatures experienced in the past 10,000 years, and that staying below the upper boundary seemed technically and politically feasible. Over the 1980s and 1990s, the 2°C number solidified as a useful goal for organizing climate policy. Meanwhile, in 1992, the United Nations Framework Convention on Climate Change adopted the official objective of stabilizing global temperature at a level that would “avoid dangerous anthropogenic interference with the climate system.” In 2009, there was pressure to be more precise on what constituted dangerous anthropogenic interference, and the Copenhagen Accord adopted a definition that said “2°C.” The 2015 Paris Agreement affirmed this 2°C goal, but the more ambitious goal of 1.5°C also emerged out of some of the political negotiations among delegates.

Only after the official adoption of the 1.5°C target in the next few years did the United Nations solicit a report from the IPCC to assess its value. The resulting 2018 report compared impacts at 1.5°C and 2.0°C, finding that they were worse at 2.0°C. What the report did not do was claim that global warming becomes catastrophic beyond 1.5°C, and it certainly did not claim to have conducted an exhaustive cost-benefit analysis that determined that the optimal rate of decarbonization resulted in net zero emissions by 2050 and 1.5°C of warming.

The Science can say even less about how to address the climate problem in practice

Prescribing the path forward on emissions reductions is, in principle, outside of the realm of science. But that hasn’t stopped it from trying in practice.

Scientific research can, of course, inform our understanding of impacts and potential decarbonization pathways. But in order for this research to be maximally useful, the field has to have sufficient intellectual diversity to ensure that a wide variety of research questions are being asked and that the peer-review and knowledge synthesis processes are truly rigorous. Instead, the field is organized to select for research results that amplify the negative impacts of climate change and downplay the costs of imposing rapid decarbonization.

In part, that’s because the researchers who choose to study climate change impacts and mitigation are a group who self-selected into dedicating their professional lives to doing so. It makes sense that they would come into the field with the presumption that climate change impacts are large and severe. Meanwhile, their professional incentives are shaped by enormous pressure to be published—to inject themselves into the conversation—and they are disproportionately rewarded for publishing studies that seek out and highlight the most severe impacts they can find.

And impacts can be found anywhere you look. For example, temperature is such a fundamental variable—the speed of movement of molecules in a substance—that it can be plausibly said to have some effect on virtually everything. With sufficient data and sophisticated-enough methods, some statistically significant relationship between temperature and a negative impact can almost always be found.

Contributing to research bias, researchers generally have the freedom to change the specific research question indefinitely until a relationship is found. This and similar practices amount to what is known as p-hacking: searching through different datasets, models, methods, variables, time periods, scenarios, and so on until a desired relationship is discovered. A competent researcher can almost always make a plausible-sounding case to justify the methodological choices that ended up giving them their result. Combine this with a large amount of pressure to publish concise, positive results and to quickly move on from analyses that found no relationship—and never publish them—and you have a pretty high likelihood that the balance of the scientific product will mostly confirm the discipline’s priors.

To be sure, when a study eventually wends its way from a researcher’s laptop to the inbox of a journal, reviewers will generally ask for minimal demonstrations of the robustness of conclusions to methodological choices. But this does not prevent non-robust research from being published. That’s because researchers have unlimited time to come up with justifications for their choices and, more importantly, have unlimited chances to submit to journals. If their paper is rejected from one journal, they can tweak it and submit it to another journal with fresh reviewers. With enough persistence, the study will be published somewhere. Publication and truth are thus not synonymous—not even close.

These types of biases also pervade decarbonization research. Much published research does indeed indicate that a rapid energy transition away from fossil fuels is technically feasible and even economically desirable. But when evaluating these findings, it’s worth remembering that the vast majority of researchers in the field are highly motivated to come to precisely that conclusion. They went into the field with a desire to find pathways toward decarbonization, and most of their institutions, publishing journals, and peer reviewers share this goal. Combine goal-oriented research with the luxury of working mostly in hypothetical model worlds, and it becomes not at all surprising that there are numerous studies suggesting we could give up all fossil fuels tomorrow and walk away much the happier.

But imagine, for a moment, a different world where an army of smart, creative researchers are trying to produce studies to show that rapid decarbonization would be undesirable. It’s easy if you try. A researcher could, for example, note that many medical emergencies require swift actions to save people’s lives and that there is some additional risk of death for every minute of delay in getting a person professional medical attention. Then the researcher could note that even under the most optimistic scenarios, it takes longer to charge an electric car than to refill a gas car. They could even calculate the portion of all emergencies that would be affected by the additional delay, and that number could be converted into the additional lives lost per every new electric vehicle. Now imagine that this is just one of hundreds of ideas that researchers came up with to show the undesirability of decarbonization. Soon, journals and media headlines would be flooded with results of a similar tone.

You don’t have to imagine too hard because this is not very far from the current state of climate impact research. It results in headlines like “Global heating is cutting sleep across the world: Data shows people finding it harder to sleep, especially women and older people, with serious health impacts” when the underlying study, even taken at face value, showed that 1°C of warming was associated with a loss of only 20 seconds of sleep per night for females and 16 seconds of sleep per night for males. Other studies supposedly show the detrimental effects of climate change on plane turbulence, the taste of wine, the price of tampons, trust in political leaders, and hate speech. A lot of the field’s output appears to be exercises in brainstorming. It’s not that the literature is even wrong; it’s just that the aggregate message is only loosely constrained by the data and it is heavily influenced by the proclivities and priors of the researchers doing the writing.

Those who endorse Neil deGrasse Tyson’s view of science, or who enthusiastically participated in the March for Science or display “Science is Real” signs in their yards, may be reluctant to accept the extent to which non-objective social forces influence supposedly scientific results. But we’ve had to grapple with this fact many times before; geocentric astronomy, eugenics, alchemy, phrenology—all were “scientific” until they weren’t. This doesn’t imply that contemporary climate research is remotely equivalent to phrenology, which was heavily influenced by racist and sexist social norms, but it does highlight that all forms of knowledge creation are shaped by forces outside of pure reason and objectivity.

It’s time to discard “The Science Says”

Overall, The Science on climate change is only equipped to authoritatively address a relatively narrow set of questions—for example, approximately how much the climate has warmed since the Industrial Revolution, how much of that warming is attributable to increases in greenhouse gases, what that entails for certain weather phenomena, and other narrowly defined downstream impacts. The Science, however, cannot definitively weigh all the costs and benefits of fossil fuel use or calculate the optimal rate of decarbonization. Such calculations involve technological, economic, ethical, philosophical, and moral questions outside of anything that can be answered with the scientific method.

And so, in conversations about climate change, we need to abandon the phrase The Science Says and replace it with the more honest “I think that …,” followed by supporting evidence. This would eliminate the rhetorical sleight of hand intended to define opinions as facts, and it would make it harder to dismiss legitimate opposing arguments as misinformation.

This framework allows us to see our slow progress on decarbonization policy in a new light. The United Nations has hosted international decarbonization negotiations annually for almost 30 years, and global emissions have only possibly begun to plateau recently. Instead of seeing this as a failure of governments to acknowledge scientific fact, we can see this as the result of governments grappling with both information provided by the IPCC and countervailing forces like the infeasibility of forcibly transitioning proven energy, agricultural, and industrial systems to less proven and more costly systems in the near term. When we acknowledge that the idealized conception of The Science was always a fantasy, we can come to understand that this system in which mostly elected officials, rather than a technocratic ruling class, hold the most sway in important societal matters is actually the best system for collective decision-making we can hope for.