{"id":87245,"date":"2026-07-15T16:09:33","date_gmt":"2026-07-15T15:09:33","guid":{"rendered":"https:\/\/www.lse.ac.uk\/granthaminstitute\/?post_type=news&p=87245"},"modified":"2026-07-15T16:09:33","modified_gmt":"2026-07-15T15:09:33","slug":"scientific-uncertainty-in-forward-looking-climate-change-litigation","status":"publish","type":"news","link":"https:\/\/www.lse.ac.uk\/granthaminstitute\/news\/scientific-uncertainty-in-forward-looking-climate-change-litigation\/","title":{"rendered":"Scientific uncertainty in forward-looking climate change litigation"},"content":{"rendered":"\n

The Dutch Court of Appeal has misunderstood the nature of scientific uncertainty around emissions reduction pathways in the Milieudefensie v. Shell<\/em> case, argue Noah Walker-Crawford, Elbert de Jong and Valentin Jahn.<\/p>\n\n\n\n

A growing number of legal cases now seek to compel major fossil fuel companies to reduce their greenhouse gas emissions. Unlike damages claims for past harm, these forward-looking cases ask courts to intervene before the worst impacts of climate change materialise, drawing on human rights and tort law to argue that large companies bear a responsibility to cut emissions in line with international climate goals. These cases are ambitious in their intent: if successful, they could reshape how the largest greenhouse gas emitters plan their transitions to low-carbon energy.<\/p>\n\n\n\n

But these cases face challenges. The German Bundesgerichtshof (Federal Court of Justice) held in a case against BMW and Mercedes that it lacks the institutional competency to set carbon budgets for individual companies. In contrast, other European courts have ruled that<\/path><\/svg><\/a> they have the institutional capacity to assess the reduction obligations of individual companies. Yet, courts that accept that a company has a duty to reduce emissions in principle face difficulties when asked to specify by how much and how fast, citing scientific uncertainty. That pattern has played out in Milieudefensie v. Shell<\/em><\/a>, which has examined whether Shell has a civil law obligation to reduce its emissions. The Court of Appeal of The Hague recognised<\/path><\/svg><\/a> Shell’s forward-looking duty of care in a 2024 judgment, but declined to set a specific reduction target for downstream Scope 3 emissions (i.e. those linked to use of Shell’s products), concluding that the science lacked sufficient consensus on a specific figure (see our commentary, ‘Scope 3 on trial<\/a>’). The case is currently on appeal at the Dutch Supreme Court.<\/p>\n\n\n\n

We argue that the Dutch court’s difficulty in setting a specific figure was not a failure of the underlying science. The same scientific framework that the court found too uncertain to set a reduction target is routinely used by policymakers and by the oil and gas industry itself to set reduction targets. Instead, we believe that the problem was a misidentification of where uncertainty lies and a limited understanding of the criteria for assessing uncertainties in legal terms. Whether<\/em> emissions pathways can inform specific corporate targets is not a question of scientific uncertainty. Rather, there is scientific uncertainty around how<\/em> specific targets are calculated and verified. That distinction matters for how courts handle the science in future.<\/p>\n\n\n\n

The Shell case: acceptance, then retreat<\/h2>\n\n\n\n

The Milieudefensie<\/em> v. Shell<\/em> case is a legal landmark. In 2021, the District Court ordered Shell to reduce its CO₂ emissions by 45% by 2030 relative to 2019 levels, including Scope 3 emissions. On appeal, among other issues, the court examined whether a reduction target could be derived from the scientific literature on emissions pathways for the oil and gas sector. In its 2024 judgment, the court did not find a “sufficient degree of consensus” on a specific reduction figure and concluded that the science was insufficient to support a binding Scope 3 target.<\/p>\n\n\n\n

The implicit assumption behind this decision is significant: the court treated the absence of consensus on a single definitive number as evidence that the science cannot support an emissions reduction duty at all. With this approach, the courts misunderstood the nature of scientific reasoning, and its inherent uncertainties and ambiguities.<\/p>\n\n\n\n

With this approach, the courts misunderstood the nature of scientific reasoning, and its inherent uncertainties and ambiguities.<\/p><\/blockquote><\/figure>\n\n\n\n

From a precautionary principle<\/path><\/svg><\/a> perspective – a key tenet of environmental law, which takes the view that a lack of full scientific certainty should not be used as a reason to postpone action – it also has cynical implications. The precautionary principle is precisely intended to prevent what happened in the Shell case: namely, that in cases of serious and irreversible damage, reference to scientific uncertainty is used to justify the failure to take any adequate risk-reducing measures at all.<\/p>\n\n\n\n

Does the court’s approach hold up against the science and how industry uses it?<\/h2>\n\n\n\n

The court’s reasoning assumes that the absence of a single definitive number means the science cannot be used to justify a legal standard at all. However, as has been argued<\/path><\/svg><\/a>, the relevant question is not whether emissions pathways can supply a precise target, but whether they can establish a minimum standard of conduct below which corporate emissions reductions are clearly inadequate. If that floor exists, courts do not need consensus on a single number. They need to at least determine whether a company’s actions fall below what the science, on any reasonable reading, indicates. To show why that floor exists, it helps to look more closely at where uncertainty in emissions pathway science lies.<\/p>\n\n\n\n

Where is the uncertainty in emissions pathway science?<\/h2>\n\n\n\n

Integrated assessment models produce emissions pathways: these are model-based explorations of how global emissions can be reduced within a given cumulative-emissions constraint. Scenario analysis relies on ranges rather than point estimates, because scenario specifications and model assumptions differ. However, the underlying factors driving variation in these ranges are not all equally uncertain, and not all equally consequential for determining minimum standards. <\/p>\n\n\n\n

Three sources of uncertainty in emissions pathway science<\/h3>\n\n\n\n
Category<\/strong><\/td>Examples<\/strong><\/td>How robust?<\/strong><\/td>Effect on required reductions<\/strong><\/td><\/tr>
Carbon budget<\/td>Remaining carbon budget, derived from the transient climate response to cumulative emissions of carbon dioxide<\/path><\/svg><\/a> (a metric that relates the primary cause of climate change, cumulative CO2<\/sub> emissions, to global mean temperature change)<\/td>The remaining carbon budget is well established as a probability distribution whose range has narrowed over successive assessments<\/path><\/svg><\/a> (and whose median estimate has changed little over the last decade<\/path><\/svg><\/a>)<\/td>High: sets the envelope within which all credible emissions pathways fall. Precautionary treatment favours carbon budgets with a 50% or higher chance of limiting global mean temperature rise by 2100<\/td><\/tr>
Temporal allocation<\/td>Assumptions about how we can remove carbon dioxide from the atmosphere using new technologies; bioenergy availability<\/td>Subject to significant and partly irreducible uncertainty<\/path><\/svg><\/a>, driven by limits in our ability to predict future deployment of technologies and land-use competition. Unlike the carbon budget, these parameters cannot be narrowed by geophysical observation alone<\/td>Moderate: affects the pace of reductions within the envelope the carbon budget has already fixed, not the necessity of those reductions


Precautionary treatment favours earlier action. Delayed reduction increases dependence on speculative large-scale CO2removal technologies<\/td><\/tr>
Technology and cost mix<\/td>Specific low-carbon technologies; relative costs; policy<\/td>Uncertain<\/path><\/svg><\/a>, but largely a function of policy design and investment decisions rather than unresolved scientific questions<\/td>Low: determines<\/path><\/svg><\/a> the price<\/em> at which the carbon constraint is met, not whether deep reductions are required<\/em><\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

The first category, the carbon budget, is well established as a probability distribution whose central estimate has been remarkably stable over at least 13 years of assessment, with the range only narrowing as we have become more confident of our estimates. That stability supports its use as a basis for legal reasoning. Additionally, the distribution is skewed: Earth-system feedbacks that are not yet fully represented in climate models, such as permafrost thaw and land carbon sink instability, would, if accounted for, shift the distribution further towards smaller carbon budgets. This means that future scientific updates are more likely to tighten the constraint rather than relax it. The carbon budget sets the envelope within which all credible pathways must fall, unambiguously requiring deep reductions in emissions. The second category, temporal allocation, is genuinely uncertain, but operates within the envelope the carbon budget has already fixed. The third category – the technology and cost parameters that dominate public debate about the energy transition – has a comparatively limited effect on aggregate emissions outcomes under a binding carbon budget.<\/p>\n\n\n\n

There may not be consensus about specific reduction percentages, but the science can establish a minimum floor below which no credible pathway falls. The uncertainty that remains is about pace and implementation, not about the necessity or approximate scale of reductions in emissions.<\/p>\n\n\n\n

The uncertainty that remains is about pace and implementation, not about the necessity or approximate scale of reductions in emissions.<\/p><\/blockquote><\/figure>\n\n\n\n

Lastly, while the carbon budget is a global constraint on cumulative emissions, the trajectories consistent with it are not uniform across regions. International Energy Agency (IEA) scenarios consistently show the high-income countries in the Organisation for Economic Co-operation and Development (OECD) decarbonising faster than the global average, reflecting their greater access to transition capital and technology. This pattern reflects a widely accepted principle in international climate law: that wealthier, historically higher-emitting countries bear a greater responsibility to act and should decarbonise faster. It means that global emissions pathways are likely to represent a conservative benchmark for companies headquartered and primarily operating in advanced economies. A minimum standard derived from global averages would, if anything, understate what is expected of such companies.<\/p>\n\n\n\n

Industry convergence confirms the framework<\/h2>\n\n\n\n

The strongest evidence that emissions pathways can be used to support corporate targets comes from the companies themselves. At least 17 major oil and gas companies<\/path><\/svg><\/a> have set targets covering Scope 3 emissions, and the majority explicitly reference Intergovernmental Panel on Climate Change (IPCC) or IEA scenarios.<\/p>\n\n\n\n

These companies can be grouped by the role scenario analysis plays in their target-setting:<\/p>\n\n\n\n