Foundations

Foundations

We are interested in foundational issues relevant to ethical prioritization. This involves pursuing and promoting philosophical work on questions raised by the project of prioritisation, especially where progress could plausibly make a significant practical difference and where good work might not otherwise be done. A wide range of areas of philosophy may be relevant here, including ethics, political philosophy, decision theory, epistemology, philosophy of science, and philosophy of mind and cognitive science.

Below, we present a sample of foundational or cross-cutting topics that we are interested in, with a focus on ethics, epistemology, and decision theory.

Deep Uncertainty and Unawareness

We can’t know for sure what will happen in future, and we often seem to lack a basis even for probabilistic predictions (Knight 1921; Levi 1974; Joyce 2010; Greaves 2016; Bradley 2017: 215-41). We may even be unaware of what kinds of things might happen, let alone how likely they are (Grant and Quiggin 2013; Karni and Vierø 2013; Bradley 2017: 252-61; Steele and Stefánsson 2021). These problems are exacerbated when we look beyond the proximate effects of our actions, over long time-scales or across periods of rapid technological change (Lenman 2000). How should we navigate this deep uncertainty in practice? What are the most defensible criteria or decision procedures in our actual epistemic circumstances? And are there areas or kinds of interventions where our ignorance matters less?

High-Stakes Decisions and Risk-Aversion

In principle, what we do today can have a huge counterfactual impact. But for most agents in most circumstances the probability of having such an impact may be very small. How should we weigh small probabilities of enormous benefits or harms against smaller but more certain effects (Wilkinson 2022; Beckstead and Thomas 2024; Russell 2024)? Should avoiding catastrophe have any special weight, either because of risk-aversion or for some other reason (Steel 2014; Buchak 2019)? More generally, what kinds of risk attitudes are appropriate for altruistic agents acting for the sake of others (Buchak 2017; Blessenohl 2020; Pettigrew 2024)?

Beneficence and Beyond

Even when it comes to impartial considerations of wellbeing, there are many unresolved questions in moral theory. For example, how should wellbeing be aggregated (if it should) and how should we evaluate the claims of people whose existence depends on our choices (Parfit 1984: 351-441; Greaves 2017)? We are especially interested in what difference resolving such issues would make for actual decision-situations. Taking a broader point of view, what are the values, besides wellbeing, that matter most for impartial prioritisation (Hurka 1993; Temkin 1993; Crisp and Pummer 2020; MacAskill 2026)? What difference do they make in practice? And are there deontic constraints (or other similar considerations) that significantly affect prioritization in some domains? As to the general project of prioritisation: to what extent, and in what circumstances, are different agents obliged to maximise their positive impact (Pummer 2016; Horton 2017; Sinclair 2018)?

The World at Large

Much policy-making focuses on the impact on present or near-future humans. Things could look quite different when we consider the position of the current generation in the world at large. For example, given the potential number of future people, long-term considerations could be decisive in many decision-situations. What normative views support such a conclusion (Greaves and MacAskill 2025; Mogensen 2021; Thomas 2023)? Do we have special obligations to our contemporaries and immediate descendants, and how significant are these (Parfit 1984: 485-6; Mogensen 2022; Lloyd 2026)? Democratic institutions most directly reflect the interests of present-day adults. Are there ways to design political institutions that more systematically take into account the interests of future people (Kavka and Warren 1983; Caney 2016; Karnein 2016), as well as non-human moral patients (Donaldson and Kymlicka 2011: 153-4; Shulman and Bostrom 2021)? How would such institutions fit into standard political theory?

Zooming out still further, many axiological views violate the principle of separability (e.g., Hurka 1983; Temkin 1993), making prioritisation sensitive in principle to the existence of moral patients elsewhere in the universe, beyond our causal influence. Non-causal views in decision theory may have similar implications (Jeffrey 1990; Ahmed 2014; Levinstein and Soares 2020). Are such views tenable, and is the sensitivity they introduce important in practice (Thomas 2022; Tarsney and Thomas 2024)? A distinct but connected set of issues arise from the possibility that the universe contains infinitely many moral patients. Even supposing that we can only influence finitely many of them, serious difficulties arise for ethical theory with potentially infinite populations (Vallentyne and Kagan 1997; Bostrom 2011; Arntzenius 2014; Askell 2018; Tarsney and Wilkinson 2025).

Normative Uncertainty and Disagreement

As this research agenda makes clear, there is wide scope for uncertainty and disagreement about the normative principles that underlie and delimit the project of prioritisation. How should we navigate this, as individuals, institutions, or societies (Lockhart 2000; Sepielli 2014; Hedden 2016; MacAskill, Bykvist, and Ord 2020)? Are there some options that seem robustly good across a range of moral views, and can we identify them more systematically? Many of the issues highlighted earlier have apparent analogues at the level of normative uncertainty. For example, are we bound to defer to normative theories that portray our current predicaments as especially high-stakes (MacAskill, Bykvist, and Ord 2020: 150-55)?