Collective harms – harms resulting from the aggregate actions of numerous agents –challenge standard approaches to establishing moral accountability. It is commonly claimed that an agent is not accountable for something that she did not individually cause, or to which she individually could have made no difference. However, in cases of collective harm, claims are frequently made about the accountability of individual agents across the cone of causation.
This thesis develops an approach to making sense of these competing claims. I explore a range of accountability theories to identify where each succeeds (or fails) in populating the catchment of consideration – the pool of candidates for downstream moral
assessment – and in providing explanatory value. I begin with a representative version of the standard individually directly causal account. I then change the subject of accountability from
the natural person to more complex agents – joint agents and corporate agents. Next, I change the basis of accountability from causation to Kutzian participation. Finally, I set out a new
conception of participation – a model I call Participation on its Own Terms. In each case I apply the model to examples drawn from a detailed account of a significant real-world collective harm: the Global Economic Crisis of 2007-9. I argue that adopting a pluralistic (supra-model) approach to applying these models makes it possible to broaden the catchment of consideration for accountability for the crisis to a few million contributions of varying magnitude. This approach produces a breadth of coverage better aligned with standard pre-theoretic views while avoiding the trap of making everyone responsible. At the same time, the application of each model contributes substantively to our explanation of how the crisis occurred. Furthermore, this supra-model approach broadly applies to other collective harms, such as the production of inequality, structural injustice, and human-driven climate change.