EDIT April 1, 2020: updated link and slightly revised abstract to the new version

I have posted a new preprint to the PhilSci Archive:

The ‘Major Transitions in Evolution’ (MTE) framework has emerged as the dominant paradigm for understanding the origins of life’s hierarchical organization, but it has been criticized on the grounds that it lacks theoretical unity, that is, that the events that make up the category do not constitute a natural kind. I agree with this criticism, and I argue that the best response is to modify the category so that it does approximate a natural kind. Specifically, I recommend defining major transitions as all those, and only those, events and processes that result in the emergence of a new level of selection. Two sorts of changes will be required to achieve this. First, events and processes that do not meet this criterion, such as the origins of the genetic code and of human language, should be excluded. Second, events and processes that do meet the criterion, but which have generally been neglected, should be included. These changes would have the dual benefits of making MTEs a philosophically coherent category and of increasing the sample size on which we may infer trends and general principles that may apply to all MTEs.