Chapter 8 concludes the book. I first consider two hypotheses about how we ended up with underdetermination in ethics. One hypothesis refers to advanced dialectics, that is, the way in which defenders of the main traditions have modified their theories in order to avoid counterexamples. The other hypothesis points to the method of reflective equilibrium and how methodological changes might have led to a convergence between the traditions. If correct, these hypotheses help alleviate some doubts about the fact that the analogy to science has only become relevant quite recently. Finally, I return to the puzzle and explain why I take the underdetermination view to offer the best answer to it. Two features make the underdetermination interpretation preferable to its alternatives. First, it can uphold a more plausible semantics, allowing us to take moral explanatory claims at face value. Second, it can uphold the adversarial spirit of the Textbook View, making it less of a miracle why this view has been accepted by a majority of ethicists.

Chapter 5 takes stock and considers the bigger picture in normative ethics. I first make explicit the analogy to the three strategies outlined in Chapter 1 and what we can learn from this about the analogous strategies in ethics. I discuss the charge of gimmickyness against conseqentializing and deontologizing, arguing that these algorithms in ethics are more plausible than similar ones in science. Still, I acknowledge the need for case studies, arguing that, even though much has been missing so far, we have a very strong case study in Parfit and that, due to the completeness of moral theories, this might already establish more than one might think at first sight. Next, I draw a map of underdetermination in ethics, assessing each version of the thesis that has been identified for the scientific realm in Chapter 1 in terms of its status in ethics. To finish, I outlinethe three most important impacts of moral underdetermination for the future of normative moral theorizing, having to do with falsification, the role of theoretical virtues, and the justification of moral theorizing itself.

Chapter 4 looks at two recent projects that aim to establish far more radical conclusions than Parfit, so-called consequentializing and deontologizing. Proponents of these projects try to come up with a simple mechanism to produce deontically equivalent counterparts to any consequentialist or non-consequentialist theory. I first explain how the two projects work on a technical level, that is, what steps are required to achieve deontic equivalence. I then reject several interpretations of the results of these projects, from the idea that we are all consequentialists (or deontologists) to the idea that consequentialism (or deontology) turn out to be empty traditions. Finally, I introduce the underdetermination interpretation of these projects and argue why it is to be preferred to yet another interpretation, the notational variants interpretation. If consequentializing and deontologizing can be interpreted in this way, we are presented with a much more far-reaching version of moral underdetermination.

The first Chapter is the Introduction. I start with a puzzle stemming from the fact that even though the Textbook View of moral theories conceives of the main traditions as both explanatorily and extensionally incompatible, many philosophers have recently argued that much more agreement is possible. I argue that this puzzle might be best explained by the fact that just as scientific theories can be underdetermined by the evidence, so can moral theories be underdetermined by their extension. This analogy to the philosophy of science has received little attention so far, and the book provides the first comprehensive analysis of it.I characterize the main idea in some more detail by distinguishing it from several others. The idea is neither that the moral is underdetermined by the non-moral, nor that moral theories leave our particular choices underdetermined, nor that theories are underdetermined because the evidence is somehow lacking. Finally, I give an outline of the structure of the book.