What is the Role of Explainability in AI Ethics?