There are generally understood to be three major schools of ethical thought: consequentialism, on which the morality of an action is judged based on its (expected or actual) consequences (measured by amount of suffering, pleasure, welfare or other difficult-to-quantify metric); deontology, which deals in rights and obligations and people-as-ends-in-themselves; and virtue ethics, whose focus is on what sort of person it is good to be, rather than what sort of things it is good to do.
There are variants and hybrid systems. Rule consequentialism, for example, is deontological in form — it lists rights and obligations which it is morally imperative to observe — but it derives its rules by determining what rules, if followed and/or imposed, would have the best overall consequences.
I believe the reason the three systems have survived alongside each other is that they do not, or should not, actually compete, but rather complement each other. In the course of researching and constructing arguments for my thesis (and forthcoming book) on reproductive ethics, I stumbled on a natural, intuitive way of delineating appropriate domains for consequentialism and deontology. In brief, it has to do with the nature of personal identity, particularly as described by Derek Parfit in Reasons and Persons (1984) and On What Matters (2011). By analogy, a similar delineation can be made for a domain for virtue ethics.
Explaining how the three systems can (and should) apply to complementary domains would be a book in itself, which I intend to sit down and write once the reproductive ethics one is finished. But I recently found myself having to defend the model in conversation, and I thought it appropriate to share the brief description of it I gave here.
It’s my belief that a comprehensive moral theory should be a hybrid system, which would involve consequentialist policy, deontological rules, and virtue education.
So if you’re deciding what policies to adopt, or what sort of society to have, on a scale of lots of people, you use a consequentialist approach. Public policymakers have to use some sort of scale on which lives are commensurable, or else they’d get nothing done; there’s an argument in my book as to how this can work, but the basic principle is illustrated very clearly in Caspar Hare’s 2007 paper “Voices From Another World”.
But such a system is inappropriate on the level of interpersonal interactions, for a number of reasons such as information asymmetry, limited processing ability, the action/inaction distinction and so on. Both the laws that apply to such interactions and the informal/personal/moral rules people follow in them should instead have a structure based around people-as-ends, rights and responsibilities, clear-cut obligations and so on.
This should all, of course, be taught to the members of such a society, but even more important on the intrapersonal level are the notions of virtue ethics — what kind of person is it good to be? What kind of life is good to lead? There are crucial senses in which each person must answer these questions for herself, and each person’s answers will be different; but a good moral education will give her the mental scaffolding necessary to do so, on top of teaching her about the other two systems described above.