Systematic moral theory in its purest and most idealised form aims at an explanation of morality that is not only recapitulatory but projective. The aim is not just to explain all first-order moral verdicts already given, but also to predict all future moral verdicts; it is not just to give a fully cogent account of why those past verdicts were right, but also to explain, with the greatest possible certainty, what future verdicts on possible novel cases will be right, or would be right, if and when they come up. (Compare a stronger and a weaker conception of scientific explanation: the weaker conception might aim only to account (fully) for why things have happened, but deny that an equally full explanation can be given of what will happen; whereas the stronger conception will contend that both what has happened and what will happen are equally susceptible of equally complete explanation.) In either case the deal is maximum explanatory power for minimum posit: the fewer the basic axioms, and the quicker the route from those axioms to complete explanations of particular verdicts (past or future), the closer the theory comes to the systematiser's ideal.
This ideal of moral systematisation seems to be an essentially modern fixation, though maybe Spinoza, Kant, and Bentham are in one way or another historical precursors of it. Whether Sir David Ross still counts as a modern, or is by now a historical figure, is not something that I will try to adjudicate here. But certainly those writing today who call themselves Rossian pluralists, as Garrett Cullity does, are backing away from the systematiser's ideal in at least three ways.
First, as pluralists, they abjure the idea of a single unitary foundational principle for morality. Rossians recognise a variety of principles, concerns or duties, or (as here) three types of reasons, all basic, so that no one of them can be taken as a starting-point for deriving the other two.
Secondly, Rossians will be more explicit than purer systematisers that simplicity is not the only theoretical virtue. As Cullity puts it (p. 1), “My aim will be to show how a theory built on these three foundations can have enough simplicity to be explanatory, while allowing for enough nuance and detail to be plausible”; yet Cullity also hopes to go at least some way towards satisfying “Williams’ Principle” that “morality is inexhaustibly complex” (p. 2).
Thirdly, Rossians will dilute – though not abandon completely – the systematiser's ambition to provide both recapitulatory and projective explanations of morality. This retreat is consequent on the first two points. Given the possibility of conflicts between a plurality of basic “foundations”, and given the general complexity of moral life, we cannot usually go straight from a moral consideration to a moral verdict. So a gap opens up here that needs to be filled by what is variously called judgement, phronesis or intuition. Cullity makes little or no use of these terms, and has no index-entry for any of them, but he does say this (p. 2):
Aristotle's Principle. There is no usable algorithm for good moral judgement that eliminates the need for moral discernment in determining what is morally good or right.
Most of Cullity's excellent book, and especially its Part III (which contains a chapter each on paternalism, instrumentalisation and consumption), consists in an exercise of “moral discernment” in just this sense. So Cullity's explanation of morality is mostly recapitulatory rather than projective. Most of the time he is showing how his three types of foundation, plus moral discernment, explain how we have arrived at certain moral verdicts. Since moral discernment always has a non-algorithmic and therefore unpredictable role to play, it is bound to be harder, and will often be impossible, to say how the three kinds of consideration central to Cullity's view will bear on novel cases that have not yet come to view; on cases that are not simply duplicates of already-judged cases.
This suggests a calibration, and a moderation, of the moral systematiser's ambitions that we might welcome anyway. Even in science, we might say, explanation is mostly recapitulatory rather than projective – and where scientific explanation is projective, it is necessarily more tentatively offered than recapitulatory explanation. A fortiori, then, with morality. Some, including me, think it is over-ambitious to seek a metaphysics of science based on laws that eliminate ceteris paribus clauses. If so, then we might think it even more over-ambitious to seek a moral system that eliminates discernment. For moral explanation, we might suggest, our model should be not the scientific theory – on a certain idealised conception of it that is contestable anyway. It should be the legal judgment.
A legal judgment can have a recapitulatory explanation in just the sense I mean: there can be an explanation of an already-given verdict that shows how it is based on reasoning plus discernment. For almost any typical, and competent, legal judgment has two contrasting characteristics. First, it is structured: it references and deploys a variety of legal considerations that support it, but which are not just logically or otherwise trivial corollaries of the judgment itself. Secondly, it is almost always contestable: it is almost always possible to say that the judgment misuses the legal considerations that do support it, or that it fails to take proper account of other legal considerations that do not.
Hence legal explanation is typically not projective. Especially with novel cases, but often also with cases that are not particularly novel, even experienced lawyers cannot always predict which way a pending judgment will go. Yet they can usually recapitulate why it went the way it did, once they know how it has gone. If Rossian pluralists are right, then moral explanation will be like this too. (Maybe, come to that, so will scientific explanation.) One of the best things about Cullity's book is his painstaking delineation of what recapitulatory explanation looks like, in practice and in detail. This discussion – to which I cannot do anything like justice in the compass of this short review – proceeds partly via the three case-studies of Part III, and partly by Cullity's careful work, especially in Part II, to show how on his view we can derive more particular ethical views from his three types of foundation; or how such derivations fail, e.g. in the two main types of undermining that he considers in Chapters 5–6.
What then are the three independent and mutually irreducible foundations on which Cullity wants to build his pluralism? They are the three items of his title – concern, respect and cooperation. In his own words, with my labellings added in bold (p. 161):
First [concern], there is the way we treat others as patients, when we respond to our awareness of the ways they are affected by the world, for better or worse, and of our ability to influence that. Second [respect], there is the way we relate to others as agents: the responses we make to our recognition of their equal possession of the capacity for making reason-governed responses, and the interdependence of our exercise of capacity. And third [cooperation], there is the way we respond to others as partners, recognising our ability to respond to the world together. It is because there are these three basic possibilities – the possibilities of relating to you as an object to which I respond, a responder whom I join, and as someone equally governed by norms for responses – that interpersonal morality has its three foundations, giving us three fundamental moral relationships to each other.
These words signal that Cullity is not building his three-part view on any of three possible justifications, each of which might seem dubious, at least on its own – though these considerations might conceivably be allowed to add some contributory weight to a different and less dubious justification. He is not (though this was my own first impression on hearing his book's title) attempting a Parfit-style conciliatory project between the already-existing traditions of consequentialism, deontology, and contractarianism, understood as the monistic moral theories based respectively on concern, respect and cooperation. Nor is he building a view on a possible, if somewhat eccentric, interpretation of Kant's three formulations of the categorical imperative, as having to do respectively with concern (the formula of the law of nature), with respect (the formula of humanity), and with cooperation (the formula of the kingdom of ends).
Nor yet, however, is he simply picking his three foundations out of the air – commending them to us merely on the grounds of their intuitive appeal. Ross himself arguably does this, and it can be said that it is to the detriment of his view, which thereby ends up appealing to intuition twice: both to justify our weighting of the various prima facie duties when they conflict with each other, and also to introduce the list of seven, or later five, basic kinds of prima facie duty in the first place. (Compare two apparent arbitrarinesses in Aristotle's ethical thought: first the appeal to intuition to determine the mean and the balance between the virtues in particular cases; and secondly, the ungrounded list of virtues that Aristotle helps himself to in the first place.) Cullity, by contrast, builds a more robust argumentative structure, on the basis of a kind of necessity. If I read him right, his idea is that there are, and are bound to be, just his three foundations of morality, because of the necessary shape of any conceivable life together.
That Cullity's Rossian pluralism has this sort of necessity built into its foundations is certainly one of the reasons why, in my opinion, his version of Rossian pluralism is the most attractive and interesting version that has yet existed. Yet his architecture inevitably raises questions; and two of the questions, with which I will close this review, are these.
First, why just this partition of the possibilities? We could also say of our life together, and arguably with equal truth and theoretical significance, that the three fundamental possibilities for our attitudes to others are love, hate and indifference, and similarly for their attitudes to us; and that these are the necessary facts about the structure of our life together that determine what is of foundational moral significance. Or again, rather similarly, we could say that the three fundamental possibilities are that I harm others, that I benefit them and that I am neutral with respect to them; and they likewise to me; and that these are the most fundamentally determinative facts. And of course indefinitely many other possibilities exist as well. So what makes Cullity's carving-up of the moral landscape more faithful to reality than those others?
I suppose Cullity's answer to this could just be “Fine – do the work, then.” That is, he might encourage his critic with: “If you think that's a direction worth looking in, carry on and look in it, and see what you come up with.”
Maybe he would give the same reply, or retort, to my second question. This second question – which reinforces the suspicion that Cullity's tripartition is not the only one available – begins from the point there is a fourth kind of possible foundation for morality that Cullity himself considers, but does not make central to his argument in this book. In his words (p. 58):
(P) Precious objects call for protection, appreciation, and the communication of that appreciation.
(P) looks like it might be the (or a) foundation for five kinds of philosophy of value that, at least in their own right, do not get much of a look-in in this book: aesthetics, the philosophy of the environment, the philosophy of childhood, the philosophy of disability and animal ethics. Cullity's focus in this book is squarely on agents – “normal” intelligent rational adult human interacting agents. But that naturally raises the question what he might say about others (other beings, other things) that we rightly treat with concern, as patients, but with whom/which cooperation is not possible, or not in the same way, and towards whom/which respect is not appropriate, or not in the same way. (Almost the only time his book touches on environmental philosophy is a brief discussion of the demandingness of the climate crisis, pp. 218–219, in his chapter on consumption.)
I do not in the least suggest that Cullity cannot have answers to these sorts of question; but I would be very interested to know what his answers would be. Perhaps he will explore them in another book. If so, I hope to have the good luck, and the privilege, of reviewing that book too.
And next time, on time. This review unfortunately appears over half a decade after the book itself, and a word of explanation is in order. I agreed to write this review on 29 May 2021. Over the next three years Utilitas asked at least twice for a review copy to be sent to me. No review copy ever turned up, so Garrett Cullity himself posted a further copy to me, which never arrived either. I finally received the copy that I now possess in May 2024. Given that the unknown factors behind this serious delay were outside our control, no apology is in order from any of us; still, this review's late appearance is an injustice to Garrett, for which I hereby express my regret.
