Suppose that there is a God of the kind affirmed by Christianity, Judaism and Islam. What difference does that make to morality? I shall argue that the existence and actions of God make no difference to the fact that there are moral truths, but that they make a great difference to what those moral truths are.

Actions may be (objectively) morally good, bad, or indifferent. Among good actions are those which are obligatory (that is, are duties), and ones which go beyond obligation and which are called ‘supererogatory’. I am obliged to pay my debts, but not to give my life to save that of a comrade – supremely, supererogatorily, good though it is that I should do so. Likewise among bad actions, there are those which it is obligatory not to do – these are wrong actions; and there are bad actions which are not wrong, and which I call ‘infravetatory’. It is wrong to rape or steal, yet (plausibly) it is bad but not wrong (barring special circumstances) to ‘waste one's talents’ or sit by oneself watching soft pornographic DVD's. By saying that some action is ‘objectively’ good, bad or whatever, I mean that it is so, whether or not the agent believes it to be so. If I have borrowed £100 from you, I have an obligation to repay you the £100, whether or not I believe that I have that obligation. By saying that some action is ‘subjectively’ good or whatever, I shall mean that the agent believes that it is objectively good or whatever. Our subjective obligations are to fulfill what we believe to be our objective obligations. (Though the main point of this paper does not depend on this) I suggest that someone is only blameworthy for failing to perform their subjective obligations, and (at any rate normally) only praiseworthy for doing actions which are subjectively supererogatorily good, that is actions which they believe to be objectively supererogatorily good. In this paper – until the last paragraph – I shall be concerned only with the objective moral qualities of actions.

As a result of discussion and experience over many centuries humans have grown in their understanding of which kinds of actions are morally obligatory, or wrong, or whatever; that is which claims of the form ‘such and such an action is obligatory (or whatever)’ are true. It is evident to almost all of us at the beginning of the twenty first century that slavery is wrong, and so is encouraging suttee (a widow burning herself to death on her dead husband's funeral pyre), and so is killing someone just because of their race; that it is not obligatory to fight a duel to defend your honour; that it is obligatory to keep your promises at any rate when it causes you little trouble; that it is good to feed the starving and talk to the lonely, barring quite extraordinary counter-considerations. And so on, and so on. And if those of some other culture think otherwise, they are obviously mistaken – just as obviously mistaken as are solipsists and flat-earthers.

Now the moral properties (moral goodness, badness etc.) of particular actions are supervenient on their non-moral properties. What Hitler did on such and such occasions in 1942 and 1943 was morally wrong because it was an act of genocide. What you did yesterday was good because it was an act of feeding the starving. And so on. No action can be morally good or bad independently of its other properties; it is good or bad because it has certain other non-moral properties. And any other action which had just those non-moral properties would have the same moral properties. The conjunction of non-moral properties which gives rise to the moral property may be a long one or a short one. It may be that all acts of telling lies are bad, or it may be that all acts of telling lies in such and such circumstances (the description of which is a long one) are bad. But it must be that if there is a world W in which a certain action A having various non-moral properties is bad, there could not be another world W* which was exactly the same as W in all non-moral respects, but in which A was not bad. If capital punishment for murder is not bad in one world, but is bad in another world, there must be some non-moral difference between the two worlds which makes the moral difference – for example that capital punishment deters people from committing murder in the first world but not in the second world. Moral properties, to use philosophical terminology are supervenient on non-moral properties. And the supervenience must be logical supervenience. Our concept of the moral is such that it makes no sense to suppose both that there is a world W in which A is wrong and a world W* exactly the same as W except that in W* A is good. It follows that there are logically necessary truths of the form ‘If an action has non-moral properties B, C and D, it is morally good’, ‘If an action has non-moral properties D, E and F, it is morally wrong’ and so on. If there are moral truths, there are necessary moral truths – general principles of morality. I re-emphasise that, for all I have said so far, these may often be very complicated principles. All moral truths are either necessary (of the above kind) or contingent. Contingent moral truths (e.g. that what you did yesterday was good) derive their truth from some contingent non-moral truth (e.g. that what you did yesterday was to feed the starving) and some necessary moral truth (e.g. that all acts of feeding the starving are good).

Theists and most atheists alike are introduced in childhood to this common concept of morality by being shown many of the same paradigm cases – keeping promises, talking to the lonely etc are both morally good actions, and so on; and they recognize these are morally good actions in virtue of what is involved in making a promise or being lonely. If theists and atheists did not have this common understanding of what makes many particular actions morally good or bad, we would not agree so much about which kinds of action are good or bad, or be able to argue – as so often we can – about the morality of particular actions. Disputes about the morality of corporal punishment or war, for example, seldom turn on whether or not there is a God or what he has done. Hence theists and atheists may agree – as clearly they do – both about the moral status (good or bad, as the case may be) of many particular actions, and also about the reasons why those actions have the moral status that they do.

Just because necessary moral truths are necessary truths, the existence and actions of God can make no difference to them, but the existence and actions of God can make a great difference to what are the contingent truths. As creator and sustainer of the universe God brings about the circumstances which (in virtue of some necessary moral truth) make an action of some kind good or bad – by making it the case that in some society capital punishment does not deter, (if its deterrent effect is the only feature which makes it not bad) he would make capital punishment in that society contingently bad. Among the necessary moral truths, which atheists as well as theists may come to recognise, is that it is very good to reverence the good and the wise who are truly great, and obligatory to thank and please benefactors; and (within limits) to please those benefactors on whom we depend for our very existence by obeying their commands. Parents, who are not merely biological parents but nurturing parents, have (within limits) the right to tell children to do things (to do the washing-up, or to help a neighbour with her shopping, for example), and these commands make actions obligatory which would not otherwise be obligatory. Likewise a state which provides a fairly just system of law and order has (within limits) the right to tell us to do things (pay taxes or do military service), and again its commands make actions obligatory. In virtue of the necessary truth that (within limits) people are obliged to obey benefactors of a certain kind, and a contingent non-moral truth that a benefactor of that kind has commanded us to do A, there is a contingent moral truth that we are obliged to do A.

If there is a God of the kind we are considering, he is all-good and all-wise, and truly great, and for that reason alone it is very good to reverence him. But he is also our supreme benefactor. He is so much more the source of our being than are our parents. God keeps us in existence from moment to moment, gives us knowledge and power and friends; and our other benefactors can only provide the benefits which they provide because God sustains in them the power to do so. Hence it becomes a duty to thank him abundantly, and also to obey his commands. If children have limited obligations to obey parents, humans will have obligations far less limited in extent to obey God. His command will make it contingently the case that some action which otherwise would be only supererogatorily good or morally indifferent is now obligatory; and his forbidding it will make an action contingently wrong when previously it was only infravetatorily bad or morally indifferent.

Neither parents nor God have the right to command someone to do what is wrong (in virtue of some other necessary moral truth). And there are further quantitative limits to the rights of parents over children – parents do not have the right to command children to serve them day and night; and so, beyond a certain point, parental commands would impose no obligation. I suggest that there are also quantitative limits (though necessarily ones far wider than those which constrain parents) on God's right to command us to do things. If he chooses to create free rational beings, thereby he limits the extent of his right to control their lives. It will then follow that in virtue of his perfect goodness, God will not command us to do actions outside those limits – for to command what you have no right to command is wrong.

What God does not command, he may commend. And since (perhaps up to a limit) it is supererogatorily good to please benefactors more than you are obliged to, God's commendation can make an action supererogatorily good, when it does not make it obligatory. And because being omniscient, God sees what is good and obligatory for reasons other than his command and commendation, and we do not always, he can inform us which actions are good or obligatory for such reasons. And, like human parents, he may command us to do what is obligatory anyway (e.g. keeping our just promises to other humans), and commend us to do what is good anyway. And his command and commendation can add to the obligation or goodness of the act. But, if what I have written earlier is correct, there are limits to what God can make to be good or obligatory. And because of the quantitative limits to the obligations which God can impose on us, there is scope for ‘works of supererogation’ as the Catholic tradition has maintained in contrast to classical Protestantism.

In Plato's dialogue Euthyphro, Socrates asked the famous question: ‘Is that which is holy loved by the gods because it is holy, or is it holy because it is loved by the gods?’ Put in monotheistic terms (and phrased simply in terms of command and obligation), the Euthyphro dilemma becomes: does God command what is obligatory for other reasons, or is what is obligatory obligatory because God commands it? Kant gave the simple answer of taking the first horn of this dilemma; other thinkers in the Christian tradition (perhaps William of Ockham, and certainly Gabriel Biel) have taken the second horn. But the view which I am putting forward takes the first horn form for some obligations and the second for others. I suggest that we ought not to rape, or break a just promise (that is one which we had the right to make), whether or not there is a God; here God can only command us to do what is our duty anyway. By contrast only a divine command would make it obligatory to join in communal worship on Sundays rather than Tuesdays. That there are very general principles of morality not dependent on the will of God, including not only the principle of the obligation to please benefactors but other principles as well, was recognised by both Thomas Aquinas and Duns Scotus. Aquinas held that ‘the first principles of natural law are altogether unalterable’ (Summa Theologiae Ia. 2ae. 94.5.) He does not tell us much in the Summa Theologiae about which these principles are, but he does write that they are principles too general to be mentioned in the Ten Commandments, principles such as that ‘no one ought to do evil to anyone’, which he says are ‘inscribed in natural reason as self-evident.’ (Summa Theologiae Ia. 2a. 100.3.) Scotus tells us that the only moral obligations from which God could not dispense us are the duties to love and worship God himself, and the duty not to worship idols, which he sees as constituted by the first three of the Ten Commandments. (Ordinatio III suppl. dist. 37.) And while both writers held – and, I have claimed, were right to hold – that there are necessary moral truths independent of the will of God, they also held that there are many contingent moral truths brought about by the commands or other acts of God.

But, although obviously God has good reason to inform us of those moral truths which hold independently of his will but which we are not clever enough to discover, what reason would he have for adding to our moral burdens by issuing commands? Three reasons I suggest. First, to give us further motivation to do what is obligatory anyway. As I have noted, parents often tell their children to do what they ought to do anyway – sometimes no doubt because children may not realise what they ought to do anyway; but on other occasions, when children do realise this, to reinforce the obligation, to make doing the act obligatory for two different reasons. Parents care that their children do what they ought to do (for reasons other than the parent's command). So if there is a God, does God. It does not need God to command us not to murder in order for it to be wrong to murder, but his command may add to our motivation not to murder. Secondly, God may issue commands for the purpose of coordination. Often we can only attain good goals which we have an obligation to promote if the actions of each of us are coordinated with those of others. We have an obligation to avoid crashing each other's cars, and to enable us to fulfil this obligation the state lays down a coordinating rule such as ‘always drive on the left’. If God founds an institution which he permits us to join (for example, marriage or the church) where membership involves an obligation to show loyalty to the institution, he needs to tell us how we are to show that loyalty – for example, whose decisions we should obey where there is a dispute (irresoluble by discussion) about what members of the institution should do. And thirdly God may issue commands in order to get us into the habit of doing what would otherwise be only supererogatorily good. When children are young, parents often command them to do such acts as doing the shopping for a neighbour, for this reason. Commands often have more effect than good advice but once children get into the habit of doing (what would otherwise be) supererogatory good acts, the need for command diminishes. God rightly want humans to be holy, and so he has this third reason for imposing obligations on us (by way of commands) – to make it come natural to us to do supererogatory good acts. For example, he might command everyone to do heroic acts of a certain kind which would not otherwise be obligatory, in order to serve as examples to those who are tempted not to fulfil obligations of a similar kind. Consider those whose marriage has for a while not been very successful but is such that its difficulties are clearly not beyond repair. Plausibly they have some obligation to each other, to their children, and to society not immediately to get divorced. Then God might command those whose marriages have much more serious problems not to get divorced, in order to encourage those with lesser problems to persist in their efforts to solve them. That would make it obligatory for some couples not to divorce when otherwise it would be at most a supererogatory good that they should not do so.

So (if there is a God) God has reasons to command us to do various acts, and his command to do them would impose on us an obligation to do them. But we need a revelation well-authenticated by a divine signature in order to know what God has commanded. Such a signature would be provided by a miracle (involving a violation of natural laws which God alone can bring about) accompanying the teaching of some prophet who purports to tell what God has commanded, such as the resurrection of that prophet from the dead fulfilling and forwarding the prophet's teaching. So, if there is a God, we cannot be objectively good without obeying his commands. But – given my suggestion at the beginning of the paper, we can still be subjectively good, without a stain on our characters that is, without obeying God's commands – so long as we do not believe that there is a God, or that he has issued certain commands to us. And if there is no God, clearly we can still be both objectively and subjectively good.