CHAPTER III

                            The Divine Command Theory

1. Moral Data.
 We are all quite familiar with moral decisions, beliefs, and judgments about particular actions, and with the general principle adduced in support of these decisions, beliefs, and judgments.  So, for example, one might decide that failing to keep a particular appointment would be morally wrong because people are obligated to keep their commitments, or that disconnecting a comatose individual from various life-support systems would be morally permissible in a particular set of circumstances would be morally permissible because doing so would simply be "letting nature take its course" and hence would not count as homicide.  As these two examples illustrate, while some moral principles are accepted as true by everyone (or almost everyone, at any rate), while others are quite controversial.  While there can be little doubt that people are, for example, morally required to keep their commitments, refrain from harming others, and respect others’ rights (even if these requirements are overridden by more stringent moral considerations in particular circumstances), a great deal of doubt surrounds issues such as euthanasia, abortion, and capital punishment.
 How (if at all) might these more difficult an complex issues be resolved?
 According to some people, we need what is sometimes called  a “moral theory”-a very general and fundamental statement that indicates what morality is about.  If we knew what morality is about, or what constitutes the basic content of morality, then might be able to determine whether such things as abortion, euthanasia, and capital punishment fit within or conform to that content.  For example, morality is about promoting the general welfare, then we could determine whether, say, abortion is moral or immoral by determining whether abortion promotes the general welfare.  Or if morality is about what God wants us to do or not to do, then we would only need to discover God’s wishes in regard to abortion, say, in order to know whether abortion is moral or  immoral.
 We cannot determine what is and what is not moral by appealing to any old theory, of course.  Rather, we need a theory that is correct-one that gives accurate information regarding the fundamental content of morality.  But how are we to determine which, of all possible theories, is the correct one?  How are we to determine, for example, whether morality is about promoting the general welfare, or is about what God wants, or is about something else entirely?  Well, how does one choose among theories outside of morality?  How does one choose among theories in the physical sciences, for example?
 One test of a scientific theory is whether it conforms to scientific data, where scientific data are pieces of information whose credibility is much higher than that of theories.  The most familiar pieces of scientific data are observational in character, and a familiar test of scientific theories consists in determining whether they conform or fail to conform to relevant observations.  So, for example, let’s suppose that some theory implies that the orbit of mercury has a particular shape, and let us also suppose that mercury’s orbit is observed to have a different shape.  Then this result counts heavily against the correctness of the theory.  Of course, if the theory did conform to this piece of data regarding mercury’s orbit, this would not by itself prove that the theory is correct although it would constitute confirming evidence.
 But how do these remarks about scientific theories and scientific data bear on the question of how to determine the correctness of moral theories?
 They help us by highlighting the idea of testing theories against relevant data.  These data are pieces of information to which theories can conform of fail to conform; and the data are so credible that, given a choice between a theory and conflicting data, the reasonable thing to do is to abandon the theory.  The question immediately arises, of course, of whether there are such things as “moral data,” and the answer is “yes.”  We noted above that there seems little question of whether people are required to keep their commitments, to refrain from harming others, and to respect others’ rights.  Properly understood, these are pieces of moral data to which any acceptable theory of morality must conform.  If some theory fails to conform to these data-if the theory implies, for example, that people are not required to keep their commitments, then the theory is seriously defective and should probably be rejected.  That is, given the choice between the theory and the data, the reasonable thing to do is choose in the data’s favor.
 The preceding remarks are open to certain misinterpretations, and these need to be cleared up before we proceed.
 First of all, the idea that people are required, say, to keep their commitments-and that this is a piece of moral data-does not mean that people are required to keep their commitments no matter what-that keeping commitments is an absolute moral requirement.  As was indicated above, a moral requirements can be overridden by some more stringent moral consideration in a particular situation.  For example, your obligation to help a friend with some project as you promised might be overridden by your obligation to care for a seriously ill family member.  Hence, to say that people are morally obligated (or otherwise morally required) to keep their commitments or to care for members of their families is not to say that they have absolute obligations to perform such actions.  Such obligations should rather be thought of as “presumptive” (or prima facie).  A presumptive requirement is a requirement “other things being equal”: if an action is presumptively required and other things are equal, then the action is required “all things considered” or “on balance.”
 A second possible misinterpretation of what has been said about moral theories and moral data, is that the idea of tenting the former against the latter is obviously the way in which to test moral theories.  This test is not being proposed here as either obviously correct or unique.  There clearly are other tests that theories must pass to be acceptable (for example, they must be internally consistent), and some particular teat might be better than the one proposed.  This much is clear, however: not all moral theories can be correct, and hence there must be criteria that provide a reasonable basis on which to choose one theory over others.  The test-theories-against-the-data approach to making this choice is being proposed as a reasonable basis for doing so, partly because of its analogy with established practice in the physical sciences.
 In any case, this is the test we will employ here.  We will also rely on a piece of moral data that is even more credible than any suggested above, namely the idea that inflicting pain on children is (presumptively) immoral.  If some theory implies that inflicting pain on children is not even presumptively immoral, then the theory is clearly defective.  We will assume, moreover, that inflicting pain on children is necessarily (presumptively) immoral: inflicting such pain cannot possibly be moral unless doing so is rendered moral by other, more stringent moral considerations-considerations that are themselves pieces of moral data.  For example, inflicting the pain of a vaccination on a child is morally permissible (and even required) if doing so is necessary to protect the child from a serious illness.
 

2. The Divine Command Theory.
 The theory referred to above according to which morality is about what God wants is commonly known as the “divine command (moral or ethical) theory”.  For now, we will interpret the theory as follows: If an action is morally required, then this is in virtue of God’s commanding that it be performed; and any action commanded by God is morally required in virtue of that command.  I say that we will interpret the theory this way “for now,” because we will consider an importantly different version of the theory later.
 A point that needs emphasizing at the outset is that the divine command theory is a philosophical theory and not a piece of religious doctrine.  It is therefore subject to scrutiny in the way that any philosophical theory is, and in a manner that is appropriate to moral theories in particular.  As was noted above, we will assume that moral theories must in general conform to moral data, and hence that this “conformity with moral data” test can be applied to the divine command theory.  That is, we will assume that if the divine command theory has implications that are incompatible with pieces of moral data such as those mentioned above, then the theory is seriously defective and should probably be rejected.
 Because the divine command theory is philosophical in character, an evaluation of the theory must proceed by way of reason rather than faith.  This is in no way to denigrate religious faith, however.  It is simply to distinguish the realm of philosophy from that of religion, and to emphasize that reason is the appropriate mode of procedure in the former even if faith is appropriate to the latter.  Since, moreover, we are engaged in moral philosophy rather than in philosophy of religion, we will not examine the issue of whether God exists.  We will assume that God does indeed exist and, to facilitate the discussion, we will adopt some assumptions about God’s nature.  Specifically, we will assume that God is a person and is therefore capable of interpersonal relationships, and also that God is perfectly moral, omniscient (all-knowing), and omnipotent (all-powerful).  As the discussion proceeds, we’ll look more closely at some of these “background assumptions,” and we will also consider others.
 We have now identified the three main components of our inquiry: the divine command theory itself, moral data, and certain background assumptions.  It is important to recognize that these elements are separate form each other.  In particular, neither moral data nor the background assumptions is a part of the divine command theory.  Rather, moral data will be used to test the theory, and we will discuss the theory and these data in the context of our background assumptions.  That is, we will consider whether the divine command theory conforms to moral data and, in doing so, we will assume that the divinity to which the theory refers is a person who is perfectly moral, omniscient, and omnipotent.

3. Is the Divine Command Theory a Correct Account of Morality?
 The piece of moral data that we will use to test the theory was mentioned above: the (necessary, presumptive) immorality of inflicting pain on children.  We will be assuming, in other words, that inflicting pain on a child is necessarily immoral unless there are more stringent moral considerations that justify doing so.
 Now, how do we use this item of moral data to test the divine command theory?
 We imagine a situation in which some person (call him Al) is commanded by God to inflict pain on a child; and we assume that, in this situation, inflicting the pain has no beneficial consequences for anyone.  So we will be assuming that, for example, inflicting pain on the child is not connected with the child’s being or becoming ill, is not necessary in order to prevent other children from suffering even greater pain, and so on.  Now, according to the divine command theory, an action is morally required if commanded by God; and we are assuming that God commands Al to inflict pain on a child.  The obvious conclusion is that Al is morally required to inflict pain on the child.  Yet we have a piece of moral data that implies that his doing so would be immoral.  It would appear, then, that the divine command theory fails to conform to our moral datum: according to the datum, inflicting pain on a child cannot possibly be morally permissible in the absence of overriding moral considerations, and yet the divine command theory evidently accommodates this possibility.  If so, the theory is seriously defective.  Indeed, the theory seems so far out of line with moral data that it should probably be rejected.
 It is important to realize that the preceding line of reasoning does not presuppose that God has ever commanded anyone to inflict pain on a child in circumstances like those we are imagining, or that any permissible action of this sort has actually occurred.  Granted, there is the story of Abraham and Isaac, in which God commands Abraham to kill Isaac; but in that story, God is merely testing Abraham, and sends an angel who prevents him from obeying the command.  In our hypothetical case, we can assume that God allows Al to implement the command and to inflict pain on the child.  Since we are assuming that such an action cannot possibly be morally permissible in the imagined circumstances, we need only consider whether the divine command theory implies that the action is possibly permissible.
 Now, given the divine command theory, Al’s action is possibly permissible only if it is possible for God to command Al to inflict pain on the child in our imagined scenario.  And it would appear that our background assumptions prevent this possibility from arising.  We are assuming, after all, that God is perfectly moral, and there is surely no way that a perfectly moral being would command someone to inflict pain on a child without justification.  It would appear, in other words, that our background assumptions make it impossible for God to command Al to inflict pain on the child: our “imagined” scenario is actually inconceivable-which implies that the action depicted in it cannot possibly be morally permissible.  If this line of reasoning is sound, then our story of Al and the child doesn’t really show that the divine command theory fails to conform to moral data.
 Note that this approach to solving the problem we are considering is not equivalent to claiming that God could not command Al to inflict pain on the child because, in doing so, God would be contradicting God’s own rules about how we should treat other people.  We will later examine the idea that God’s are transmitted by rules rather than by commands that apply to particular actions.  Right now we are assuming that the moral status of actions is individually determined to be moral by God’s respectively commanding that those specific actions be performed or not be performed.
 Let’s pause briefly to summarize:
 We developed an argument aimed at showing that the divine command theory is defective on the ground that it fails to conform to moral data.  According to this argument this failure consists in this: that the divine command theory allows for the possibility that inflicting pain on a child is morally permissible even in the absence of overriding moral considerations, and yet such an action cannot possible be morally permissible.  We then formulated a reply to this argument based on our background assumption that God is perfectly moral, and therefore could not possibly command Al to inflict pain on the child in the envisaged circumstances.  If this reply succeeds, then the divine command theory does not in fact imply that Al’s action would be morally permissible, and hence the original argument does not demonstrate that the divine command theory fails to conform to moral data.  We must now consider whether the reply does in fact succeed.
 Consider first of all the idea that our background assumptions-particularly the assumption that God is perfectly moral-implies the impossibility of God’s commanding (and allowing) Al to inflict pain on the child.  Exactly why does our assumption have this implication?  Presumably, because a perfectly moral being would not command (and allow) someone to inflict pain on a child without some very compelling justification.  That is, a perfectly moral being would not command someone to do something that is clearly immoral.  But this presupposes that there are moral standards to which God’s actions must conform, whereas, according to the divine command theory, God determines what is moral.  According to the theory, if God commands Al to inflict pain on the child, then Al’s doing so in moral in virtue of God’s command.
 To claim, therefore that God would not command Al to inflict pain on the child because God is a perfectly moral being presupposes the existence of moral standards that are independent of God’s commands; and this presupposition directly contradicts the divine command theory.  In other words, the objection to the divine command theory raised above-that the theory doesn’t conform to moral data-cannot be avoided by claiming that God could not possibly command Al to inflict pain on the child.  Of course, it is undeniably true that God, as a perfectly moral being, could not issue such a command to Al. But the explanation of why this is true (that a perfectly moral being could not command that an immoral action be performed) contradicts the divine command theory.
 The points at issue here can be put another way.
 In his dialogue Euthyphro, Plato depicts Socrates as raising the question “What is piety?”-a question that we can interpret as concerned with the nature of morality.  One answer suggested by Euthyphro is that piety consists in what pleases the gods.  In reply to this answer, Socrates asks another question: Is something pious because it pleases the gods or does it please the gods because it is pious?  Now, let’s assume that Plato’s gods are generally decent folks and, as such, are pleased when people act piously.  This assumption suggests that the gods are pleased by pious actions because they are pious-and not that the actions are pious because they please the gods.  Analogously, to say that God could not command the performance of an immoral action because doing so would be contrary to the divine nature is to imply that God could not command immoral actions because they are immoral-which is quite different from saying that actions are moral because commands them.  Now, the first of these replies is the one offered above in response to the claim that the divine command theory fails to conform to moral data, whereas the second reply is the one implied by the divine command theory.
 We seem therefore to have encountered a dilemma:
 If God commands Al to inflict pain on a child in the circumstances we have imagined, then Al’s obeying would be morally permissible according to the divine command theory-in which case the theory implies something that is incompatible with moral data.  To reply, however, that God could not issue a command because the action would be immoral and the command would therefore be incompatible with God’s nature, would be to imply that there are moral standards for which God is not responsible-and this contradicts the divine command theory.

4. The Issue of Omnipotence.
 At the end of Section 3, we described the following dilemma to which the divine command theory gives rise:
 If God commands a person to inflict pain on a child in a situation that contains no other morally relevant considerations, then the person’s obeying would be morally permissible according to the divine command theory-in which case the theory implies something that is incompatible with moral data.  To reply, however, that God could not issue a command because the action would be immoral and the command would therefore be incompatible with God’s nature, would be to imply that there are moral standards for which God is not responsible-and this contradicts the divine command theory.
 In fact, it appears that we have two dilemmas to deal with.  For if we say that God commands actions because they are moral, then we seem to be implying that there are moral standards that even God must obey.  And this evidently implies that God is not above morality and that morality is therefore above God.  If all of these implications are true, then God’s power appears to be limited, and this contradicts our background assumption that God is omnipotent.  Consider an analogy: we are accustomed to saying that no one-not even the president-is above the law; and this makes perfectly good sense because our legal system treats everyone equally.  Unlike the laws of absolute monarchies, ours don’t derive their authority from any particular individuals, and hence the require everyone to comply.  Our background assumptions presuppose inequality rather than equality, however, and they seem to imply that God is above the moral law.  If so, then it can’t be the case that God commands actions because they are moral.
 So here’s the problem: if the divine command theory is true, then actions are moral because God commands them and this leads to the conflicts with moral data described in the previous section; but if God command actions because they are moral, then this seems to place limits on God’s power-which contradicts our background assumption that God is omnipotent.  Does this problem have a solution?  To answer this question, we must look more closely at the concept of omnipotence.
 According to one interpretation, omnipotence is the ability to do anything whatever; according to another, it is the ability to do whatever is possible.  If God’s omnipotence is of the first (“unqualified”) type, then God can do the impossible, whereas if God’s omnipotence is of the second (“qualified”) type, then God cannot do the impossible.  So, for example, if God’s omnipotence is unqualified, then God can create plane triangles whose angles add up to more or less than 180 degrees and can create a person who is simultaneously both six feet tall and five feet tall.  God can do neither of these things, however, if God’s omnipotence is qualified.
 Now recall what we said above about moral data: if the fact that a type of action is (presumptively) immoral is a piece of moral data, then that type of action is necessarily (presumptively) immoral-performing an action of that type cannot possibly be (presumptively) moral.  If, therefore, God’s omnipotence is of the qualified type-if, in other words, God is able to do whatever is possible, then God’s inability to change moral data is entirely compatible with God’s omnipotence.  Only if God’s omnipotence is unqualified must we say that God things are moral because God commands them, and that God therefore has the sort of control over moral data that is incompatible with the idea that God commands things because they are moral.  As was noted above, similar results obtain outside of morality.  If God’s omnipotence is qualified, then God is omnipotent even if God cannot make a plane triangle whose angles add up to more than 180 degrees.  Only if God’s omnipotence is unqualified must we say that God can create a plane triangle whose angles don’t add up to 180 degrees.
 So which interpretation of omnipotence is the correct one?
 This is not the place to attempt to answer this question, but we can say this much: the idea that God’s omnipotence is qualified has a long and respected history within religious traditions that include belief in an omnipotent God, and we could therefore interpret this omnipotence as qualified without fear of being relegated to the fringes of such traditions.  If, moreover, we do interpret God’s omnipotence as qualified, then we have a solution to the problem described above.  That is, we could say, for example, that God would not command someone to inflict pain on a child because such an action would be immoral, and our saying this would not contradict our background assumption regarding God’s omnipotence.
 Of course, none of this would change our earlier conclusion that the divine command theory fails to accommodate moral data, and hence fails our test of acceptability for moral theories.
 

5. Divine Commands and Moral Principles.
 Suppose that the divine command theory is true.  Then particular actions are immoral or immoral according to whether God commands that they be performed or not be performed; and in order for us to know whether our actions are moral, we must know whether they are commanded by God.  Notice that this applies to any action that we are contemplating performing, and the question therefore arises of whether we have the sort of moment-to-moment contact with God that would enable us to determine the moral status of our actions.
 The point being made here can be put another way.
 To the extent that the divine command theory provides a useful guide to conduct, it presupposes that we are in constant contact with God, and that this contact provides us with accurate information about how God views the innumerable particular actions that we perform during the course of our lives.  Even assuming, however, that we can easily establish personal contact with God (through prayer, for example), the idea that we are constantly receiving information from God about the morality of our actions is extremely implausible.  So either the divine command theory is true but not very useful as a guide to conduct, or it is mistaken.
 One very natural reply to the preceding argument is that our moral decision-making doesn’t require information about how God views our particular actions: we can base our moral decisions on knowledge of general principles or rules that God lays down.  So, for example, a person can know that a particular act of disrespecting her parents is immoral without knowing how God views that very action, because she knows that divine law contains a requirement that people honor their parents.  True, questions arise regarding how this latter knowledge is obtained, but answering such questions wouldn’t require us to presuppose that we are receiving a constant flow of information about how God views each of our actions.  Before considering possible new problems that arise from our solution to the old ones, we need to be clearer about certain changes in the divine command theory that this solution implies.
 Recall that, as the divine command theory was originally formulated, it applies directly to particular actions.  It states that if an action is morally required, then this is in virtue of God’s commanding that it be performed; and any action commanded by God is morally required in virtue of that command.  The theory makes no mention of divine law or of principles or rules laid down by God.  While this might seem to be a minor point, it is actually of considerable importance.  Indeed, to accommodate the idea of moral rules that are part of divine law, we must reformulate the divine command theory.
 In its revised form, the divine command theory has two components.  The first component concerns the justification of moral rules or principles.  It states that such rules are justified if and only if they are laid down by God-if and only if they are part of divine law.  So, for example, if the rule that it’s wrong to kill is part of divine law, then this is a justified moral rule.  The divine command theory’s second component is concerned with the moral status of particular actions: it states that this moral status is determined by justified rules.  So, for example, if the rule that killing is wrong is justified, and if some particular action involves killing someone, then that action is morally wrong.  Since this version of the divine command theory has God laying down moral rules, we’ll refer to it as the “rule divine command theory.”  The original theory, which centers on God’s commanding that particular actions be performed or not performed will be referred to as the “act divine command theory.”
 On important advantage that the rule theory has over the act theory is that, unlike the latter, the former doesn’t require that we be in constant contact with God in order to acquire information we need to guide our moral decisions.  All we need to know are the rules that God has laid down; and, while questions certainly arise about how we acquire this knowledge, these questions seem easier to answer than the parallel questions that arise in connection with the act divine command theory.  We will return to this point presently.  First, however, we must consider another kind of difficulty that is associated with the rule divine command theory.
 This difficulty is, in fact, exactly analogous to the one we encountered above in connection the act divine command theory.  Recall that we imagined a case in which God commands Al to inflict pain on a child in circumstances that contain no other morally significant features.  We noted that the divine command theory (the “act” version, that is) implies that Al’s inflicting the pain would be morally right, and also that this result counts heavily against the divine command theory because in conflicts with moral data.  Notice that this difficulty doesn’t carry over directly to the rule divine command theory because, on that theory, God doesn’t issue commands regarding particular actions.  Rather, God lays down moral rules that we use to draw conclusions about the morality of particular actions.
 But now suppose that God lays down the rule “Inflict pain on children whenever you have an opportunity.”  The rule divine command theory implies that this rule is justified, and yet the rule certainly conflicts with moral data.  The rule in question is not only unjustified, it can’t possibly be justified.  It conflicts with moral data, and its being derivable from the rule divine command theory reveals that the theory is seriously defective.  Moreover, this difficulty is not avoidable by claiming that God wouldn’t lay down such a rule because-according to our background assumptions-God is perfectly moral.  Attempting to avoid the difficulty in this way should be familiar, since we encountered its exact analogue in our discussion of the act divine command theory.  As we noted in that discussion, this appeal to God’s being perfectly moral is self-defeating, since it contradicts the basic idea underlying all divine command theories-namely, that God determines morality.  To say that God wouldn’t lay down a certain rule because God is perfectly moral and a perfectly moral being wouldn’t lay down such a rule, presupposes that the rule is justified independently of its having been laid down by God; and this is incompatible with the rule divine command theory.
 

6.  Morality, Faith, and Reason.
 Let’s set the preceding discussion aside for the moment, and assume that the rule divine command theory is true.  Then our moral decisions should be guided by rules laid down by God, and
the question immediately arises of how we know what these rules are.  Typical answers to this question would point to familiar sources of religious doctrine such as religious books (like the Bible or Koran) and pronouncements by individuals (such as popes an prophets) whose views about morality are regarded as authoritative.  It is at least unclear, however, that these sources are capable of providing the raw materials that we require for making moral decisions-particularly in areas that are very complex.  Note that this is not to question the reliability of the sources in question on matters of religious doctrine.  Our concern here is exclusively with moral issues.
 Let’s begin with a moral issue that might appear at first glance to be quite straightforward, namely, the issue of homicide.  The natural assumption is that homicide is wrong, and that this information is conveyed in clear an unambiguous terms by standard religious sources.  “Thou shalt not kill” is, after all, one of the Ten Commandments, and it is easily found in the Bible.  In fact, however, the morality of homicide isn’t at all straightforward, as becomes clear on considering possible exceptions to the prohibition against homicide.  Most people believe, for example, that killing in self-defense is morally permissible; others believe that-at least under certain conditions-capital punishment, “mercy killing,” or killing in warfare is permissible.  The moral issues surrounding these three latter areas are obviously very complicated, but even the moral status of homicide in self-defense far less clear than it might appear at first glance.  It isn’t at all obvious, for example, whether people are permitted to kill others only defense of their own lives, or whether killing in defense of property, or to prevent oneself from being seriously injured is also permissible. The point of this discussion is especially well illustrated in regard to the morality of capital punishment, however.
 Because so much attention is paid to the morality of capital punishment, people tend to overlook the moral problems that arise in connection with all forms of punishment.  Thus, consider the familiar idea that capital punishment is immoral because, in executing criminals, we are doing to them exactly what we are punishing them for.  But much the same can be said about imprisoning kidnappers: we are relocating and confining them against their will for performing acts of that very kind.  So if capital punishment is immoral because it is legalized murder, imprisonment is immoral because it is legalized kidnapping.
 The fact is that punishment of all forms harms people, and there is clearly something wrong with inflicting harm on others. Recognition of this fact has led philosophers to search for and propose justifications of punishment, most of which imply that capital punishment is justifiable at least in theory even if not in practice.  One such justification grows out of the more general moral theory called “utilitarianism.”  It centers on the idea that punishment-including capital punishment-is justifiable provided that it is sufficiently effective as a deterrent to further criminal acts.  Another possible justification is called “retributivism,” which places no weight at all on deterrent effectiveness, but focuses exclusively on the idea that wrongdoers deserve to be punished.  Reasons are offered for and against these and other theories of punishment, and some of these reasons bear on the question of whether capital punishment is justified.
 The point of mentioning these various theories of punishment is to emphasize that determining whether capital punishment is justified is part of the much larger problem of justifying punishment in any form; and solving this latter problem is a very difficult business indeed.  Is it reasonable to believe that standard sources of religious doctrine are capable of dealing adequately with such complex issues?  Almost certainly not.  And consulting religious sources when looking for guidance regarding complex moral matters is therefore likely to have an unfortunate result-namely, that the complexities of these issues are ignored, and the answers that are proposed are badly oversimplified.
 This result is inevitable if the divine command theory is accepted because the theory implies that moral truth is a subcategory of religious truth; and since faith provides a perfectly appropriate route to religious truth, it must lead also to moral truth.  But let’s try for a moment to set aside the question of what morality is about-whether, in particular, it’s about what God commands-and let’s consider a moral issue that isn’t typically addressed in religious contexts even by deeply religious people.  Affirmative action is such an issue: it concerns the question of whether certain kinds of laws and policies are unjustly discriminatory, or whether they are in fact required in order to eliminate injustices in hiring, admissions, promotions, and so on.  How should we set about attempting to resolve this issue?  Well, we must surely look closely at the matter of discrimination, and determine the conditions under which it is just or unjust.  And how do we make these determinations?  By examining the concept of justice, particularly as it applies to distributions of goods like jobs and positions in medical and law schools.  Now, does it really make sense to conduct this sort of examination within a religiously doctrinal context?