This is the third in a series of posts about the Euthyphro dilemma. In this series, I am making a case that the Euthyphro dilemma provides the basis of a definitive objection to DCT. This case will take several posts to present fully. In part 1, I explained what the two options of the dilemma are and showed that these options are mutually exclusive. In part 2, I began to consider the problems that arise for any view that, like the divine command theory, takes the second of the two options.
In part 2 of this series, I considered the Arbitrariness Argument (AA), which serves as an objection to DCT:
Premise 1: Either God has reasons for his commands or else his commands are arbitrary.
Premise 2: If God’s commands are arbitrary, then they do not ground moral obligations (since arbitrary commands cannot ground moral obligations).
Premise 3: If God has reasons for his commands, then those reasons would be reasons for us independently of God’s commands.
Premise 4: If there are reasons for us that are independent of God’s commands, then those reasons, rather than God’s commands, ground our moral obligations.
Therefore, 5: Whether God has reasons for his commands or not, his commands do not ground moral obligations.
As I pointed out, Premises 3 and 4 are the controversial ones. In the previous post, I made a case for Premise 3. In this post, I will make a case for Premise 4. There are two aspects of Premise 4, which are important to treat separately as much as possible. They are: (A) the reasons for our actions (which, importantly, are independent of and prior to God’s commands) ground our moral obligations; and (B) God’s commands (which are grounded in these reasons) do not ground our moral obligations. I will make a separate case for each of these aspects.
Recall that part of my case for Premise 3 involved what I called the Normative Independence Thesis (NIT):
(NIT): There are normative reasons (for action) that exist independently of and prior to God’s commands.
As I argued in part 2, in order to avoid the arbitrariness problem, DCT must accept NIT. Since, in order for God’s commands to be non-arbitrary, there must be normative reasons that favor his commands, a defender of DCT who insists that God’s commands are non-arbitrary is committed to the existence of normative reasons that are independent of and prior to God’s commands.
This is a highly significant result and in addition to supporting Premise 3, it serves as an entry point into a compelling case for Premise 4. What it tells us is that normative relations are independent of divine commands. If God’s commands are to be non-arbitrary, there must be such divine-command-independent normative relations. Since moral reasons are a species of normative relation, if normative relations can exist independently of and prior to divine commands, then it is reasonable to suppose that moral reasons can as well.
We can thus make the following concise case for (A): When we are morally obligated to do something, there are strong reasons of a particular kind to do that thing. So, plausibly, being obligated is just a matter of having reasons of the right kind and strength. Sure, it is not the case that anytime that I have reasons to engage in some action, I have a moral obligation to engage in that action. But reasons come in differing degrees of strength and, plausibly, when I have a moral obligation, I have it in virtue of having reasons that are particularly strong, so strong as to be binding.
A defender of DCT might push back against this argument by observing that a moral obligation is not merely a matter of having reasons. There is a vast difference between having a reason to Φ and being morally obligated to Φ. A moral obligation is binding; it is not a mere suggestion. Further, moral obligations are always obligations to someone. Having reasons is not the same as nor sufficient for being bound to do something for the sake of another person.
These observations are correct and important; there are important differences between merely having reasons and having a moral obligation. But the differences do not amount to an insurmountable gulf. As I noted above, in my concise case for (A), reasons come in differing degrees of strength. Some reasons amount to little more than suggestions, others are so powerful as to be overriding and demanding. Some reasons concern the welfare of sentient beings; other reasons are fairly disconnected from any being’s welfare. Some reasons involve the effects that an action would have on some person(s), some reasons do not. It is plausible that we have moral obligations precisely when there are very powerful reasons to perform some action, which reasons have to do with the well-being of a person (or, possibly, any sentient being).
I will therefore make a case that, once we grant that there are reasons that exist prior to and independent of God’s commands, there is no obstacle to granting that some such reasons are strong enough and of the right sort to constitute moral obligations. Before I make this case, I need to make some preliminary points and define some terminology.
First, as I understand them, moral obligations are a special kind of norm. Moral obligations have the following features:
(i) They are binding: when we are obligated, something is required of us, it is not merely suggested or encouraged of us.
(ii) They involve reasons: when someone is morally obligated to do something, she has reasons to do that thing.
(iii) They have a social character: moral obligations involve things we owe to others (and perhaps ourselves). Obligations are always obligations to someone(s).
A reader who is looking for places to criticize my argument might want to start here. I am offering an account of the features must be present for someone to have a moral obligation. We might object that I have left some important feature of moral obligations out or included some feature that not all moral obligations possess. I encourage anyone inclined toward any such objections to let me know in the comments section.
With the above features in mind, let me stipulate some terminology, with the caveat that I am not intending, with the following definitions, to capture standard or common usage. I am stipulating how I will use these terms and am doing so solely for ease of exposition. Nothing in my argument depends on using this terminology, but I have found that this is the best way that I can make the points I want to make. Here is the terminology that I will use:
A welfare reason is one that involves the welfare of a sentient being.
A person-involving reason is one that counts in favor of or against actions that affect the well-being of some person(s).
A recommending reason is one that counts in favor of (or against) some action, but does not do so strongly.
A demanding reason is one that strongly counts in favor of (or strongly against) some action.
A moral reason is one that is either a welfare reason or a person-involving reason and that is stronger than a recommending reason.
While I will talk of demanding reasons and recommending reasons, we should not assume that this is a simple dichotomy. Rather, the strength of reasons exists of a spectrum; on one end are reasons that merely suggest (without requiring an action or being binding) and on the other end are reasons that are so strong as to make demands of us.
I will also understand a moral obligation as involving an all-things-considered ‘ought.’ In other words, when we say of someone that she is morally obligated to do something, we are making a judgement that takes all relevant factors into consideration; i.e., it is a judgement that, given present circumstances and all relevant considerations, this is what she ought to do. Claims that a person has (some) moral reason(s) to perform some act, I take to be (at least in most circumstances) pro tanto judgements. They are claims that, to some extent, a person ought to do something; but they are not all-things-considered judgements. As such, a claim that there are some moral reasons to do something is not the same as a claim that it is morally obligatory to do it.
An all-things-considered judgement, as the name suggests, takes into consideration all the relevant considerations. Reasons can conflict with one another. That is, in one and the same circumstance we can often have reasons that are opposed; one reason or set of reasons counting in favor of some course of action, some other set of reasons counting against the same course of action, and yet other reasons favoring an altogether different course of action. Reasons can also support and/or enable other reasons. The way in which reasons combine and interact is quite interesting and complicated, much too complicated to fully account for here. I recommend Jonathan Dancy’s Ethics Without Principles to those readers who want to understand this issue better. What matters for this discussion is just the idea that, an all-things-considered judgement results from the (often complex interactions) of pro tanto factors.
My case for Premise 4 rests on the idea that when we have moral reasons that are sufficiently strong, which are not canceled or counterbalanced by opposing reasons, that is sufficient to make it the case that we are morally obligated. In other words, if, all things considered, a person has sufficiently strong moral reasons to do something, then, just in virtue of this, she has a moral obligation. If this is correct, then having a moral obligation is just a matter of having, all things considered, reasons of the right kind and strength. Once we grant the existence of reasons that are independent of and prior to God’s commands (as even defenders of DCT must if the arbitrariness problem is to be avoided) there is no obstacle to granting the existence of reasons of the types I described above. In particular, there is every reason to expect that there are demanding moral and person-involving reasons. But, then, in some instances in which such reasons are present, there will be people who have reasons, the force of which is not canceled by opposing reasons (that is, all things considered), that are so strong as to be binding and that concern the well-being of persons. Therefore, in such situations, these people will have moral obligations.
I think that these claims can be best supported by thinking about some imagined but realistic examples. These examples will all involve people who have reasons and we will be able to make what I take to be fairly obvious comparative claims about the strength and content of these reasons.
Sally has been given a new job offer. It involves similar work and responsibilities to her current job, but has a slightly different compensation package. In addition, the new job is five miles closer to her house than her current job.
The fact that the new job is closer to her home counts in favor of Sally’s accepting the new job. But it doesn’t count very strongly in favor; the amount of time she will save on her commute is of relatively minor importance compared to other considerations, such as how pleasant the work environment will be compared to her current job, whether she will be more fulfilled at the new job, and what her compensation will be compared to her current job. That it is not a significant factor by comparison does not mean that it is no factor at all. The relative proximity of the new job, therefore, is a reason for Sally to accept the new job even if not a very powerful one. This reason is merely a recommending reason. Note also that this reason is not obviously a welfare reason since it is not directly related to the welfare of anyone. Conceivably, Sally’s own welfare might be enhanced by saving a couple of minutes every day on her commute, but any such effect on her well-being is minor.
INCORRECT CHANGE 1
Tom is purchasing a small amount of groceries. The total cost is $29.60. He pays the cashier with two $20 bills and receives a $10 bill, a quarter and two dimes. Tom does not immediately see the error that the cashier has made, but as he is walking out of the store, he counts his change and realizes that there has been a mistake.
The fact that Tom has been given five cents more in change than he should have been given counts in favor of his going back to the cashier so that the mistake can be corrected. Thus, it is a reason for Tom to do so. But it is not a very powerful reason. Five cents is not that big of mistake and the supermarket will almost certainly overlook such an insignificant error. In addition, this reason is quite far removed from any person’s welfare. The cashier will not suffer when she realizes that her cash drawer is five cents short. She may briefly wish that it was not, but she will not suffer because of the error. Nor will Tom greatly benefit from the extra five cents he now possesses.
INCORRECT CHANGE 2
Susan is at REI purchasing supplies for her upcoming backpacking trip. She buys a new tent, backpack, sleeping bag, clothes, food, and miscellaneous supplies. Her total comes to $1138.50. She hands the cashier twelve $100 bills and receives four $20 bills, a $1 bill, and two quarters in change. Susan is in a hurry and does not notice the error until she gets back to her car, where she looks at her receipt and counts her change.
The fact that Susan has been given twenty dollars more than she should have been given counts in favor of her going back to the cashier so that the mistake can be corrected. Thus, it is a reason for Susan to do so. It is a more powerful reason that that which Tom has in INCORRECT CHANGE 1. In addition, Susan’s reason is more directly connected to some person’s welfare, namely that of the cashier who made the error. A $20 error is not tremendously significant, but it is significant enough that we can reasonably expect the cashier to suffer some (probably minor) consequences. Thus, Susan’s reason to return the $20 is a fairly strong person-involving reason. However, the reason is probably not so strong as to be demanding and it is probable that it can be outweighed by other considerations.
Barbara is taking a short hike in the local hills. A few miles into her hike she finds a wallet lying on the trail. She opens the wallet and finds $125 in cash, credit cards, and a driver’s license. There is nobody else nearby on the trail .
The fact that Barbara has discovered the wallet counts in favor of her taking it to the nearest police department so that the police can contact the owner and return it to him (or in some other way ensuring that the wallet is returned to the owner). Given the amount of money contained in the wallet, and the importance of a driver’s license and credit cards, this reason is stronger than the reasons present in the previous examples. Further, this reason is more directly related to some person’s welfare than any of the reasons present in the previous examples. If the owner does not get the wallet back, he will lose $125, be forced to cancel his credit cards, and will need to visit the DMV to get a new license. These are merely inconveniences, but they are not negligible and they will have a negative effect on his welfare. Barbara has strong person-involving reasons to turn in the wallet (and thus, she also has moral reasons to do so).
John has planned a multi-day backpacking trip through a remote part of the Arizona desert. Three miles into his hike he encounters a small child alone on the trail. The girl is clearly tired and suffering from exposure to the elements. She is sunburned, dehydrated, and malnourished. John tries to communicate with her, but the child is barely conscious and cannot answer John’s questions.
The fact that the child is suffering and will most likely die if she does not receive medical attention is a reason for John to cut his trip short and transport the child to the closest hospital. Given the amount of suffering she is experiencing and the value of her life, this reason is very powerful. Further, it is obviously directly connected to the child’s welfare. This reason is so powerful that it is difficult to imagine any factor that could outweigh or override it. John thus has a demanding, person-involving reason (thus, a moral reason) to render aid to the child. And, all things considered, this is what he ought to do.
The reasons present in DESERT HIKE are much stronger than any of the reasons present in the previous examples. Consider NEW JOB: Plausibly, if Sally ignores the fact that the new job is five miles closer than her current job, she has not done anything morally wrong. Sally is under no obligation to take this factor into consideration when she is deciding whether to accept the job offer. Consider INCORRECT CHANGE 1: Plausibly, if Tom decides to refrain from returning the extra five cents to the grocery store, he has not done anything morally wrong. Tom is not under a strong obligation to correct such an insignificant error. Even those of us who think Tom is under some obligation to return the five cents, we will think that it is not a very strong obligation. Consider INCORRECT CHANGE 2: Plausibly, if Susan decides to refrain from returning the $20, she has done something wrong. Plausibly, Susan has at least some obligation to correct the error. However, whatever obligation Susan is under, it is not a powerful obligation. If, for example, Susan decides not to return the money because she needs to get to an important appointment and correctly foresees that she will be late to the appointment if she returns the money, it is plausible that her decision is both understandable and forgivable. She will have committed no serious moral violation. On the other hand, John’s reason to help the child is very powerful, it is certainly stronger than a mere recommendation. Further, it is stronger than any other reason that I identified in any of the other scenarios. It seems reasonable to say that John has reasons to help the child that are so strong as to be demanding. If John fails to render aid for some reason, say, for example, because he does not want to cut his trip short, he will have done something very morally wrong.
My case for Premise 4 is something like an existence proof: by describing the above realistic scenarios, I am demonstrating how the existence of reasons in certain circumstances can make it the case that a person has a moral obligation. In DESERT HIKE John (a) has reasons to render aid to the child, (b) has reasons that are so strong that they constitute requirements; i.e., they are not merely suggestions, and (c) owes something to someone, namely, John owes it to the child to try to save her life. Thus, in DESERT HIKE, we have a situation in which the elements of a moral obligation are present and are present solely in virtue of the existence of reasons of the right sort. So, it is plausible that John’s moral obligation in DESERT HIKE is fully constituted by the reasons that he has to render aid to the child. Much the same can be said about many other cases of moral obligation.
The points that I have made thus far strongly support the claim that reasons of a certain strength and type are alone sufficient to ground moral obligations. But we will not be able to accept this conclusion if we think that God’s commands add something essential to the equation, something that can only be supplied via those commands. With that in mind, let’s turn to the second aspect of Premise (4), namely, (B): God’s commands, which are grounded in these reasons, do not ground moral obligations.
The suggestion that God’s commands make us obligated is implausible precisely because we know that an arbitrary command cannot make an action obligatory. And why is it that an arbitrary command cannot do so? I suggest that it is because a command is not the right kind of thing that could be what makes an action obligatory. After all, that Φ is commanded (by God or by anyone) is an extrinsic feature of an action. That Φ is commanded by God has nothing to do with (a) the kind of action Φ is, (b) the actual consequences of Φ, (c) the intended consequences of Φ, or (d) the reason Φ was performed.
Consider: Let Φ = John’s act of bringing the lost child to a hospital so that she can receive medical care. Let’s compare two scenarios: In scenario 1, God commands Φ. In scenario 2, God does not exist and so it is not the case that God commands Φ. The only difference between Φ-as-performed in scenario-1 (Φ-ap1) and Φ-as-performed-in-scenario-2 (Φ-ap2) is that Φ-ap1 has the feature commanded-by-God and Φ-ap2 lacks this feature. But Φ-ap1 and Φ-ap2 share every other features in common; they have the same actual consequences, the same intended consequences, they are acts of the same type, they are performed at the same time, by the same people, in the same location, with the same motives, etc. The one feature whereby Φ-ap2 differs from Φ-ap1 is an extrinsic feature. It concerns attitudes that a rational agent has taken toward the action, which attitudes, it is important to note, must, given that they are the attitudes of a perfectly rational agent, be grounded in reasons (which reasons must themselves must involve features of Φ). Given that the only difference between Φ-ap1 and Φ-ap2 is an extrinsic feature and that this extrinsic feature is one that is itself reason-responsive, it is implausible that this feature (rather than the reasons it is responsive to) is what makes Φ morally obligatory. Isn’t it more plausible that it is one of the other features that I mentioned above (its actual and/or intended consequences, the kind of act it is, the reasons for it, etc.) that make the action obligatory? How strange it would be if, rather than these other features, Φ is obligatory in virtue of a reason-responsive attitude that a rational agent has taken toward the action. In other words, that Φ is commanded by God is not the kind of feature that could make Φ obligatory.
In the following clip, Dennis the constitutional peasant makes a point similar to the one I am trying to make:
That a strange woman lying in a pond distributed a sword to you cannot be what makes you the supreme executive in the government. But this is precisely what Arthur Pendragon claims as the ground of his executive power. Note what he says: “The Lady of the Lake, her arm clad in the purest shimmering samite, held aloft Excalibur from the bosom of the water; signifying, by divine providence, that I, Arthur, was to carry Excalibur. That is why I’m your King.” The ‘why’ here is not the ‘why’ of reasons but the ‘in-virtue-of’ ‘why.’ When he says, “That is why I’m your King,” he is saying that this is what makes him King. Arthur is claiming that he is King in virtue of the fact that the Lady of the Lake held Excalibur out to him. Dennis claims that this claim is absurd. That a lady in a lake lobbed a scimitar at you is not the kind of thing that can make you the supreme executive.
In making this point, Dennis says the following: “If I went ‘round saying I was an emperor just because some moistened bint had lobbed a scimitar at me, they’d put me away.” If, instead of getting angry and using violence to silence his critics, Arthur had calmly explained to Dennis that the Lady of the Lake is perfectly wise and would never hold out a sword to a peasant, or any other person unqualified to be a leader, he would have missed the point. Dennis is not saying that, according to Arthur’s theory of executive power, it is possible for an unqualified person to become King (since it is possible for the Lady of the Lake to give a sword to any person, even an unqualified one). No. Dennis’ point, rather, is that getting a sword from a woman in a lake is not the kind of thing that could make one King.
Similarly, we may say to the divine command theorist that an action’s being commanded by God is not the kind of property that could make the action morally obligatory. And we know this because of the following observations (which, by the way, are agreed to by most defenders of DCT):
- The mere fact that Φ is commanded cannot make Φ obligatory. After all, any person could issue an absurd command (e.g., don’t brush your teeth before 7:00 am) and this would obviously not make for an obligatory action.
- The mere fact that Φ is commanded by someone powerful cannot make Φ obligatory. After all, powerful people often command horrendous things. That Hitler commanded the extermination of Europe’s Jewish population did not make it obligatory to do so.
- An arbitrary command, no matter who it is issued by, cannot make any action morally obligatory.
All of this strongly suggests that being commanded by anyone, even God, is not the kind of feature that could make an action morally obligatory.
It is useful to focus our attention on point (3). We know that if a command to Φ is arbitrary, then no matter who issues it, the command cannot make it the case that Φ is obligatory. This strongly suggests that commands, on their own, are never sufficient to make it the case that a person is obligated to do something. In other words, since a command that is not grounded in reasons cannot make it the case that anyone is under a moral obligation, bare commands are morally impotent. If that is right, then in cases in which a person is obligated to do something that he has been commanded to do, that person must be so-obligated in virtue of something other than the fact that he has been so-commanded.
Undoubtedly, there are cases in which a person that has been commanded do something is obligated to do that thing. One reason that this might be so is that the person is obligated in virtue of something independent of and prior to the command and the command serves as something like a reminder of this independent and prior obligation. But this sort of account will not cover all cases in which a person is obligated to do what she has been commanded to do.
There are other kinds of cases that we need to consider: cases in which some person is obligated to do something that she has been commanded to do but would not have been so-obligated if she had not been so-commanded. That is, there are circumstances in which the fact that a commander, C, commands someone, S, to Φ implies (in some sense of “implies”) that S is obligated to Φ. Military contexts provide examples of such cases. When a military subordinate is commanded to do something, he is obligated to do what he is commanded to do and, at least in some instances, would not have been obligated in the absence of the command. How do we square this with my above claim that bare commands are morally impotent?
I suggest that when the fact that C commands that S Φ makes it the case that S is obligated to Φ this is because of the existence of a prior obligation to obey C, which S is under. When a military subordinate is obligated to obey the commands of her superior officer, this is because, as a member of the military, she has an obligation (which is independent of the authority of any particular commanding officer) to obey the (lawful) commands of superior officers. In the absence of any such prior and independent obligation, the fact that S is commanded to Φ cannot establish any obligation that S Φ. This is consistent with my claim that bare commands are morally impotent. When commanding officer issues a (lawful) command to a subordinate, the subordinate is obligated to obey the command; but the command does not, all by itself, make the subordinate obligated. It does so only given the prior obligation that the subordinate has, namely the obligation to obey his superior officers. The command, considered in and of itself (and thus outside of a military context), is impotent. And we can see this if we imagine that the command is given, not to a subordinate member of the military, but to me or any other person who is not in the military. If a US Army general orders me to do something, this does not and cannot make it the case that I am obligated to do what he orders me to do. Without the prior obligation to obey the lawful commands of a superior officer, all military commands are impotent in the sense that they cannot make any person obligated to do anything.
If the above is correct, then all commands, no matter who issues them, are morally impotent considered in and of themselves. The only time a command can make a person obligated is when the commanded is under a prior obligation to obey the commander. And this fact will not help a defender of DCT respond to the objection that I am raising against it. If a defender of DCT insists that God’s commands ground moral obligations in virtue of the fact that we are obligated to obey God, then she is defending a version of Normative Divine Command Theory (NDCT) rather than Metaethical Divine Command Theory (MDCT). As I indicated in part 2 of this series, my target is MDCT and I am not here interested in raising objections to NDCT. The claim that our moral obligations are ultimately grounded in a basic moral principle, according to which we are obligated to obey God, is an interesting one. But as I have pointed out in other contexts, MDCT cannot rely on such a principle. On MDCT all moral obligations are grounded in divine commands, thus there can be no command-independent obligation to obey God. Therefore, appeals to our obligation to obey God cannot help MDCT here.
The upshot is that, with respect to the factors that make an action morally obligatory, the fact that God commands the action is irrelevant. A command, no matter who issues it, is not the kind of thing that could, in and of itself, make an action obligatory. Thus, God’s commands do not ground our moral obligations.
 Whether moral obligations extend to actions that effect all sentient beings or only persons is an interesting question. But it is one that it is beyond the scope of this essay. My point here is only that it is plausible that moral obligations are grounded in the presence of powerful reasons of the right sort (what sort that is will be determined through careful philosophical reasoning).
 This distinction is inspired by Joseph Raz’s distinction between mandatory and non-mandatory norms. See his Practical Reason and Norms.
Dancy, Jonathan (2004) Ethics Without Principles. Oxford: Oxford University Press.
Raz, Joseph (1999) Practical Reason and Norms. Oxford: Oxford University Press.