St. Thomas Aquinas, Summa Contra Gentiles I, chaps. 37-41 & 92-94
Thomas V. Morris, "Duty and Divine Goodness"
Robert M. Adams, "Must God Create the Best?"

A. The Elements of Goodness

1. SCG I, 37-41 & 92-94

    a. God's goodness follows variously from (i) God's perfection, (ii) God's being the first desired being, and (iii) God's being pure act without admixture of potency.

    b. God has virtues literally, but analogously.

      i. The negative analogy is as follows: God has no habits that are perfective of powers (instead He has the virtues as active perfections); God has no virtues that govern the use of one's body, or bodily passions; God has only metaphorically virtues which are directed at goods like honor, glory, etc. and their attainment.

      ii. The positive analogy is as follows: God can have that in the following virtues which does not in any way connote an imperfection:

        art: prevolitional and potentially practical knowledge of creatable things.
        prudence: prevolitional knowledge as practical knowledge that determines the will to one effect.
        justice: willing what is necessary for the perfection of each thing.
        liberality: willing good in such a way that it does not benefit oneself.
        truthfulness: showing oneself as one is.

B. Duty and Divine Goodness

Morris in part misstates the force of his argument. In fact, what the argument suggests is that there is a problem with God's being a perfectly good moral agent just because it seems impossible for anyone to be a moral agent unless that person has the power to do wrong freely, however one construes acting wrongly, i.e., whether one takes a morally wrong action to be one which is contrary to duty or simply contrary to virtue. So it is not really the duty model of moral goodness that is at stake here, but more centrally the connection between moral agency and freedom.

1. Morris's argument:

Argument from duty and freedom for conclusion that God is not a perfectly good moral agent:

    (1) An agent A is a perfectly good moral agent only if A has duties and freely refrains from acting contrary to duty. [premise based on duty model of moral goodness]

    (2) An agent A has duties and freely refrains from acting contrary to duty only if A is capable of acting contrary to duty. [premise based on notion of freedom]

    (3) So an agent A is a perfectly good moral agent only if A is capable of acting contrary to duty. [from (1) and (2)]

    (4) But God is necessarily good or impeccable and so is necessarily incapable of acting contrary to duty [premise based on divine goodness]

    Therefore, God is not a perfectly good moral agent.

So the conclusion is that whatever sort of goodness God might have, it cannot be moral goodness, since this demands the ability to contravene duty as a necessary condition. From this it follows that God is not worthy of a certain sort of praise which is reserved for those who do good instead of evil. Or at least God would not be worthy of praise for the sort of behavior which we would ordinarily think to be simply the fulfillment of duty rather than supererogation. Note, by the way, that Morris immediately focusses on the satisfaction of duty as opposed to the demands of virtue, and indeed his final position seems more consonant with a virtue model of divine goodness rather than what he calls a 'duty' model of divine goodness.

2. Possible solutions to the argument

    a. Deny premise (2) by adopting some notion of freedom which does not have the ability to do otherwise at its core. Objections: (i) Even determinists seem to admit some sort of ability to do otherwise; (ii) such a view of freedom might well create problems for the interpretation of other Christian doctrines (analogia fidei).

    b. Deny premise (4) and adopt the view that while it is metaphysically necessary that God be good in some sense, the being who is God is at least capable of perpetrating sin or moral evil and hence of ceasing to be God. Objections: This claim seems theologically repugnant to many of us and, besides, it may well be demonstrably impossible for any being to be God at one time and not God at a later time, thus ceasing to be God, just as it is now impossible for me to cease have the property of having at one time taught at Notre Dame, even though it is metaphysically contingent that I should have this property at all.

    c. Deny premise (1) and reflect more deeply on just how the duty model of divine goodness might be appropriated. Let us now explore this option

3. Morris' solution

Morris questions the first premise, viz.,

    (1) An agent A is a perfectly good moral agent only if A has duties and freely refrains from acting contrary to duty. [premise based on duty model of moral goodness]

He claims that the duty model of moral perfection can be salvaged even if we deny literally speaking that God has moral duties and hence deny that moral perfection strictly implies the possibility or capability of acting contrary to duty. For it may be that God necessarily acts in accord with moral duty without being obligated to follow moral duty.

    "We human beings exist in a state of being bound by moral duty. In this state we act under obligation, either satisfying or contravening our duties. Because of his distinctive nature, God does not share our ontological status. Specifically, he does not share our relation to moral principles--that of being bound by some of these principles as duties. Nevertheless, God acts in accordance with those principles which would express duties for a moral agent in his relevant circumstances" (p. 117).

Notice immediately that Morris's point would hold even if we thought, with St. Thomas, that it is better to think of God acting in accord with virtue rather than duty, since on St. Thomas's view the very idea that God has moral duties or obligations would be a non-starter. God is the source of obligation because He is the law-giver. On the other hand, it might very well be that we can expect from God action in accordance with those virtues which we "share" in some sense with all beings having understanding and free will.

Morris seems to go too far in claiming that the fulfillment of duty never warrants praise, strictly speaking, even though he is probably right in claiming that the fulfillment of duty does not always or necessarily warrant praise. It seems to depend upon the circumstances. In some cases the mere fulfillment of duty might require heroic fortitude; in other cases it might require relatively little effort.

Morris actually overlooks an important argument that would strengthen his case. What is our moral ideal with respect to any given virtue--presumably to become the sort of person who has become so habituated to thinking and acting in the right sort of way that acting contrary to that virtue is something we do not think of or do not entertain as a serious alternative. We want to become the sort of person who "could not bring himself" to be cowardly, or unchaste, or gluttonous, or drunk, or greedy, or power-hungry, etc. But then our freedom is seen primarily as a means to an end. Finally, we are praised for having attained this end-state just as much as for having attained it freely. So freedom is only a means to becoming by "second nature" as God is by "first nature". And this is another way of seeing that while the freedom to go wrong might be an essential feature for moral virtue among creatures, it is not a pure perfection or anything that we would want to attribute to God. In this sense, to use Morris's terminology, moral goodness is a kind of metaphysical goodness, viz., the full metaphysical goodness possible to a being with intellect and will--and God has this sort of goodness by nature whereas creatures endowed with intellect and will must under the present dispensation attain it. That is, these creatures have some metaphysical goodness by virtue of being members of a given species and have full metaphysical goodness only to the extent that they acquire moral perfection.

One way to make this clearer to the students is to distinguish roughly two senses of metaphysical goodness. The first is the goodness one has simply by virtue of the fact that one has actuality (esse) determined by a certain essence (essentia), i.e., insofar as one is simply an existing member of a natural kind. The second is the goodness one has in virtue of being a perfect(ed) or excellent member of that kind. So (roughly):

    x is metaphysically good1 = x is a member of natural kind K, but x is not a perfect (or: excellent) instance of K.

    x is metaphysically good2 = x is a member of natural kind K, and x is a perfect (or: excellent) instance of K.

Now we can define moral perfection in terms of this second notion of goodness in the following way.

    x is morally perfect = x has intellect and will, and x is metaphysically good2.

Now we can distinguish between our moral perfection and God's moral perfection in such a way that premise (1) of the argument is shown to be false. Both God and human beings can be morally perfect in the sense just defined. BUT

    FOR US: Moral perfection is an achievement which requires freedom as a means to bring us from being good1 to good2. Because we begin with this gap between what we are and what we ought to be, and because our good is (in part) a good peculiar to beings endowed with intellect and will, our freedom is an essential ingredient in our rise toward moral perfection.

    FOR GOD: Moral perfection is a natural attribute, since God is good2 by nature. Hence God does not become morally perfect through his freedom; indeed, it is a mark of imperfection to have the freedom to fall from being good2 or to become good2.

C. Must God Create the Best?

1. The Argument

Adams's paper is a dialectical masterpiece. Notice how he puts his opponents on the defensive from the beginning. One might have thought that the burden of proof is on Adams, since his position on the surface seems to be "untraditional" (even though this is not so in reality). But Adams basically asserts his thesis and then challenges his opponents to come up with objections--even though he does not give any really impressive "positive" argument for his claim. Actually, the best positive argument he gives occurs in the reply to the objections. So since you are familiar already with the structure of the paper, I will present the argument in a somewhat different way, presenting first a "positive" case for Adams's thesis and then showing how he responds to the objections.

Adams's main thesis is that (P) is false, where (P) is the following thesis:

    (P) If a perfectly good moral agent created any world at all, it would have to be the very best world that He could create.

Now it is worth reflecting for a moment on just what it means to hold that (P) is false. Here some of the concepts we've already introduced in this part of the course will be helpful. Imagine that God is in a creation situation with a full complement of prevolitional knowledge. Now it does not matter whether one is a Thomist or Molinist as far as (P) is concerned. For what (P) says is that in any given creation situation a perfectly good moral agent will actualize the best alternative world open to him, i.e., the best feasible world. If Thomism is true, then the best feasible world is always the best possible world (assuming for the moment that there is such a world), whereas this need not be so on the Molinist position. So (P) says that in any given creation situation a perfectly good moral agent would opt for the best feasible world. So to show that (P) is false one has to defend the claim that in some given creation situation a perfectly good moral agent chooses to actualize some world other than the best feasible world. Adams further stipulates that he will try to do this from within a Judaeo-Christian ethical theory. How does one go about trying to show this?

    a. The most obvious way is simply to deny that there is any such thing as a best feasible world or a best possible world. And this seems eminently plausible. Just as there is no highest number and for any number you please we can come up with a better one, so too it seems that on any plausible measure of goodness you choose, no world is going to be unsurpassable in goodness. Adams realizes all this, but he could not have gotten the paper published if he had just pointed this out. So he wants to show that even if there is a best feasible world, a perfectly good moral agent would not have to create it. So Adams is really arguing that the following is false:

      (P*) If a perfectly good moral agent created any world at all, and if there were a very best world that He could create, then He would have to create that world.

    b. So now the question is: how does one show that (P*) is false. Well what Adams does is to present the following as a possibility: God chooses a world that meets the following conditions:

      (1) None of the individual creatures in it would exist in the best of all possible (feasible) worlds.

      (2) None of the creatures in it has a life which is so miserable on the whole that it would be better for that creature if it had never existed.

      (3) Every individual creature in the world is at least as happy on the whole as it would have been in any other possible world in which it could have existed.

Notice immediately that Adams is not claiming that the actual world, i.e., our world, satifies these three conditions. He does not have to claim this in order to argue that (P*) is false. All he has to claim is that in some creation situation God could have created a world that satisfies these three conditions. It is also worth noting that the Christian faith gives us good reason to believe that in our world neither (2) nor (3) is satisfied. (Judas is a counterexample to both.) So if it is at all plausible (and it is) to suppose that in some creation situation God could create a world satisfying (1) - (3), then Adams wins, and that's the end of the matter.

2. Objections:

a. If (P*) were false and if God were to create a world other than the best feasible world, then He will have either (i) violated someone's rights or (ii) treated someone unkindly.

    Reply to a (i):

    Just whose rights would be violated if God created a world that exemplifies conditions (1), (2), and (3) above? Surely not the denizens of the best possible (feasible) world, since those people do not exist and hence have no rights to be violated. Surely not those whom God created instead, since they are as well off as they possibly could be, given that (3) is satisfied.

    Further objection: But is it so clear that (3) could be satisfied by any world other than the best world (or indeed even by the best feasible world)? And if it cannot be, then God will have violated the rights of any being who would have been happier had He created another world.

    Further reply:

    As long as (2) is satisfied, then no one has a just cause for complaint. After all, if your life is on the whole not so bad that it is better for you never to have existed, then what right have you to complain? It may very well be that you do not exist in any world in which (3) is satisfied and, if so, then it follows that in order for you to as happy as you could be, someone else would have to less happy than he or she might be.

    What's more, it's not even clear that the best feasible world satisfies (3). Most people who have thought about this matter have concluded that it does not satisfy (3). So there cannot be any general requirement that God create a world that satisfies (3) on pain of violating the rights of all those who would have been more happy in another feasible world. Even if I make the complaint not just on my own behalf but on behalf of everyone in my situation, it still follows that if God created a world in which (3) were satisfied, then on the current assumptions neither I nor they would have existed. So how can our rights have been violated?

    Reply to a (ii):

    If (3) is satisfied, then God has obviously not acted unkindly. So let's assume that the world God has created satisfies only (1) and (2) but not (3). Now the question is: Has God treated anyone unkindly by creating a world that does not satisfy (3).

    Once again, the first reply is that probably even the best feasible world does not satisfy (3); so God can hardly be faulted for creating a world that does not satisfy (3).

    Even if it is the case that God has been less kind to a person P than He could have been to P, it still does not follow that He has been unkind to P or that He has not been very kind to P. God has to worry about the universe as a whole, and various considerations might lead Him to let some be less happy than they would otherwise be for the sake of some other good He wants to preserve. But this sort of consideration undoubtedly applies to the best of all worlds as well.

b. The second objection is that if (P*) were false and God created a world other than the best feasible world, then He would be exhibiting a defect of character--like an athlete or student who doesn't do his best.

    Reply to b: We have to distinguish between doing one's best and bringing about the best state of affairs one can. We all recognize this distinction as applied to ourselves. But does it apply to God? Adams uses the notion of grace here:

    "A God who is gracious with respect to creating might well choose to create and love less excellent creatures than he could have chosen. This is not to suggest that grance in creation consists in a preference for imperfection as such. God could have chosen to create the best of all possible creatures, and still have been gracious in choosing them. God's graciousness in creation does not imply that the creatures he has chosen to create must be less excellent than the best possible. It implies, rather, that even if they are the best possible creatures, that is not the ground for his choosing them. And it implies that there is nothing in God's nature or character which would require him to act on the principle of choosing the best possible creatures to the object of his creative powers" (p. 98).

    Grace, Adams claims, is part of the Judaeo-Christian moral framework, and so God does not exhibit any moral defect by virtue of choosing to create a world other than the best world He could create. We are grateful that God created us even though He could have created much more impressive creatures instead of us. So we certainly do not think that God has exhibited a defect of character.

    Raise some problems here about the notion of goodness as applied to worlds. How do we even begin to think about it?

3. The examples:

CASE A: A couple knowingly and deliberately brings a retarded child (call her Susanna) into existence (what they do is such that if they had not done, that particular child would not have been conceived).

Here Adams claims that it is wrong to think that the child has been wronged or had its rights violated, even though it is uncontroversially true that the parents have done something wrong. For we are assuming that the child does not have such a miserable existence that it would have been better for her never to have existed; and we are also assuming that Susanna is as happy as she would have been in any other feasible worlds. (If these conditions are not fulfilled, then the case is not analogous to God and a less than best world.) And further it is the case that the child would not have existed unless it were retarded.

Yet everyone agrees that the couple has done something wrong, i.e., has at least exhibited some defect of character or acted wrongly or viciously. Why exactly? What principle might they have violated which is both true and such that by violating it they have done something wrong. Adams begins by suggesting the following principle:

    (Q) It is wrong to bring into existence, knowingly, a being less excellent than one could have brought into existence.

However, it is obvious that (Q) is not true or such that one who violates thereby does something wrong. This is clear from the second example:

CASE B: A man knowingly and deliberately breeds goldfish instead of dogs.

Another case to think about and which undermines (Q) is the following:

CASE C: Parents who refuse to breed a superintelligent child even though they have the opportunity to do this.

What this suggests is that the following principle has been violated:

    (R) It is wrong for human beings to cause, knowingly and voluntarily, the procreation of an offspring of human parents which is notably deficient, by comparison with normal human beings, in mental or physical capacity.

But this principle obviously has no application to God and hence cannot be used to show that God does something wrong by creating less than the best world.

Don Turner's comments (Don was my assistant when I taught this course):

First, Adams is not assuming that the happiness of creatures is up to God, but rather that God has comprehensive knowledge of the futures of all feasible worlds. (Notice that this commits him, it seems, to either Thomism or Molinism, neither of which are consistent with his later views on providence and foreknowledge.

Second, one begins with graciousness, but Don asks graciousness to whom? (Is this a fair question?). If Adams says to those he is going to create, then these are merely possible people. If he says to those whom he has already created, then God would have been just as gracious in creating others. (This is not quite right). Also, Adams assumes but has no right to assume that the best world contains more excellent creatures than the other worlds. Even those in the best world could consider themselves the scum of the earth and ask why God would have bothered to create them.

Also, it is not clear that the examples are replied to adequately. Perhaps variety is a desideratum in a world and this is why it is alright for Jones to breed goldfish rather than dogs. But if there were an oversupply of one and the other were dying out, then there might be a requirement to breed the one or the other. So too it might be a requirement of God's moral perfection that He create a world with just the right sort of variety. But then this gives us no reason to think that God would create less than the best world he could create.