3.1: The Various Traits of God (Tom Metcalf)

16 The Various Traits of God
Tom Metcalf²⁶

16.1 Defining ‘God’

Who, or what, is God? We’re going to be talking about the god that philosophers refer to as the ‘Anselmian’ God, after St. Anselm of Canterbury, who was one of the first philosophers to define God explicitly this way. These philosophers also generally refer to God using masculine pronouns, so I’ll follow that tradition, but I don’t mean to imply anything in particular about God’s gender.

This god occupies center stage in debates about the divine attributes: God’s traits. Anselm described God in a few ways, but we’ll use this one:

Conceptually Maximal Greatness: God is a being greater than which none can be imagined (Anselm 1998: Chapter II).

Anselmians are theists who believe in the Anselmian God.

Of course, this definition doesn’t tell us very much yet about which attributes God actually has. But Anselmians generally believe that some conclusions follow. First, God would be omnipotent or all-powerful, because such a being would be greater than one with limited power. Second, God would be omniscient, or all-knowing, for the same reason. Third, God would be morally perfect, because an immoral or weak-willed God would be less great or otherwise worse. Last, God would be an eternal, immutable, omnipresent, and necessarily existent person: God exists throughout all of time and space (and possibly outside of or beyond it); God cannot change; God cannot possibly not exist; God isn’t a physical object; and God has intentions and a mind. And to be a person is greater than to be a mere, inanimate object (Anselm 1998: Chapters VI-VIII).

Therefore, we say that Anselmians believe that God is an omnipotent, omniscient, morally perfect, eternal, immutable, omnipresent, necessarily existent, immaterial person who created the universe. Case closed? Not so fast. You may have already realized that each of those terms will, itself, inspire further controversy and debate. In the rest of this entry, I’ll present what I take to be the most-defensible analyses of the main divine attributes.

16.2 Omnipotence

The most-straightforward way to think about being omnipotent or all-powerful would be that God can do absolutely anything. Still, most Anselmians today (and Anselm himself) would reject this view (cf. Anselm 1998: Chapter VII; Aquinas, I, Q. 25, Art. 3, ad. 2.; Flint and Freddoso 1983: 99, Hoffman and Rosenkrantz 2002: 172; but see Descartes 1991: 25). Why? Because it doesn’t seem plausible that God could do what is literally impossible, and it might even seem to make him less than great (Anselm 1998: Chapter VII). Could God create a square circle? Could he destroy himself? Could he compose a guitar solo so difficult that even he couldn’t play it? (Isn’t it impossible to compose a guitar solo too difficult for God?) Therefore, Anselmians are generally okay with weakening God’s omnipotence to say that God can only do what is possible.

The debate doesn’t end there, however. For there seem to be many things that are possible, in some sense, that God still can’t do. Can God bring it about that I freely choose to order the oysters for dinner? Maybe not, because if he caused me to do it, then maybe I didn’t do it freely. But of course it’s possible that I bring it about that I freely choose to order the oysters, at least on the assumption (shared by most philosophers (Bourget and Chalmers 2014: 476)) that free will of some sort or another exists. Similarly, can God bring it about that for the first time ever, a butterfly flies? No; at least, not now, because it’s already happened. But surely it’s possible that for the first time ever, a butterfly flies. It’s not just possible; it actually happened.

These considerations have led many Anselmians to accept a definition of omnipotence that looks basically like this (Flint and Freddoso 1983: 83; Hoffman and Rosenkrantz 2002: 172):

Omnipotence

A person is omnipotent at a certain time in a certain world if and only if for any situation that it’s possible at that time in that world to bring about, either:

(1) that person can bring about that situation at that time in that world; or

(2) that person can’t bring about that situation, but only because that situation comprises someone else’s freely choosing something.

If no butterflies had ever flown, then God could bring it about that for the first time, a butterfly flies. And it’s just not possible in any world for anyone to bring it about that God be destroyed or that there is a guitar solo too difficult for God.

Here’s one way, however, that we shouldn’t amend our definition of omnipotence:

A person is omnipotent at a certain time in a certain world if and only if for any situation that it’s possible for that person at that time in that world to bring about, that person can bring about that situation at that time in that world (cf. Wierenga 1989: 16-17).

Most philosophers who have commented on this think this change won’t work, because it makes too many beings omnipotent. Consider the person Mr. McEar. He’s a being who can only ever scratch his ear; that’s all it’s possible for him to do (Plantinga 1967: 170). Yet that amended definition makes him omnipotent, which seems clearly incorrect.

16.3 Omniscience

As with omnipotence, defining ‘omniscience’ gets a little complicated. It basically means that someone knows everything. For any true proposition, God knows that proposition. (I don’t just mean that he knows of it; I mean that he knows it, in the same way you know the proposition that 2+2=4 and the proposition that parrots exist.) And it’s normally thought that God also doesn’t have any false beliefs. (Maybe God’s perfect rationality is also required by maximal greatness.)

Is it possible for anyone to know everything? Maybe not. There are many potential problems with defining omniscience (cf. Grim 2000), but let’s just consider three of them.

(1) Consider the proposition that

God does not know this proposition.

Yes, that proposition refers to itself, in the same way that ‘this sentence has five words’ refers to itself. But does God know the proposition? If you say ‘yes,’ he does know the proposition, then the proposition is false, and most philosophers think that it’s impossible to know a false proposition. (You can believe it, but you can’t know it.) So we can rule out ‘yes.’ But if you say ‘no,’ i.e. that he doesn’t know the proposition, then the proposition is true, and he’s failed to know a true proposition. That seems incompatible with omniscience.

(2) Consider the proposition that

I have exactly three cats.

I know that proposition: it’s true; I believe it; and my evidence for it is sufficient for my belief to count as knowledge. My belief isn’t false, so that proposition is indeed true. But does God know that proposition? Well, yes and no. Surely he knows the proposition that

Tom has exactly three cats.

Isn’t that enough? Some philosophers would say ‘no.’ They would say that those two propositions are different propositions. After all, someone could believe one without believing the other. In fact, I bet that unless you happen to have exactly three cats, as long as you trust my report about my cat-possession, you believe one proposition but not the other. And if for some reason I forgot that I was Tom, I could believe one and not the other. So the two propositions can’t be the same proposition. Yet presumably God doesn’t have exactly three cats. So if he were to believe the proposition that

I have exactly three cats,

then he’d have a false belief, and not be both omniscient and perfectly rational. Yet if he fails to believe that proposition, then there is a true proposition that he’s failed to believe. In short: the proposition that I have three cats and the proposition that “he” has three cats (“he” referring to me) might be different propositions, but God can really only know the second one.

(3) Let’s suppose that libertarian free will exists. That would mean that in some situations, there’s more than one possible choice for me to make (cf. O’Connor and Franklin 2018). So consider my (future) choice about which shoes I will wear to work tomorrow. If I have libertarian free will, then there’s more than one real possibility. Does God know what I will choose? Well, if he does, then arguably, there’s only one possibility, because it’s impossible that God’s beliefs be false, and my choosing something other than what God expected would mean he had a false belief. But then I don’t have libertarian free will. In turn, if libertarian free will exists, then maybe omniscience is impossible (cf. Zagzebski 2018).

Therefore, maybe the Anselmian should define ‘omniscience’ as something like this:

Omniscience

A person is omniscient at some time if and only if: for any true proposition that it is possible to know at that time, either:

(1) that person knows that proposition at that time, or

(2) that person doesn’t know that proposition at that time, but only because the proposition:

(a) contains a paradoxical reference to itself; or

(b) contains an indexical reference to a particular knower; or

(c) is a proposition that it would be impossible for a rational being who knew every true proposition to know.

This definition may handle the problems above; it seems to indicate maximal knowledge, even if it doesn’t indicating knowing absolutely everything. Of course, each of the terms (a)-(c) will have to be spelled out in more detail to handle all the problems that could arise.

16.4 Moral Perfection

What would moral perfection be? Well, ‘perfection’ implies that nothing could be better. So if God ever did anything less than what’s morally best, then he wouldn’t be morally perfect.

Maybe it’s fairly easy to imagine that a being might not ever do anything morally wrong. Still, many ethicists believe in a moral theory according to which ever doing anything less than the best is, itself, morally wrong (cf. Kagan 1991). In any case, to truly count as ‘perfect,’ maybe a being really would have to always create what’s best. Otherwise, we could imagine a slightly better being.

Here’s a potential problem: It’s not completely certain that there is always a fact of the matter about what’s morally best to do. Let’s assume (with most philosophers (cf. Bourget and Chalmers (2014: 476) on moral realism) that some things in the world are good and some are bad. Bunnies are good. Suffering is bad. Human virtue is good. Broken promises are bad. And so on. Then we can ask, of a particular being, what sort of world it would create if it had the chance.

If you were going to create an entire universe, how many happy baby bunnies would you create? Suppose your answer is ‘one million.’ Well, wouldn’t a being who would have created more bunnies, e.g. a billion of them, be a morally better being? If we’re creating an entire universe, after all, we can fit as many bunnies in it that we want. So if we thought God might create a trillion bunnies, then we could imagine a morally better being, one who would have created a quadrillion bunnies.

You might already be thinking of a possible solution to this problem. Why not just say that God would create infinitely many bunnies? Surely no amount is higher than infinity. Are we really going to say there could be a being who would create infinity-plus-one bunnies?

Well, unfortunately, most philosophers who have written about this (and most mathematicians) think that in a sense, there is something higher than infinity (cf. Bagaria 2018: § 3). Or, more precisely, there is a smallest infinity, and there are infinities that are larger than this smallest infinity. Unfortunately, in fact, there are infinitely many infinities, and no highest-quantity infinity. So we have the same problem.

At the very least, we see that there could be a serious problem here. Maybe the Anselmian should define ‘moral perfection,’ then, something like this:

Moral Perfection

A being is morally perfect if and only if: For every choice such that there is a morally best choice to make, the being makes that morally best choice; and for every choice such that there is no morally best choice to make, the being doesn’t make any blameworthy choice.

Maybe there is no morally best choice to make in the ‘how many bunnies’ question, but that there are at least blameworthy choices, such as to make zero bunnies. That’s not going to be perfect; after all, if you could create a trillion bunnies, it might be blameworthy to create only a million, and so on. But as with many other cases, this might be the best we can do.

16.5 Eternality, Omnipresence, Immutability, Necessary existence, Immateriality, and Personhood

So far, we’ve looked at omnipotence, omniscience, and moral perfection, and found that matters are much more complicated than they might initially seem. Indeed, I’d say that none of definitions we’ve looked at so far has really been perfect. I’ve saved six more attributes for last: eternality, omnipresence, immutability, necessary existence, immateriality, and personhood. But fortunately, none of these attributes is ultimately very troubling compared to the others. I do want to say something about necessity and immateriality, however.

First, as for necessary existence, it’s not completely clear that a person could be a necessarily existing being. If necessarily existing beings exist at all—which some philosophers doubt—we normally imagine them to be such entities as numbers, properties, relations, and propositions (cf. Balaguer 2018). Maybe it doesn’t seem possible for the number nineteen to go out of existence. But God himself—a person—might be different. It certainly seems possible for persons to go out of existence. And if it’s possible for God not to exist, then given God’s definition as a necessarily existent being, it follows that God actually doesn’t exist.

Second, as for immateriality, some philosophers believe that there are simply no immaterial objects at all (cf. Balaguer 2018: § 2), or at the very least, that we couldn’t know about them (cf. Balaguer 2018: § 5). But this is a deep, complicated debate, which we mostly need to save for another time.

16.6 Conclusion

We have seen definitions of the main divine attributes, definitions that seem defensible but also to capture an intuitive idea of each of these attributes. However, one can always question whether limiting the attributes in this way is nevertheless compatible with the idea of a being greater than which none can be conceived. Our answer will depend on further, difficult issues about what exactly it means to conceive of something. We may end up concluding, with a few philosophers (cf. Almeida 2012: Chapter 1), that Anselmians should define God, instead, as a being greater than which none can exist. Whether this is plausible depends on whether you think such a being is worthy of the title ‘God.’

References

Almeida, Michael J. 2012. Freedom, God, and Worlds. Oxford, UK: Oxford University Press.

Anselm, “Proslogion.” 1998. In Anselm of Canterbury: The Major Works, ed. Brian Davies and G. R. Evans (Oxford, UK: Oxford University Press), pp. 82-104.

Bagaria, Joan. 2018. “Set Theory.” In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), URL = <https://plato.stanford.edu/archives/...es/set-theory/>

Balaguer, Mark. 2018. “Platonism in Metaphysics.” In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), URL = <https://plato.stanford.edu/archives/...ies/platonism/>

Bourget, David and David J. Chalmers. 2014. “What Do Philosophers Believe?” Philosophical Studies 170(3): 465-500.

Descartes, René. 1991. The Philosophical Writings of Descartes (Volume 3: The Correspondence). Ed. John Cottingham, Robert Stoothoff, Dugald Murdoch, and Anthony Kenny. Cambridge, UK: Cambridge University Press.

Flint, Thomas P. and Alfred J. Freddoso. 1983. “Maximal Power,” in The Existence and Nature of God, ed. Alfred J. Freddoso (Notre Dame, IN: University of Notre Dame Press), pp. 83-113.

Grim, Patrick. 2000. “The Being that Knew Too Much.” International Journal for Philosophy of Religion 47(3): 141-154.

Hoffman, Joshua and Gary S. Rosenkrantz. 2002. The Divine Attributes (Malden, MA: Blackwell).

Kagan, Shelly. 1991. The Limits of Morality. Oxford, UK: Oxford University Press.

O’Connor, Timothy and Christopher Franklin. 2018. “Free Will.” In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), https://plato.stanford.edu/archives/...ries/freewill/

Plantinga, Alvin. 1967. God and Other Minds: A Study of the Rational Justification of Belief in God. Ithaca, NY: Cornell University Press.

Wielenberg, Erik W. 2000. “Omnipotence Again,” Faith and Philosophy 17 (1): 26-47.

Wierenga, Edward R. 1989. The Nature of God: An Inquiry Into Divine Attributes (Ithaca: Cornell University Press).

Zagzebski, Linda. 2018. “Foreknowledge and Free Will.” In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), https://plato.stanford.edu/archives/...foreknowledge/

For Review and Discussion

1. What traits does God have according to Anselmians? Does this match up with how you generally think about God’s traits?

2. Why might it be a problem for God to have certain traits? Is that being still “God” if it doesn’t have, for example, omnipotence?

3. There are a number of traits mentioned but not covered in detail, like immutability, omnipresence, and eternality. Are these traits as problematic for God as the others that are covered in more detail? Why or why not?