God, Faith and Infinity » Lecture 6
There is a broad … consensus that the main divine attributes are:
- Omnipotence
- Creatorship
- Omniscience
- Eternity and omnipresence
- Personhood
- Goodness ⁄ [moral] perfection
- Non-physicality
- Necessary existence
- Simplicity
- Immutability
- Impassibility [not susceptible to transient suffering and pleasure].
… perfect being theism, or the ‘maximal greatness’ tradition, regards God as the uniquely perfect being or the maximally great being, and the attributes then specify what characteristics count as perfections, or which aspects of God are maximally great. (Everitt 2010: 78)
It is standardly assumed that these features are essential properties of any PBT God (Oppy and Scott 2010: 236).
Maybe the fact that God’s properties are part of his essence also follows from his perfection. (Recall here Plantinga’s (1974) modal ontological argument.)
It may also be thought to follow from the divine attributes that God has no accidental properties, since no accidental feature could be explained by something intrinsic to God’s own nature.
The simplest proposal:
But note that this depends pretty crucially on what an action is.
So we might prefer to define omnipotence in terms of the agent’s options. An option is a proposition such that an agent can ‘act at will so as to make it hold’ (Lewis 1981a: 7).
(O\(x\)) is a schema; different definitions will impose different conditions \(C\).
The most liberal view is that there is literally no restriction at all on what can is able to do:
If this is what omnipotence means, then an omnipotent being is able to ensure the truth of contradictions and impossibilities, among other propositions.
This liberal view cannot be right. Suppose X is an omnipotent being according to (O1):
So we might say: let’s impose the condition that \(\phi\) must be possible:
Following our definition of omnipotence, the argument talks of possibly true propositions. But are the propositions in question really possible?
when Jane is an essentially omnipotent agent … the state of affairs of Jane’s being non-omnipotent is impossible. Therefore, Jane cannot bring it about that she is not omnipotent. Since, necessarily, an omnipotent agent can move any stone, no matter how massive, [the proposition that there is a stone which Jane cannot move] is impossible. But, as we have seen, an omnipotent agent is not required to be able to bring about an impossible state of affairs. (Hoffman and Rosenkrantz 2017: §2)
If this is right, (S3) is false: the proposition that an essentially omnipotent being \(x\) makes it true that there is a stone that \(x\) cannot lift is impossible. Making that true would ensure the being loses their omnipotence – which it cannot, because it is essential.
If an omnipotent agent \(x\) can’t make it true that there is a stone \(x\) cannot lift, then it is not possibly true that (there is a stone and \(x\) can’t lift it).
Substituting, using the logical equivalence of \(\exists y (Sy \wedge ¬Lxy)\) and \(¬\forall x (Sy \to Lxy)\), yields:
Since ‘not possibly not’ is equivalent to ‘necessarily’, we can substitute:
That allows us to reformulate (S3) in a way that reveals its falsity quite clearly.
That proposition is clearly not possible!
(O2) is not such an account.
it is not possible for an efficient cause to occur later than its effect. However, an agent’s bringing about a state of affairs is a kind of efficient causation. Therefore, it is not possible for an agent to bring about anything that is in the past. In other words, it is impossible for any agent to have power over what is past. Hence, no agent, not even an omnipotent one, can bring it about that [‘a raindrop fell’] obtains. (Hoffman and Rosenkrantz 2017: §2)
But the proposition that a raindrop (already) fell is clearly possibly true.
So if this is right, we need to further restrict our definition (O2) – because we would have a need to ensure that omnipotent beings are not required to be able to affect the past, and so some possible propositions must be outside the scope of God’s control.
If this is right, we get this third definition: a definition which tries to emulate the scope of (O2), but for a God who is embedded in time rather than outside it.
This is not subject to constraints imposed by time or causation, constraints which (intuitively) should not be constraints on an omnipotent agent, though they might well be constraints on us.
The question remains: why opt for (O3) rather than (O2)? Reasons for thinking God is within time are not, to my mind, compelling; so I’ll deal with the simpler (O2) from now on.
If a plurality of coexistent omnipotent agents were even possible, then possibly, at a time, \(t\), some omnipotent agent, \(x\), while retaining its omnipotence, endeavors to move a feather, and at \(t\), another omnipotent agent, \(y\), while retaining its omnipotence, endeavors to keep that feather motionless. Intuitively, in this case, neither \(x\) nor \(y\) would affect the feather as to its motion or rest. Thus, in this case, at \(t\), \(x\) would be powerless to move the feather, and at \(t\), \(y\) would be powerless to keep the feather motionless! But it is absurd to suppose that an omnipotent agent could lack the power to move a feather or the power to keep it motionless. Therefore, neither \(x\) nor \(y\) is omnipotent. This line of reasoning appears to reduce the notion of a plurality of coexistent omnipotent agents to absurdity. If such a reductio ad absurdum is sound, then a plurality of coexistent omnipotent agents is impossible. (Hoffman and Rosenkrantz 2017: §1)
Let us say that one being X is dominated by another Y just in case ‘that other being [Y] will be able to do all that the described being [X] can do, and more besides’ (Oppy 2007: 81)
With this notion, we may find this domination principle plausible:
God, on the standard conception, dominates us – that is enough to exclude us from the ranks of omnipotent beings, without having to tally up what we in fact can and cannot do.
If God is only contingently good, then
we have adopted a religiously unappealing conception of the creator of the world. … In view of the horrendous evils of this world, why should we suppose that the creator is perfectly good if we have already acknowledged that, at best, the creator is merely contingently perfectly good? (Oppy 2007: 82)
One response to this argument from dominance revisits our definition on omnipotence in terms of options.
An omnipotent agent can bring about any possible state of affairs at will. But then what an omnipotent agent would will seems to be relevant to their options. This gives us a further narrowing of (O2):
Noting God’s essential goodness, we might think God would never will evil. So an evil state of affairs that God cannot bring about turns out not to be an obstacle to God’s omnipotence, so long as God would not will that state of affairs into being.
Adopting (O4) leads us to reject the domination principle (D): God is dominated by the morally indifferent agent, but remains omnipotent by (O4), because God is able to enact at will anything he would will.