2

Twice this week, in academic works, I came across the "idea" that omnipotence implies omniscience. I don't remember the first place I encountered it (I'm trying to remember, and if I do I'll add it to my question), but it was mentioned in passing so I just shrugged it off. But I just came across it again, in Prospects for a Sound Stage 3 of Cosmological Arguments by Jerome Gellman:

I conclude that Gale and Pruss's argument, in addition to being promising for stage(1), is also potentiating for showing that the necessary being who created the world is essentially omnipotent (given stage 2). (According to the view that omnipotence entails omniscience, we can conclude that the necessary being is essentially omniscient as well.) If I am right, Gale and Pruss's argument has advanced the cause of optimal cosmological arguments in an important way.

[Emphasis mine]

I'm assuming Gellman wouldn't include this parenthetical remark unless there were people who actually do hold this view and argue that omnipotence entails omniscience. So... who does argue for this and how?

Adam Sharpe
  • 3,764
  • 2
  • 9
  • 35
  • While omnipotence implies the *ability* to be omniscinet doesn't it imply the ability *not* to be as well? – Conifold Feb 01 '20 at 08:52
  • @Conifold If we're talking about omnipotence in general, it seems that way to me too. But if we're talking specifically about God's omnipotence, I'm guessing the answer would be that God doesn't have the ability to make himself non-omniscient since there would be an implicit contraction (maybe non-omniscience contradicts another essential property of his, like perfection). And God can only do what is logically possible. – Adam Sharpe Feb 01 '20 at 16:26
  • If we are talking about Christian God, at least on some conceptions of free will, God gives up omniscience to let his creatures co-create the world. Omnibenevolence is also essential to his nature. [Boethius and Aquinas](https://plato.stanford.edu/entries/free-will-foreknowledge/#BoetSolu) even moved omniscience into somewhat obscure (some say nonsensical) timeless interpretation to preclude theological fatalism. – Conifold Feb 01 '20 at 20:18
  • @Conifold I don't know much about the theology here... But, my initial thought is if omniscience is defined as "knowing everything that it's logically possible to know" (Swinburne defines it this way; I don't know if he was the first, but it seems sensible to me), and it's not logically possible to know what a creature will freely choose, God not knowing what a creature will freely choose doesn't detract from his omniscience. Likewise, if it is logically possible to know what a creature will freely choose, then God knowing what a creature freely chooses doesn't make the choice any less free. – Adam Sharpe Feb 01 '20 at 20:59
  • The problem with that is circularity a la the ontological argument. One has to *first* offer a conception of what omniscience is and what God can know, and *then* verify that it is logically consistent. Logical consistency is a second order constraint, it can not be used as a first order predicate in the definition of what omniscience is, any more than existence can be similarly used. – Conifold Feb 01 '20 at 21:05
  • @Conifold I'm not following. Suppose God could create a truly random number generator. A truly random number generator is such that it's not possible to know what number it will generate. And so, it's not possible for an omniscient being to know what number it will generate. Maybe, it's not knowable because all propositions of the form "the RNG will generate number x" lack a truth value until the number is actually generated, and it's only possible to know true propositions. I don't see the circularity. – Adam Sharpe Feb 01 '20 at 21:24
  • It may not be possible to predict what a truly random generator generates, but God can know it regardless, since he can know every truth w/o deriving it. On other conceptions of free will (e.g. Molina's) God *can* know "counterfactuals of freedom". More generally, what is or is not logically possible depends on the whole conception, it can not be used as a justification for anything in particular within it. Declaring in advance that something is to be logically impossible and thereby not detract from omniscience is simply a cheat. Consistency is not a predicate, to paraphrase Kant. – Conifold Feb 01 '20 at 21:38
  • @Conifold But it isn't a cheat, it's an argument. Either the future is open or not. If it's not open, then propositions about the future have truth values now, and God must know those propositions about the future. OTOH if the future is open, then propositions about the future lack truth values. If a proposition lacks a truth value, and a necessary condition of knowledge is truth, it follows that propositions about the future cannot be known now. I don't know if free will or true RNGs imply an open future, but either they do or they don't, and in either case God's omniscience can be preserved. – Adam Sharpe Feb 01 '20 at 22:51
  • Exactly, whether they do or they don't depends on moving parts of one's conception, and conception's ability to declare something logically impossible is highly dubious. God is not restricted to time, so relevance of temporality of truth values is questionable, as is the relevance of "true randomness". It is true that theologians often offer "logical impossibility" as an "explanation" for restricting omnipotence and/or omniscience, but the impossibility is of their own conceptual creation, hence not "logical". From neutral position, it is simply a covert way to impose restrictions. – Conifold Feb 01 '20 at 23:03
  • @Conifold I'm not quite sure if you're saying that 1. it's just semantics, or 2. that conceptual analysis is not useful for telling us what is logically necessary/possible. If it's 1., then I agree. The exact definition of "omniscience" doesn't matter as much as what follows from it, and whether we can argue that God exists and has omniscience in the relevant sense under consideration. If you're saying 2., then I would just disagree. Round squares and married bachelors are impossible, because they involve contradictions (when defined in there "normal" way). – Adam Sharpe Feb 02 '20 at 00:20
  • I do not think that I am saying either. I am saying that conceptual analysis of theological arguments shows that justifications of restrictions on omniscience by consistency are structurally flawed, just as the justification of God's existence by "perfection" in the ontological argument. Round squares are nonsensical and if God was only said not to "know" the nonsensical that would be one thing, but "justification-by-consistency" stretches it far beyond that. I am saying that such talk in the sources is more fruitfully rejected and different kinds of theological restrictions discussed upfront. – Conifold Feb 02 '20 at 00:32
  • @Conifold You said: *"...if God was only said not to "know" the nonsensical that would be one thing, but "justification-by-consistency" stretches it far beyond that."* Maybe I'm being dense, but I'm sincerely not seeing the difference. Inconsistencies *are* nonsensical. I'm also not seeing the parallel with the ontological argument from perfection, which I do agree is probably a flawed argument (though I don't know if I'd call the reasons "structurally flawed")... So you're clearly seeing something here that I'm failing to appreciate. – Adam Sharpe Feb 02 '20 at 13:35
  • Logical inconsistencies perhaps, but "God knows what creatures would freely do" is not. One can declare that impossible, but the impossibility is at best metaphysical, not logical. And the next question will be what metaphysical force stands in God's way there. The structural flaw of the ontological argument was pointed out by Kant, it uses existence as a predicate. – Conifold Feb 03 '20 at 00:42
  • One could argue that a being can't be omnipotent if it does not know of everything that there is to be done. Because if I am not cognizant of the concept of, say, snapping my fingers, I can't think of doing it, therefore I can't do it, and therefore I am not omnipotent. – armand Mar 30 '21 at 23:09

4 Answers4

3

Well, the ideas are certainly closely related. I would say that both omnipotence and omniscience are possible only for something which is purely actual, and that which is purely actual must be both omnipotent and omniscient.

Feser argues this at length in Five Proofs of the Existence of God. In Chapter 6, The Nature of God and of His Relationship to the World, in the section on omniscience, he writes:

"...since everything that exists or might exist other than God, and every state of affairs that obtains or might obtain other than God's existence, depends on God's causal activity, all propositions about such things will be true or false only because God causes the world to be such that these propositions are either true or false."

Now, it is due to God's omnipotence that everything depends ultimately on His causal activity. That bit was established in the previous section of the same chapter. Thus, God's omniscience is not to be understood as God being able to discover true facts as they come into being, such that it might be possible for some true fact to evade His notice for some period of time. Instead, nothing at all can become true unless God makes it true. How then could God fail to know any true fact, unless God somehow failed to know what He, Himself was doing?

Perhaps the point you are missing, with regard to power, is that the power that beings have are not unrelated. If you don't comprehend the principality of power (the idea that power must come from a source which has power to give), then you can imagine beings which have power that is independent from God's power. That is impossible, and entails that power can come from nothing. In reality, all power must have a source, and thus power must be sourced in that which just is power - which is God. Thus, God, as the source of all power, must be viewed as cooperating with every secondary cause down the line, at every moment.

  • Welcome! I'm familiar with Feser's *Five Proofs*. I was hoping for a more direct way to show that necessarily, if x is omnipotent then x is omniscient, without having to first assume the Thomistic principles of act-potency, proportionate causality, and so on. Both places I encountered the idea were in "analytic" philosophy of religion papers that weren't specifically Thomistic, so hopefully this is possible. Nonetheless, +1, thanks for the answer. – Adam Sharpe Jan 31 '20 at 21:51
2

Here's an argument that comes close. I have seen it somewhere, but I don't remember where.

  1. If A is omnipotent, A can bring about anything that is logically possible.
  2. A's being omniscient is logically possible.
  3. Therefore: If A is omnipotent, A can bring about his or her own omniscience.
E...
  • 6,436
  • 4
  • 20
  • 39
  • Thank you. I've heard something like that too. A's being omnipotent entails A is possibly omniscient (if omniscience is compatible with A's other properties), but I'd be really interested to know how people argue A's omnipotence entails A's actual omniscience (probably with additional assumptions). – Adam Sharpe Jan 31 '20 at 22:16
  • 1. is debatable, logic restrains the omnipotence of A. – MathematicalPhysicist Mar 31 '21 at 15:36
2

Reviving this because I'm interested. I know Rasmussen provides a (very short) argument for this conclusion on p. 146-147 of How Reason Can Lead to God. But this book was written well after Gellman's paper.

Here’s the steps to his argument:

  1. The necessary cause has maximum power. (As shown by simplicity style reasoning)
  2. The power to know something is a power.
  3. So the necessary cause has the power to know something.
  4. Only minds have the power to know something.
  5. So the necessary cause has a mind.

He then extends the argument to imply omniscience.

Anyone else know where this type of argument has been discussed?

Dean
  • 21
  • 1
  • Welcome to SE.Philosophy! In the Q&A format used on this site, new questions can be posted if you're interested! If it's related, then you can link back to this question. – Nat Mar 30 '21 at 23:44
  • That said, it looks like you've written a valid answer that just happens to also contain your own question that a commenter might answer.. which might work out, just wanted to point out that you could also ask your own question if you'd like to get more info. – Nat Mar 30 '21 at 23:47
1

According to the doctrine of divine simplicity, any of the divine attributes is equivalent (or even identical) to the others (and to God as such). This is a difficult contention to maintain, granted, but it might be part of an argument somewhere that omnipotence entails omniscience.

A partial, more direct argument might proceed: God can do anything, God can only do what It knows how to do,† therefore God knows how to do everything. If this knowledge-how is inherently tied to knowledge-that, then God knows everything simpliciter.

I don't know(!) or even remember if this argument appears anywhere in the literature, though.

†Consider how in Christian theory, God could not simply wish our sins away but had to employ the complicated machinery of the Incarnation and the Atonement to "make a way" for us to be redeemed. Thus our redemption was in God's power but He had to "know how" to effect it, in order to pull it off.

Kristian Berry
  • 9,561
  • 1
  • 12
  • 33