ADS Senior Thesis II: ADS, Omniscience and Genuine Possibilities


Now that we have a “rough-and-ready” understanding of ADS and some idea of the motivation for believing it, we will transition to Christopher Hughes’s critique of the doctrine. The first argument I will examine concerns God’s omniscience and his knowledge across possible worlds. The argument goes something like this: Because God is a necessary being and because he is essentially omniscient, he must know all truths in any given possible world. Because what is true differs across possible worlds, God’s knowledge must also differ across possible worlds. However, if God’s knowledge differs across possible worlds, then surely his beliefs must also differ across possible worlds. Because belief-states are, intuitively, intrinsic properties, it seems that either God’s intrinsic properties must differ across possible worlds, thus implying that God has intrinsic accidental properties, or one must assert that all truths in the actual world are necessary and deny that there are other genuine possible worlds. Obviously, if premises are true and the argument valid, then it seems the most reasonable horn of the dilemma to take hold of is the first, where God has intrinsic accidental properties. This would entail the falsity of ADS, so it seems the proponent of ADS has a serious problem on her hands.

One possible defense from this objection can be found in William E. Mann’s paper, “Simplicity and Immutability in God.” Mann examines a very similar problem to the one that Hughes addresses. Mann points out that the proposition, “Pharaoh is unmoved by Moses’s pleas at t2” (call this P) is distinct from the proposition “the Egyptians suffer a plague of locusts at t3” (call this L). A dilemma can be created here such that either God’s knowledge P is identical with his knowledge of L, or the two items of knowledge are distinct items of knowledge for God. It seems that intuitively, these are different items of knowledge because of the fact that they are different propositions. For us this seems to surely be the case, so the same must go for God as well. The problem is that if God is absolutely simple, it seems that He can, at most, know only one thing, and so He is not omniscient. While this problem is importantly different from the one Hughes raises, Mann’s solution is very interesting and possibly applicable to Hughes’s problem. Mann’s solution is to say that “while the content of God’s knowledge may be unlimited…(ADS) need only insist that the activity by which He knows is simple.”

So can Mann’s solution be applied to Hughes’s problem successfully? I say no. The reason is that such an answer would fail to grasp the heart of the problem. The heart of the problem is that knowledge implies (among other things) belief-states, and belief-states are intuitively intrinsic properties. So while it may be true that the activity by which God knows does not change across possible worlds, His corresponding belief-states surely will. This denies the fundamental tenant of ADS that God has no intrinsic accidental properties.

Another possible answer by the proponent of ADS can be found in William F. Vallicella’s article, “Divine Simplicity: A New Defense.” Vallicella characterizes the problem in a very similar way to Hughes. In his own words: “Simply put, the content of omniscience might have been otherwise. Hence God cannot be identical with the property of knowing the conjunction of (what in fact are) all truths.” However, Vallicella does not see this problem as especially troublesome. He suggests that the simplicity theorist admit the existence of accidental properties in God, and restrict the the claim of the identity of God with His properties to the ones God has essentially. Surely one can make such a move, but one might wonder what Vallicella has left after giving up this ground to the antagonist of ADS. It seems that he’s saved Divine simplicity at the cost of the traditional doctrine of ADS. Thus, his answer is surely an option for the proponent of ADS to deal with Hughes’s argument, but only if the proponent of ADS wishes to completely concede Hughes’s point and give up on the traditional doctrine of ADS.

It seems like the only option left for the proponent of ADS who does not wish to bite the bullet like Vallicella is to try to deny that a different beliefs across possible worlds would amount to God having different intrinsic properties. In fact, Hughes does entertain the suggestion. He looks at two stories provided by Tyler Burge meant to motivate intuitions in the direction favorable to the proponent of ADS. The first story involves a man who has a notion of arthritis, but mistakenly believes that arthritis can in occur in other areas besides the joints. Next, Burge invites us to consider a counterfactual situation in which the man is altogether the same, except for the fact that he is part of a linguistic community in which the word arthritis can be correctly used to describe painful inflammation of other parts of the body besides the joints. In the actual world the man falsely believes that he has arthritis, but in the counterfactual world, he lacks that belief due to the fact that he lacks the concept (arthritis does not mean the same thing in the different worlds). This example is meant to show that differences in the social environment of a creature can make a difference to what a creature’s beliefs are, and intuitively a difference in social environment implies a difference in extrinsic and not intrinsic properties.

Another story Burge tells involves the infamous “Twin Earth.” Twin-Earth is like earth in nearly every respect, except that instead of water, Twin-Earth has a phenomenologically exactly similar liquid called “twater.” Suppose that on Earth, Adam has the belief that water is a liquid, and on Twin-Earth, Adam* (who is an exact duplicate of Adam down to his physical and experiential history) lacks this belief. It seems that, once again, differences in the environment of a believer can make a difference in the beliefs of a believer without making a difference in the believer’s intrinsic properties.

Hughes admits that these stories do something for the plausibility of the contention that a change in God’s beliefs may not imply a change in His intrinsic properties. Hughes notes the difficulty here lies “in making sense of the role that, on the Thomistic picture, purely extrinsic properties play in fixing what God believes.” Hughes points out that this means we are to suppose that God’s belief that Pluto exists must consist in His one, necessary internal state, coupled with extrinsic relation R, and we are also supposed to think of God’s belief that Pluto does not exist as his necessary internal state, coupled with a different purely extrinsic, relational property R*. The obvious question here is just what kind of properties could R and R* be? Could something like the extrinsic properties Burge mentions be good candidates? The answer is no. First of all, I would point out that the cases aren’t exactly analogous. The Burge cases involve a difference in beliefs only on the basis of the believer lacking a certain concept. But if God is omniscient, surely He lacks no concept across any possible world. Furthermore, as Hughes points out, for the Thomist, God could never be on the receiving end causally, and in each Burge case, “the extrinsic properties are the ones that involve being causally affected in the right sort of way, either by what the belief is about, or else at least by other members of the believer’s linguistic community.” Thus, the Burge cases don’t provide us with the right kind of extrinsic relational properties.

Nor does it seem to Hughes to make sense to claim that the extrinsic relational property is merely the state of affairs in which Pluto exists. If this were so, then God’s belief that Pluto exists would consist in merely being in His necessary internal state, coupled with the state of affairs in which Pluto exists. Because God would have been the same had Pluto not existed, God’s internal state is neutral to the equation. The problem for Hughes is that the constituents of the equation do not seem to be able to add up to the sum; namely, that God believes that Pluto exists.

Here, Hughes points out that the Thomist will probably respond by pointing out that for Aquinas, God’s believing that Pluto exists consists both in his necessary state and His causing Pluto to exist by willing it with His efficacious will. However, given that Pluto intuitively exists contingently, one would have to explain the how this would work in relation to the Divine Will. One would encounter the very same problem, just moved from God’s omniscience to the Divine Will, because if the Divine Will is identical with God’s absolutely simple necessary state, then one would have to appeal to an extrinsic relational property to explain how God’s willing Pluto into existence could be contingent. What kind of extrinsic relational property could fill this role?

Finally, let us consider a suggestion by Alexander Pruss from his lecture, “On Three Problems of Divine Simplicity.” Pruss asks us to consider a possible theory of mind in which there is a sort of “mind’s eye” that contemplates the phenomena before it, yet is totally distinct from the phenomena. In the case of the sensations of hot and cold, the difference is located in the phenomena and not in the mind’s eye. As Pruss says, “One state involves my mind’s eye standing in a relation of contemplation to a feeling of heat and the other involves its standing in that relation to a feeling of cold.” The difference between these two states seems to be extrinsic to the mind’s eye in this case. Pruss wants to here ask, “why not think that the way God knows things is something like this?”

Here I have nothing perfectly conclusive to say because it is hard to say how clear our intuitions are about whether such a view of God’s mind is possible. I do have a few thoughts on the suggestion however. First, this view obviously entails that God does not have beliefs, and one might wonder in what sense one could have knowledge without belief considering the classic characterization of knowledge as justified true belief (with whatever is necessary to satisfy Gettier of course). Dallas Willard actually presents a very plausible characterization of knowledge without reference to belief in his paper, “Knowledge and Naturalism.” Willard takes “knowledge in the dispositional sense to be identical with the capacity to represent a respective subject matter as it is, on an appropriate basis of thought and/or experience,” and “in the occurrent sense it consists in actually representing, at a point in time, the respective subject matter as it is, on an appropriate basis of thought and/or experience.” Although this account of knowledge does not make reference to belief, it still will not accommodate the theory of mind and knowledge provided by Pruss because Willard’s account requires that the mind represent reality, and therefore the mind will be in different representational states depending on the various things it knows. These states will surely be accidental intrinsic properties of the mind. It seems very difficult to think of any account of knowledge that Pruss’s suggestion about God’s knowledge could fit into. There is a worry that calling what God has knowledge will be an equivocation because intuitively, knowledge involves the mind representing reality correctly. Furthermore, it seems that there are certain types of knowledge that God has that are difficult to fit into Pruss’s model. Namely, any type of knowledge of other possible worlds and other types of counterfactuals will present a difficulty because they do not exist as phenomena to be observed directly by the mind’s eye. It seems difficult for God’s mind to perceive these realities because things like possible worlds are (arguably) something like a maximal proposition and propositions seem to be the kinds of things that are only known via their representational qualities. Since God’s mind (for Pruss) cannot represent anything, it cannot represent reality as being some other way. It merely directly perceives the truths in front of it. One might here suggest that God could know possibilia through direct perception of the potentialities in His own will, but this will not work for the Proponent of ADS because there are no unactualized possibilities for God’s will because He is actus purus; there is no composition between potentiality and actuality in God.

Thus, it seems that there are some serious problems for the proponent of ADS in explaining how God can be omniscient if there are other genuine possible worlds. I have not completely defeated the type of response Pruss gives, but I have at least pressed it to the point that the defender of the Prussian suggestion has to explain why we should call what God has knowledge, and how God could know truths about possible worlds without the ability of God’s mind to represent the world in any way.


5 Responses to “ADS Senior Thesis II: ADS, Omniscience and Genuine Possibilities”

  1. Grail Seeker Says:

    Excellent article and I agree 100%. I have a question here:

    ***Because what is true differs across possible worlds,***

    Why must truth differ across possible worlds?

  2. ZSDP Says:

    Grail Seeker –

    Thanks for stopping by and taking the time to read Mark’s thesis. I think I should be able to answer your question.

    So, why would truth differ across possible worlds?

    In order to answer, we must distinguish between at least three separate things: (1) truth as such, (2) particular necessary truths, and (3) particular contingent truths. Truth as such and particular necessary truths (e.g. God and His character, the definition of a triangle as a three-sided polygon with three vertices, etc.) will remain static across possible worlds because, frankly, they could be no other way. These are not the truths Mark is referring to in the fragment you quote. Rather, he is referring to particular contingent truths. Such truths are things that could be other than we know them to be, possibly including whether the earth is inhabited by talking rabbits, the spectrum of light that is visible to humans, whether I eat a sandwich or borsch for lunch, etc. None of these things (at least arguably) necessarily has to be the case. Because they could be different in some other possible world, the composite set Mark casually designates “what is true” could also be different in that possible world.

    Does that help?

  3. Perry Robinson Says:

    ah, a man after my own heart…

  4. Grail Seeker Says:

    I meant to get back to you earlier. That helped nicely.

  5. ZSDP Says:

    Great. Glad to hear it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: