For several years Linda Zagzebski has been exploring the coherence and theological implications of an attribute called omnisubjectivity. In her mind, the theological implications of omnisubjectivity could extend to Christian doctrines such the Trinity, sanctification, and more. For this essay, I wish to consider Zagzebski’s suggestion that omnisubjectivity could be useful for dealing with the particular problem of creation. I will argue that the current models for omnisubjectivity cannot help with the particular problem of creation, but that divine imagination and simulation can be developed in order to offer some minimal assistance for dealing with the particular problem of creation.

This paper shall proceed as follows. In the first section I will articulate the general and particular problem of creation. In the second section I will distinguish creation ex nihilo from eternal creation in order to properly set up the discussion on the particular problem of creation. In the third section I will explain what is omnisubjectivity and consider Zagzebski’s three models for how God is omnisubjective. The fourth section will articulate Zagzebski’s proposal for omnisubjectivity and the problems of creation. I will argue that the proposal is lacking in several serious ways. In the fifth section I will develop some proposals for divine imagination and consider some objections. This will provide proponents of omnisubjectivity with ways to extend their view. This will also provide alternatives for people who think omnisubjectivity is incoherent but that something in the neighbourhood is correct. Finally, I will tie all of the threads together to develop some concluding thoughts on divine imagination and the particular problem of creation.

In contemporary philosophical theology there is a distinction between The General Problem of Creation and The Particular Problem of Creation. Each is related to a specific question. Why would God create anything at all? Why would God create this particular universe? These are interesting questions in their own right but they are connected to a history of serious objections to the foundations of philosophical theology and the rationality of theism. For example, there is a history of atheistic arguments that claim there is no reason for God to create the universe. This lack of a reason to create is argued to undermine inferences that our universe is the product of a divine designer (see, e.g., Bossche, 1998 and Manson, 2013).

The concept of God is that of a perfect being which is the single, ultimate foundation of reality. A model of God seeks to give a unique articulation for what it means to be perfect and in what way God is responsible for everything else in existence. A serious problem arises when theologians try to develop the account of divine foundationalism because there needs to be an explanation for why God produced anything at all. What is often at issue is God’s perfect rationality which demands that God always acts on the basis of objectively good reasons. If a model of God prevents one from identifying a good reason for God to produce something, then that undermines the plausibility of that particular model of God.

Elsewhere I have argued that classical theism is unable to provide an answer to this issue, whereas other models of God can. Classical theism says that God cannot be moved or influenced by external considerations, and that God realizes all values in Himself (see Murphy, 2017, pp. 80–81 and Murphy, 2021, p. 156). This makes it exceedingly difficult to discern what possible reason God could have to create. Whereas other models of God say that God can be moved or influenced by considerations that are external to the divine nature, and further that it is simply incoherent to suggest that God somehow realizes all possible values in Himself (see e.g., Mullins, 2020a and Mullins, 2023). After all, the value of a creator-creature relationship cannot be realized by God all alone in Himself. The potential value of such relationships can be one possible reason for God to create in general. One general motive for God to create anything at all is to bring about the realization of values that would otherwise not obtain.

As interesting as those debates are, they are not the main focus in this article. My focus is on the particular problem of creation. Out of all the possible universes that God could create, why this one? As I have explored elsewhere, God has various options such as creating a theologically deterministic universe or an open universe. Perhaps God could create a Molinist universe (See Mullins, 2024, ch. 6). Maybe God could create a universe that has a presentist ontology of time, or maybe God creates a universe with an eternalist ontology. It may even be that God is eccentric enough to create a morphing block universe in hypertime (for more, see Mullins, 2025). I very much doubt that last possibility, but the point should be clear. God has options, and perfect rationality demands that God has an objectively good reason to create the particular universe we find ourselves in.

Zagzebski has suggested that omnisubjectivity could be explored to help answer the particular problem of creation (Zagzebski, 2013, p. 36). Below I will explore this in more detail and argue that the current models of omnisubjectivity do not help answer the particular problem of creation. Then I will argue that omnisubjectivity can be developed further to deal with certain issues if we extend this attribute to include divine imagination. I need to continue setting the stage before I can do that, but I wish to offer a clarification before moving forward. As I understand Zagzebski, she is not saying that omnisubjectivity can give a full answer to the particular problem of creation. She is merely saying that omnisubjectivity can help fill in the details of an answer. I will explain this more below, but it is important that readers understand how minimal her claim is before continuing.

To help readers better understand the problems of creation, it will be important to keep in mind a common starting assumption in these discussions. Readers need to keep in mind the notion of God all alone prior to any created order. This is because a common assumption in these discussions is the doctrine of creation ex nihilo. The doctrine of creation ex nihilo says that the universe has an absolute beginning, prior to which God was all alone. According to Samuel Lebens, creation ex nihilo is the affirmation that, “The universe was created by God at some point in time (perhaps the first moment in time), before which there was nothing (except God)” (Lebens, 2020, p. 31). Since God was all alone, there is no pre-existent matter from which God produced the universe. To give but one example of this, al-Ghazali says that,

This is distinct from a doctrine of eternal creation which says that God and the cosmos are co-eternal, or that God never exists without a created universe of some sort (Lebens, 2020, p. 31). Again, to give merely one example, the contemporary theologian Thomas Jay Oord offers a series of objections to the notion of God all alone that is taught in the doctrine of creation ex nihilo. He explicitly argues against the doctrine of creation ex nihilo in order to justify his claim that God eternally creates a successive series of universes such that God and the created order are co-eternal (cf. Oord, 2023). This will be important to keep in mind when I consider panentheism later since panentheism rejects creation ex nihilo in favour of an eternal creation.

Both doctrines have a long and interesting history in their own right with philosophers on both sides laying forth their arguments. Famously Maimonides and Aquinas claim that the philosophical arguments for each doctrine are at a stalemate, so the teachings of one’s faith must decide in favour of creation ex nihilo.¹ I am not going to attempt to adjudicate this debate here. My point is that readers need to keep in mind the doctrines of creation ex nihilo and eternal creation when they are considering the subsequent sections of this essay. The general and particular problems of creation typically assume the doctrine of creation ex nihilo. They ask one to consider God all alone prior to creation, and then ask why God would create in general, and why this universe in particular.

There is one final piece of the puzzle that I need to discuss before I can consider Zagzebski’s proposal for the particular problem of creation. What is omnisubjectivity? In Zagzebski’s earlier writings, she claimed that omnisubjectivity is perfect total empathy with all creaturely conscious states (Zagzebski, 2008; 2013; 2016). In her recent book on omnisubjectivity, she has expressed doubts about this. Her current definition of omnisubjectivity is a bit different. According to Zagzebski, “Omnisubjectivity is the property of having a complete and perfect grasp of all actual and possible subjective states, past, present, and future, from the first-person perspective of the possessor of the state” (Zagzebski, 2023, p. 119). To be clear, Zagzebski has in mind God knowing propositional knowledge and phenomenal knowledge. One of the motivations for affirming the property of omnisubjectivity is that it expands omniscience beyond mere propositional knowledge because there is more to know in the world such as phenomenal knowledge.

One of the issues that Zagzebski considers in her more recent work is the question, “How is God omnisubjective?” In earlier work she seemed to assume that empathy was the only answer, but in her recent book she considers three possible answers to this question. There is empathy of course, but there is also perception and panentheism as possible answers to how God is omnisubjective. Allow me to briefly articulate each model.

On the empathy model, God’s omnisubjectivity is perfect total empathy, or “a complete and accurate copy of all a person’s conscious states” (Zagzebski, 2023, p. 66). For any conscious state that one might have, God has a perfect copy of it. For example, God really understands what it is like for you, from your own first-person perspective, to have anxiety. Zagzebski expresses doubt that empathy can deliver omnisubjectivity because there is always a difference between an empath and the one being empathized with. An empath is always aware that she is considering a copy of a person’s conscious states. The empath does not adopt the other person’s conscious states as her own. If I empathize with Zagzebski, I come to understand what it is like for her to be a world’s leading philosopher. I do not come to know, “I am the great philosopher Linda Zagzebski.” Hence in empathy there is always a bit of a gap between the empath and the one empathized with. Zagzebski worries that this gap undermines omnisubjectivity, which is why she wishes to consider multiple models for how God gets the omnisubjectivity (Zagzebski, 2023, pp. 68-69). I, however, am unbothered by this gap. This is merely the logical limit of empathy. If God came to know, “I am the great philosopher Linda Zagzebski, and this is what it is like to be me,” God would not be empathizing with Zagzebski. Instead, God would have lost His mind because God would have confused Himself with Zagzebski. I take this sort of cognitive failure to be impossible for a perfect being. My worry is that Zagzebski is asking for the impossible, but of course, the impossible is not a strike against the empathy model. But I shall set aside my sympathies for divine empathy at the moment.

On the perceptual model, God merely perceives that you have some particular conscious state. According to Zagzebski, this does not really seem to give God full omnisubjectivity since God only knows that you have some particular conscious state. Mere perception does not give God a total grasp of that conscious state (Zagzebski, 2023, p. 72). To develop this thought a bit further, the problem is that God does not get the phenomenal knowledge through mere perception. If you are in pain, and God perceives it, God merely has the propositional knowledge that you are in pain. Nothing about mere perception gets God the full phenomenal grasp of what it is like for you to be in pain at this moment. I think Zagzebski is right to have doubts about this model for how God is omnisubjective, but I will set those aside.

The final model of omnisubjectivity that Zagzebski considers is panentheism (Zagzebski, 2023, p. 73). This is an odd move since panentheism is a model of God that is incredibly difficult to demarcate. Panentheism has the catchy slogan that the universe is in God, but God is more than the universe. The problem is that it is not at all clear what this means in a non-metaphorical way that clearly demarcates panentheism from its rivals (cf. Mullins, 2016). In fact, I can find nothing in Zagzebski’s writings that help me figure out what panentheism is, nor anything that can help illuminate how God is omnisubjective. Even further, Zagzebski has admitted to me that she hopes I can figure out what panentheism is so that she can discern if it helps the case for omnisubjectivity.²

One problem is that panentheists seem to disagree on just about everything. For example, Avicenna affirms that God is completely unmoved by the world. That does not sound friendly towards omnisubjectivity. In fact, many panentheists affirm that God is impassible and in a state of pure blissful contemplation of Himself. Again, this is not remotely congenial towards omnisubjectivity. Other panentheists, such as Oord, are happy to affirm that God has a perfect grasp of all creaturely conscious states and feels everything along with creation. Of course, notice here that Oord is affirming omnisubjectivity by way of empathy and omniscience. Nothing about panentheism is doing the explaining for how God is omnisubjective. Instead, the affirmation of God’s radical empathy is what is intended to demarcate Oord’s model of God from something like Aquinas’ model of God.³

So what exactly is panentheism? To be honest, I don’t really know. Elsewhere I have suggested that the only way to clearly demarcate panentheism from its rivals is that it rejects creation ex nihilo in favor of a doctrine of eternal creation (cf. Mullins, 2023). This is intentionally ignoring the fact that one contemporary panentheist says he affirms creation ex nihilo (e.g. Clayton, 2008). But such outliers must be ignored in favor of the majority view among panentheists who explicitly reject creation ex nihilo (cf. Gocke, 2017). Otherwise, no progress can be made for figuring out what panentheism could possibly be.

With that being said, I now have three models of omnisubjectivity to consider. This will help readers grasp what Zagzebski’s proposal is for dealing with the particular problem of creation.

In her earlier writings, Zagzebski claimed that omnisubjectivity can help tackle the general and particular problems of creation. Again, my focus is on the particular problem of creation. Since Zagzebski articulates this proposal in her earlier work, readers should keep in mind that she is assuming the empathy model of omnisubjectivity. I will argue that the empathy model does not help with the particular problem of creation. Then I will consider her other models of omnisubjectivity to see if they fare any better. I argue that they do not help, but in the next section I will make a suggestion for what will help.

In her earlier work, Zagzebski takes omnisubjectivity to be God’s total perfect empathy of all creaturely conscious states. Empathy is gained by having a direct cognitive acquaintance with another person. Thus it seems a bit difficult to figure out how empathy could help God decide which kind of universe to create. When God is deciding, there are no creatures in existence for God to empathize with.

Zagzebski is not deterred by this problem, and suggests that God might have counterfactual or hypothetical omnisubjectivity. Careful readers will have noticed that the definition given above from Zagzebski’s recent book has this counterfactual cognitive grasp built into the definition of omnisubjectivity. What Zagzebski does in her earlier work is try to defend this position. So allow me to elaborate on her defense.

Any Molinist might initially find this attractive because it looks like an extension of middle knowledge. This is divine empathy with merely possible conscious states of merely possible conscious creatures. Zagzebski claims that this kind of power better captures God’s cognitive perfection (Zagzebski, 2013, p. 36). Not only that, she thinks that it would help us make sense of God’s creation and providential control. She writes,

On standard accounts of God’s omniscience and decision to create, theologians only speak of God knowing the truthvalues of the propositions about those possible universes. God is said to know all of the descriptive facts about those possible worlds, but there is hardly any mention of God knowing the phenomenal facts about what those worlds are like. Zagzebski takes this to be a cognitive deficiency that is unbefitting of God. She says,

When it comes to discerning which particular universe God should create, Zagzebski says that God’s counterfactual omnisubjectivity must come into play. Again, in her earlier work she is assuming the empathy model for how God is omnisubjective. The suggestion is that God’s empathetic knowledge will help Him compare and contrast the different possible universes that He might create. Without such knowledge, Zagzebski says she can’t see how God could adequately compare the universes. She asks us to imagine that God lacks counterfactual omnisubjectivity. In that case, God “could not know what it would be like for him to create such a world and to have an empathic awareness of it. This not only puts serious limitations on a creator God, but also limits his knowledge of his own counterfactual self” (Zagzebski, 2013, p. 37-8). This, says Zagzebksi, is a serious problem.

She doesn’t say much about how all of this works, but she does give some suggestions. For example, she asks us to consider the notion that God rewards and punishes free creatures. Zagzebski says that if God does not know in advance what it would be like for a person to have certain subjective experiences, then it is difficult to see how God could adequately have a plan for rewarding and punishing people (Zagzebski, 2013, p. 38).

The careful reader will notice that this is not a full answer to the particular problem of creation. As I said earlier, I don’t think Zagzebski intends it to be a full answer. Instead, omnisubjectivity is meant to assist with the answer. Omnisubjectivity gives some of the details about how God considers different possible universes. Grasping such phenomenal knowledge is meant to supply God with unique reasons in favor or against any particular potential universe. So allow me to extend Zagzebski’s proposal on her behalf before offering my critique.

Any answer to the particular problem of creation will need to specify what God’s goal is. In my previous discussions of the particular problem, I consider something called God’s most central purpose (Mullins, 2024, p. 82). This is God’s goal to enter into friendship with as many humans as possible. The idea is that this is not God’s only goal for creation, but it is an incredibly significant goal that God intends to satisfy. Imagine that this is in fact’s God’s most central purpose for the universe. If that is God’s goal, that will significantly narrow down the range of potential universes for God to consider creating. It should be obvious that God will no longer consider creating universes in which human life will not be possible. Instead, God will consider potential universes in which human life will be inevitable. Imagine further, that God is considering a range of different universes in which human life is inevitable. How can God whittle down His options even more? This is where Zagzebski’s proposal could come into play. It could be that God considers the phenomenal states of potential conscious creatures in each universe, and those phenomenal states supply God with unique reasons to select a particular universe. Perhaps one of the potential universes is filled with an amount of suffering that is far too unacceptable, or maybe another universe has a greater variety of unique human experiences than others. These could be important reasons that factor into God’s decision to create one particular universe. Again, it is not the full answer to the particular problem of creation, but it certainly gives more detail to the story, which is something desperately needed in philosophical theology.

It seems to me that Zagzebski is right in suggesting that this kind of phenomenal knowledge is important for answering the particular problem of creation. However, what are we to make of all of this? I have my doubts that this is coherent. Again, this proposal was developed with the empathy model for how God is omnisubjective. Omnisubjectivity here is total perfect empathy. The problem is that empathy is achieved by some sort of direct cognitive acquaintance with another conscious being. In the scenarios that Zagzebski is describing, there are no conscious creatures for God to empathize with. So this entire suggestion seems to be a non-starter. Empathy does not help with the particular problem of creation because there are no creatures to empathize with prior to creation when God is making His decision. If there are creatures for God to empathize with, omnisubjectivity comes too late to assist God in making a creative decision.

Since in later writings Zagzezbski explores other models for how God is omnisubjective, these are worth taking into consideration as well. So next consider the perception model for how God is omnisubjective. Does this help with the particular problem of creation? No. Perception does not help because this also presupposes the existence of creatures. Prior to creation, there are no creatures to perceive. If there are creatures for God to perceive, omnisubjectivity comes too late to assist God in making a creative decision. Moreover, I don’t know what it could mean to literally perceive the phenomenal states of hypothetical creatures. So this model also sounds like a nonstarter.

What about panentheism? There are several problems here in terms of answering the particular problem of creation. The biggest problem is that panentheism does not help because panentheism is incredibly fuzzy. Again, Zagzebski offers no clear model of panentheism in her writings. As stated before, omnisubjectivity does not seem to be doing anything to help explain what panentheism is, nor does panentheism seem to help illuminate what omnisubjectivity is. Another problem is that omnisubjectivity does not help here to explain why God must create in general, nor why God must create a particular universe.

In my own articulation of how panentheism can answer the particular problem of creation, omnisubjectivity plays no role. As mentioned above, I think the clearest way to demarcate panentheism from its rivals is to reject creation ex nihilo in favour of an eternal creation. God always exists with a created universe of some sort. There is no state of affairs where God is all alone to consider His options. What this means is that, once again, omnisubjectivity would come too late to factor into God’s decision to create any particular thing. God is eternally creating, and thus God is eternally grasping a variety of creaturely conscious states. So it seems that omnisubjectivity cannot do the explaining for why God creates this or that thing.

To see this further, panentheists typically have a different explanation for why God would create in general, and this explanation answers the particular problem of creation. One standard panentheist answer is to appeal to the diffusiveness of the good (Mullins, 2023). This means that God is the greatest good and as the greatest good must produce as many other good things as possible. Why does God create in general? Because goodness must produce more good things. Why does God create this particular universe? Because goodness must produce all other possible good things. This is a nice and tidy answer to the general and particular problem of creation. However, notice that nothing about grasping potential or actual creaturely conscious states does any explanatory work. All of the work is being done by goodness.

Where does this leave us? All of Zagzebski’s models for how God is omnisubjective fail to help with the particular problem of creation. This might seem to be bad news for omnisubjectivity, but I have a proposal that might help.

I want to revisit Zagzebski’s current working definition of omnisubjectivity. Zagzebski says, “Omnisubjectivity is the property of having a complete and perfect grasp of all actual and possible subjective states, past, present, and future, from the first-person perspective of the possessor of the state” (Zagzebski, 2023, p. 119). The problem for Zagzebski is how God gets this grasp of actual and possible subjective states. As Zagzebski acknowledges, empathy and perception can only help fill in the details for actually existing creaturely conscious states (Zagzebski, 2023, pp. 131-2). I suppose God’s memory can help fill in the details for past creaturely subjective states. One serious problem that Zagzebski keeps encountering is how to fill in the details for possible and future subjective states. She suggests that divine imagination might help, but laments that there has been little serious reflection on divine imagination (Zagzebski, 2023, p. 133). I think I can fill in some of those details on her behalf. The issue with my account will be the complete grasp. If Zagzebski thinks that empathy is a problem because there is a gap between the empath and the one empathized with, then she will not like the gap that I will have in the account below. However, just as I maintained that Zagzebski is asking for too much from empathy, I maintain that any complaint against what I am about to say is also asking for too much. There are logical limits to what can be grasped, and I believe that my account fills God’s knowledge to the logical limit. What is needed for God to be omnisubjective is for God to have a variety of cognitive powers that give Him the capacity to undergo experiences and simulate or imagine potential experiences. Allow me to explain.

Omnisubjectivity is an interesting attribute, but I think it is grounded in something more fundamental, namely, God’s maximal cognitive excellence. My suggestion is that how God is omnisubjective is explained by identifying the different cognitive powers that God has. God is omnipotent which I take to be the maximal set of power granting abilities which logically excludes all liabilities. One set of power granting abilities is maximal cognitive excellence. Maximal cognitive excellence covers God’s cognitive powers related to beliefs, emotions, and rationality. As I see it, God has the power to develop true beliefs in the right way, and is always appropriately responsive to objectively good reasons. God’s power to develop true beliefs covers not merely propositional knowledge but also phenomenal knowledge as He experiences new things over time. This requires that God is a temporal being, but I know of no good reason to reject divine temporality, so I see no problems with this. Emotions are felt evaluations of a situation, and God’s emotions are always appropriately fitting to any given situation. God is sad or angry when humans commit sins, and God is delighted when sinners repent. These are rationally fitting emotions to have, so a perfectly rational being will have them when it is appropriate.

I believe that God’s cognitive powers entail the ability to empathize with others, just as Zagzebski argues for in her work. This is because empathy is an epistemic state that one achieves, and a maximally excellent cognitive being will have the power to achieve this. However, I also believe that maximal cognitive excellence entails the power to imagine or simulate propositional and phenomenal states. The traditional doctrine of divine ideas says that God necessarily and eternally knows all of the possible creatures that He could produce. This is typically understood to be that God has the idea of all possibilities (cf. Ward, 2020). My suggestion is that the divine imagination be extended to include God knowing all possible phenomenal states. I don’t think this is too much of a leap for your average theist to make. If one is willing to accept that God knows all of the logical possibilities, it seems like a very small step to say that God also knows all of the logically possible phenomenal states. If I can imagine some possible phenomenal states, then surely God could imagine all possible phenomenal states.

What are the consequences of this? As I see it, this introduces a difference between what God imagines and what the experience will actually be like. As Zagzebski points out, we often imagine something, and after we experience it we can say, “That is what I thought it would be like” (Zagzebski, 2023, p. 133). Some might worry that this creates a gap in God’s knowledge because God’s imagination of what it would be like is not the same as the experience itself, but I say that this alleged “gap” is just the logical distinction between what is possible and what is actual. God can imagine what it would be like to create this or that universe. Imagining what it would be like to create universe X is a distinct phenomenal state from what it is like to actually create universe X. As perfect in cognitive power, God can tell the difference between a simulation and the actual experience. So whatever this “gap” is, it is simply a logical limit, and that is no strike against omnisubjectivity.

Where does that leave us? Here is my stance. Prior to creation, God may be able to simulate what it would be like to create a universe, but that is not the same as actually experiencing what it is like to create a universe. Prior to creation, God just cannot know what that actual experience is like because that is the logical limit of imagination. The same goes for God contemplating His ideas and simulating all the possible combinations of conscious experiences. It is one thing to contemplate the idea of Socrates’ emotional life, and quite another to actually empathize with the emotional life of an actually existing Socrates. A cognitively perfect being can tell the difference between a simulation and actuality. As far as I can tell, at best, simulation would only give God phenomenal knowledge of what it might be like for a creature to experience such and such. It would not give God knowledge of what it would actually be like for a creature to experience such and such. Subsequent to the act of creation, the cognitively perfect God can perceive and empathize with creatures that actually exist.

Does this fully answer the particular problem of creation? No. Omnisubjectivity, as Zagzebski says, is only a way of filling in some of the details about God’s decisionmaking process. While I think that divine imagination and simulation may help some with God’s decision about what to create, I think that this suggestion only goes so far. This is because God could simulate what it would be like to create Calvinist, Molinist, and open universes. So whatever limited benefit simulation might have for the particular problem of creation, it does not obviously point God in any particular direction. The main driving force for answering the particular problem of creation is whatever God’s goal is, as I suggested above. Omnisubjectivity can only fill in some of the details, and it can only do so if one affirms that God has the power of imagination. This prior imagination of God would certainly help God consider the different possible worlds to create that would best satisfy whatever His particular goal is for creation. This prior imagination would supply God with unique reasons to consider because grasping the potential phenomenal states of a possible universe are distinct reasons from merely grasping the propositional content of a possible universe.

In this paper I have considered Zagzebski’s attempts to articulate what omnisubjectivity is, how God is omnisubjective, and the implications for answering the particular problem of creation. I have argued that Zagzebski’s work can be extended and developed in interesting ways in order to fill in more details to one’s philosophical theology. In particular, theologians should develop an account of divine imagination in order to maintain omnisubjectivity or something close enough, and then fill in some of the details for the answer to the particular problem of creation.