The Creator/Creature Distinction in Debates over Models of God

Abstract

In contemporary theology, one can find an array of models of God to choose from. There are various types of arguments that one can put forth in order to reject one model, or to defend another model. In this paper, I wish to examine one popular type of argument. This argument typically says that one should reject a model of God that blurs the creator/creature distinction. The problem with this kind of argument is that it usually begs the question by presupposing one’s own model of God in the creator/creature distinction. In other words, the argument basically boils down to, ‘I reject your model of God because it is not my preferred model of God.’ On many other occasions, the arguments beg the question by confusing additional metaphysical doctrines with the concepts of creator and creature. In this paper, I shall examine several different versions of the creator/creature distinction argument in an effort to find a version that is not question begging. I shall begin with articulating what I take to be the basics of a Christian understanding of the creator/creature distinction. Then, I shall consider several recent attempts from classical theists to argue that a rival model of God violates the creator/creature distinction. In each case, I shall find the arguments wanting. First, I shall examine James E. Dolezal’s attempt to argue that non-classical models of God violate the creator/creature distinction. I will argue that Dolezal’s attempt is question begging because it sneaks in his own model of God and several questionable metaphysical assumptions into the concepts of creator and creature, and thus goes beyond the basic creator/creature distinction without justification. Second, I shall examine a recent attempt by Kevin J. Vanhoozer to argue that Thomas J. Oord’s panentheistic model of God violates the creator/creature distinction. I will argue that Vanhoozer’s attempt suffers from several conceptual errors.

1. What I Take the Creator/Creature Distinction to Be

The goal of this paper is to find a legitimate form of the creator/creature argument by examining some potential candidate arguments. As I understand it, a creator/creature argument says that a model of God is incoherent because it confuses God with a creature. I gather that the argument form is something like this.

(CC) If a model of God implies that God is a creature, then it is an incoherent model because God is a creator and not a creature.

This is a very common form of argument in Christian theology, but I am skeptical about most versions of the argument. It seems to me that most versions of the argument are question begging because they build in their preferred model of God into the concept of creator, and then use their concept of creator to justify their model of God. Alternatively, most versions seem to build in their preferred metaphysical doctrines into the concept of creator, thus confusing the concept of creator with their preferred metaphysical doctrines (Kvanvig 2021, pp. 190–91). It seems to me that it would be quite useful for theological debates if one can identify a legitimate version of the creator/creature argument. In order to identify a legitimate version of this argument, one must first answer several questions. What is the concept of God? What is a model of God? What does it take to be the creator? What does it take for a model of God to satisfy this definition of a creator? What does it take to be a creature? What does it take for a model of God to imply that God is a creature? In this section, I shall answer these questions. In subsequent sections, I shall examine different attempts to articulate a legitimate creator/creature argument.

What is the concept of God? The concept of God is that of (a) a perfect being which is (b) the single ultimate foundation of reality (Gocke 2017, pp. 1–2). Clause (b) is sometimes referred to as divine foundationalism. Divine foundationalism is the thesis that God is the first cause and fundamental ground—the diachronic and synchronic source—of all things that are distinct from God (Bohn 2019, p. 2). In other words, God is the sole asymmetrical source of all else (Kvanvig 2021, p. 11).

What is a model of God? A model of God is a particular coherent set of unique claims about the nature of God and the God-world relation. A model of God starts with the basic concept of God, and then develops a thick conception of God by identifying various essential attributes that God has in order to explain why God is perfect, and in what way God is the single ultimate foundation of reality. In other words, a model of God is a particular extension of the concept of God.

A model of God is not a fully developed philosophical or theological system, or what one might call a worldview. According to Benedikt Paul Göcke, ‘a worldview is a set of assumptions in the form of a unifying picture, or narrative, that shapes the way in which each one of us understands what is going in our lives and the world.’ We draw upon a particular model of God in order to justify our worldview, but the worldview is a much more robust picture of reality than a model of God (Gocke 2018, p. 167).

This distinction between a model of God and a worldview explains why one can find a particular model of God in multiple religious traditions, and in diverse theological systems from within a specific religious tradition. For example, it is often held that classical theism is a model of God found within different religious traditions. It is said to have proponents like Maimonides, Augustine, and Aquinas. These thinkers are said to have the same model of God, and yet their specific religious traditions vary, and their individual theological systems are quite diverse. Open theists also claim to be represented in different religious traditions, though they are most prevalent in Christianity. According to the open theist Richard Rice, open theism itself is not a systematic theology. Instead, Rice maintains that open theism makes certain unique claims about God and the God–world relation that can provide a basis for developing a theological system (Rice 2020, pp. 135–36).

My interest in this paper is with models of God that can be found within Christian thought. I shall say more about the different models of God that I have in mind in the next section. At this point, I shall focus on the minimal claims that all models of God should be able to agree upon. I say that God is a necessarily existent being who exists a se, self-sufficient, maximally powerful, maximally knowledgeable, maximally good, perfectly free, and eternal. I shall assume that readers are familiar with standard definitions and analyses of God’s power, goodness, and knowledge, so I shall say focus on these other divine attributes (Swinburne 2016; Mawson 2018). God exists necessarily in that God cannot fail to exist. God’s eternality follows from His necessary existence. To exist eternally is to exist without beginning and without end. Necessary existence and eternal existence are distinct from aseity. God exists a se if and only if His existence is not asymmetrically dependent upon nor derived from anything external to God. God is self-sufficient if and only if His essential nature is not asymmetrically dependent upon nor derived from anything external to God. One way to think about this is that God’s possession of essential attributes like maximal power, goodness, knowledge, and freedom are not dependent upon anything external to God.

The next question that I wish to consider is what it means to be the creator. One might wonder if divine foundationalism captures all there is to being the creator. I say that it does not capture the concept of being the creator. This is because divine foundationalism is a more fundamental concept than that of being the creator. In order to see this, consider the distinction between emanation and creation (Burrell 1993, p. 7; Brown 2016, pp. 18–19). As I understand it, emanation is when something distinct from God necessarily follows from the divine nature. Whereas creation is when God freely causes something distinct from Himself to exist (Burrell 1993, p. 7; Blowers 2012, p. 186; Garcia 1992, p. 192). Emanation and creation are both consistent with divine foundationalism in that both are extensions of the ways that God can be the foundation of reality. For the sake of understanding the creator/creature distinction, it is necessary to distinguish emanation and creation because these are claims that go beyond mere divine foundationalism.

In order to clarify the difference between emanation and creation, consider the distinction between worlds and universes, and a distinction between world-actualization and creation. As I understand it, a possible world is a maximally consistent proposition that is best captured by modal logic. Such propositions express the entire way things could be. A maximally consistent proposition will contain an ontological inventory of all things that exist within a world, and the relations that obtain between those objects. This maximal proposition will also include the entire history of a world’s timeline, if that particular world contains a timeline. The actual world is a maximally consistent proposition that expresses the entire way things are. Worlds are distinct from universes. A universe is a smaller domain within a world. A universe is a collection of contingently existent substances which are spatiotemporally related to one another. This is why one finds theists talking about a possible world where God exists without any universe of any sort, or a possible world in which God exists with a universe. Different models of God disagree over whether or not God can exist without a universe.

With this distinction between worlds and universes in hand, I can turn to the difference between world-actualization and creation. Creation occurs when God freely causes some contingent substances to exist. For my purposes, I shall primarily focus on creation as God freely causing a universe to exist, though it is logically possible that God create a single contingent being instead of an entire universe. According to Klaas J. Kraay, world-actualization is different from creation. World-actualization need not involve any causal activity on God’s part because the mere existence of God entails world-actualization (Kraay 2015, pp. 4–5). To state this in other terms, the mere existence of God entails that there is an entire way that things are—i.e., God exists with a particular nature.

Allow me to clarify a point before moving forward. That there is a world of some sort is necessary because God necessarily exists, and world-actualization simply follows from the way things are. One can see world-actualization as a kind of emanation. However, there may be other kinds of emanation. On some models of God, God’s nature necessitates the existence of one universe, multiple universes, or all possible universes. These would be emanations as well. On models of God that affirm a creation, however, a universe does not necessarily exist because the existence of a universe depends upon the voluntary exercise of God’s power. A creation occurs when God voluntarily exercises His power to cause a universe to exist. This could be an eternal creation in which God is freely and eternally causing a universe of some sort to exist, and whatever God is eternally causing to exist is co-eternal with God (Hoover 2019, p. 122; cf. Oord 2015). On an eternal creation, there is no state of affairs where God exists without some sort of created beings. Things are different on a doctrine of creation ex nihilo. On creation ex nihilo, God and the created order are not co-eternal, the created universe has a definite beginning to its existence, and there is a state of affairs where God exists without the universe (Broadie 2010, p. 53; Burrell 1993, p. 7; Brunner 1952, pp. 14–15; Fergusson 2014, p. 40; Leftow 2012, p. 4). This could be a timeless or temporal state of affairs in which God exists alone, depending on the details of one’s preferred model of God. On both creation ex nihilo and eternal creation, God is said to freely create a universe of some sort, despite their differences on other issues.

With this cleared up, I shall attempt a definition of what it means to be the creator.

Creator: God is the creator if and only if God freely causes a universe to exist.

Again, emanation and creation are both consistent with, and extensions of, divine foundationalism. So, this definition of what it means for God to be the creator presupposes the truth of divine foundationalism.

If that is what it means to be a creator, then what does it mean to be a creature? All it takes to be a creature is to have one’s existence be ultimately caused by the will of God. A creature is any contingent substance that has its ultimate cause in God. If the existence of some contingent substance is ultimately caused to exist by the will of God, then that substance is a creature.

Creature: A contingently existent substance C is a creature if and only if its existence is ultimately caused by the free will of God.

I take it that this is essential to the concept of being a creator and being a creature. These are concepts that can fit nicely within different metaphysical and theological systems. Anything else that one might try to build into these concepts would seem to be an unnecessary, or perhaps question begging, add on to the concept of a creator or a creature. It would be to smuggle one’s metaphysics and theology into the concept without justification. If one wishes to build off of these concepts in order to develop doctrines like eternal creation or creation ex nihilo, that is acceptable and not question begging. In this case, this is to take a concept like creation, and develop a thick conception of creation into a more specific doctrine.

That is what it means to be a creator and a creature. What would it mean for a model of God to violate the creator/creature distinction? I take it that any model of God that somehow entails God satisfying the definition of a creature would violate the creator/creature distinction.

With this distinction cleared up, I shall articulate some of the rival models of God that are on offer in the contemporary debates. Then, I shall examine some attempts to argue that particular models of God violate the creator/creature distinction.

2. Models of God

In the contemporary debates, there are multiple rival models of God. I shall focus my attention on four of the most common within Christian thought: classical theism, neoclassical theism, open theism, and panentheism. All four of these models affirm that God is the creator of a universe of some sort. Thus, all four are attempting to affirm the creator/creature distinction.

Classical theism is unique in affirming four divine attributes: timelessness, immutability, impassibility, and simplicity. It also affirms that the modal scope of God’s knowledge extends to the future. Christian classical theists affirm a doctrine of creation ex nihilo.

Neoclassical theism is a model of God that rejects one or more of the four unique classical attributes of God. It is distinct from open theism because it affirms that God has an exhaustive knowledge of the future. Like with classical theism, Christian neoclassical theists affirm a doctrine of creation ex nihilo.1

Open theism rejects all four of the classical attributes, denies that God has an exhaustive knowledge of the future, and affirms a doctrine of creation ex nihilo (Cf. Rice 2020).

Panentheism is notoriously difficult to define (cf. Mullins 2016). It claims that the universe is in God, but God is more than the universe. What that means in non-metaphorical terms is up for grabs (cf. Clayton 2019). Some panentheists affirm the four classical attributes, whilst others deny it. Some panentheists affirm a doctrine of eternal creation, whilst others embrace emanation. Some affirm that God knows the future, and others deny that God has an exhaustive knowledge of the future. What does seem to be clear is that panentheists deny the doctrine of creation ex nihilo (Gocke 2013; Stenmark 2019). For the purposes of this paper, I shall focus on Thomas Jay Oord’s version of panentheism that denies the four classical attributes and denies that God has an exhaustive knowledge of the future. Further, this panentheistic model affirms a doctrine of eternal creation. This version says that the divine nature of perfect goodness or perfect love necessitates that God must bring about the existence of a universe of some sort, but that God is free to create whichever universe He wants (Oord 2015). On this view, the nature of God is not somehow completed or made perfect by the existence of a universe. Rather, the universe exists because God is perfect goodness (Brown 2016, p. 14). I focus on this version of panentheism because it is subject to one of the creator/creature arguments that I shall consider in a later section.

In what follows, I wish to examine some recent attempts by classical theists to argue that rival models of God entail a violation of the creator/creature distinction.

3. Dolezal on the Creator-Creature Distinction

James E. Dolezal claims that non-classical models of God violate the creator/creature distinction. Dolezal lumps all non-classical models of God under the moniker of theistic mutualism. According to Dolezal, theistic mutualism is the view that God and creatures are in a symbiotic relationship of giving and receiving. Both God and creatures are capable of being ontologically moved by, acted upon, and determined by each other (Dolezal 2017, pp. 1–2). In other words, theistic mutualists deny the doctrine of impassibility, which says that God cannot be moved, acted upon, or determined by something external to Himself.2 Dolezal also asserts that theistic mutualists affirm a univocal way of being such that God and the universe exist within the same order of being, and interact with one another (Dolezal 2017, p. 2).

At this point, one might worry about the usefulness of Dolezal’s category of theistic mutualism. Dolezal places many distinct thinkers under this banner who affirm radically different conceptions of God and the God–world relationship. For example, William Lane Craig, John Frame, Charles Hartshorne, and Karl Barth are all said to be in the same category. To say that these thinkers have different models of God and different underlying metaphysical and theological systems would be an understatement.

One also might have worries about including a thesis on religious language in a model of God. After all, the Thomistic doctrine of analogy was not affirmed by classical Christian theists who lived before Aquinas, and it was rejected by subsequent classical theists like John Duns Scotus. Further, it is far from accurate that all of the non-classical models of God assume the doctrine of univocity. For example, various open theists and panentheists like William Hasker, Richard Rice, John Sanders, Curtis Holtzen, and Arthur Peacocke affirm the doctrine of analogy (cf. Hasker 2022; Rice 2020; Sanders 2007; Holtzen 2019; Peacocke 2007). Neoclassical theists like John Peckham also affirm the doctrine of analogy (Peckham 2021, pp. 33–37). I suggest that it is best to not include a philosophical thesis on religious language in a model of God since a model of God can be consistent with different stances on religious language. Instead, one should take a thesis on religious language to be part of a particular philosophical or theological system.

Despite these worries, it is important to note that Dolezal does acknowledge that these thinkers disagree over different issues. Yet, he says that they share a common denominator in that they all affirm that God is in some sense capable of being changed by creatures (Dolezal 2017, p. 4). Perhaps it is best to see ‘theistic mutualism’ as a pedagogical device employed by Dolezal to make his material easier to grasp by different audiences. After all, it is rather cumbersome to list out all the different models of God on offer today. Careful readers will notice that I have limited myself to four different models, and completely ignored pantheism. So, anyone worried about the usefulness of a category like ‘theistic mutualism’ should keep these pedagogical considerations in mind.

For the purposes of precision, I shall focus on a version of neoclassical theism that I take to fit one of Dolezal’s primary targets. Dolezal emphasizes that contemporary Calvinist theologians within the evangelical world have departed from the classical model of God (Dolezal 2017, p. 21). The neoclassical model of God that I shall have in view here rejects divine timelessness, immutability, impassibility, and simplicity. In good Calvinist fashion, this view affirms that God has an exhaustive knowledge of the future because He has causally determined what shall come to pass.3

Recall from above what neoclassical theism affirms. It affirms that God is a necessarily existent being who has the properties of aseity, self-sufficiency, maximal power, maximal knowledge, maximal goodness, perfect freedom, and eternality. These are essential properties of God, and as such it is metaphysically impossible for God to lose these properties. No thing can lose its essential properties. As temporally eternal, God never began to exist, but God can undergo succession as He freely exercises His power to create, sustain, and providentially guide the universe. As mutable, God can gain and lose accidental properties over time. For example, God comes to have accidental properties associated with making covenantal promises such as being faithful to the Abrahamic covenant. This is not a property that God had prior to making a covenant with Abraham. As passible, God can have a range of emotions that are consistent with being perfectly rational and perfectly good. When God witnesses something truly tragic, God is sad. When God witnesses something truly sinful, God has wrath. When God witnesses a sinner repent, God rejoices. These are new, temporary emotional states that God comes to have when it is morally and rationally fitting to have them.4

Dolezal makes several different attempts to argue that this model of God violates the creator/creature distinction. Here is one argument. A neoclassical theist will say that creator is an accidental property that God acquires as the result of freely exercising His power to cause a universe to exist. According to the neoclassical theist, God was not always exercising this power, and God did not have to exercise this power. Hence, God is only accidentally the creator of the universe. This fits nicely with the divine foundationalism and the definition of creator described above. However, Dolezal thinks that something is amiss here. He writes,

If being Creator should be something temporal that God becomes, it would seem to follow that His actions in and toward the world as Creator are not properly the actions of God as divine. A creatorhood that begins to be cannot be regarded as an aspect of God’s divinity as such, but, ironically, must be considered as a creaturely property.

(Dolezal 2017, p. 97)

One might find this first sentence a bit confusing at first glance. It is difficult to figure out in what sense this temporal God’s actions are not properly the actions of God as divine. This is because the neoclassical theist says that God becomes the creator by exercising His essential divine properties of maximal power and perfect freedom. The fact that a divine being who freely exercises His divine power to create a universe would strongly suggest that His actions toward the universe are divine.

This second sentence from Dolezal is also somewhat confusing. It is difficult to figure out in what sense the property creator is a creaturely property. On my definition of what it means to be a creature, I limited my focus to contingently existent substances. This is because most models of God are held within metaphysical systems which affirm that abstract objects, like properties, are things that exist in the mind of God, or necessarily emanate from the divine nature, or necessarily exist independent of God (cf. Gould 2014). One finds this among classical and non-classical theists, so it is not controversial to deny that God creates abstract objects.5 Hardly anyone thinks that God creates abstract objects, though some Cartesians might be said to affirm this view. Perhaps Dolezal thinks that neoclassical theists are committed to the notion that God creates abstract objects, but I find this implausible because there are many neoclassical theists who would outright reject the view. Certainly, none of the thinkers that Dolezal has in mind would affirm this view. Most would affirm the divine foundationalism that my definition of creator presupposes. So, it is best to say that this is not what Dolezal has in mind when he refers to creator as a creaturely property.

What else could Dolezal mean? Another interpretation is that creator is a creaturely property in the sense that it is a property that only creatures can have. However, this is incoherent since the property of creator can only be had by a being that freely causes the universe to exist. Further, as explained above, the definition of creator presupposes divine foundationalism. So, it is impossible for a creature to have the property of creator. Hence, I think it is best to say that this is not what Dolezal has in mind when he says that creator is a creaturely property.

I must confess that I am uncertain what Dolezal could mean, and thus, I am unable to identify exactly what Dolezal’s argument is. The property creator is not the sort of thing that could be created, nor is it the sort of thing that a creature could possess. Hence, I cannot see what this version of the creator/creature argument could be. This being the case, I will turn to another example from Dolezal of the creator/creature argument.

In a separate argument, Dolezal says that,

Since God is the first being from whom all other being flows, it follows that He must not derive His own being from constituent parts or elements within Himself. All things that exist in the world are said to exist by His will…If God should be composed of parts, then these parts would be before Him in being, even if not in time, and He would be rightly conceived of as existing from them or of them. His existence would, in some respect, be bestowed to Him as a gift He receives from another.

(Dolezal 2017, p. 49)

According to Dolezal, this would violate God’s self-sufficiency, and would violate the notion that God is the source of all (Dolezal 2017, p. 49).

In order to understand this particular argument from Dolezal, one must know the Thomistic metaphysics that Dolezal is assuming. First, Dolezal is assuming that a ‘part is anything in a subject that is less than the whole and without which the subject would be really different than it is’ (Dolezal 2017, p. 40). Second, Dolezal is assuming that ‘the parts in an integrated whole require a composer distinct from themselves to unify them, an extrinsic source of unity’ (Dolezal 2017, p. 40). With this in mind, I can reformulate Dolezal’s argument.

(1)

If God acquires the property of Creator, then God comes to have a part.

(2)

God acquires the property of Creator.

(3)

Thus, God comes to have a part.

(4)

If God is composed of parts, then these parts are before God in being, even if not in time.

(5)

Thus, the part Creator is before God in being.

(6)

If some thing has a part, then that thing and its part are composed by an extrinsic source of unity.

(7)

Thus, God and the part Creator are composed by an extrinsic source of unity.

(8)

No thing that has an extrinsic source can be the creator of all.

(9)

Thus, God cannot be the creator of all.

I am not certain that this is exactly what Dolezal is arguing for, but this is the best that I can piece together from his text. I gather the idea is that a neoclassical theist would have to accept premise (2) since she affirms that the property Creator is an accidental property that God acquires by freely exercising His essential divine power and freedom. Hence, the rest of the argument should follow straightforwardly.

However, the neoclassical theist can safely reject premise (1) because she is under no obligation to accept Dolezal’s incredibly permissive definition of parts. It is far from obvious that abstract objects like properties can literally be considered as parts since many metaphysicians speak of the simplicity of the soul and consider that a soul has properties. For example, the classical theist Augustine takes the soul to be simple in that it lacks parts, but he affirms that the soul has properties, powers, and capacities (Goetz and Taliaferro 2011, pp. 37–39). Further, there are Thomistic metaphysicians who claim that form and matter should not literally be considered as parts (Niederbacher 2015, p. 114). Moreover, most mereological discussions about the part–whole relation focus on concrete, material objects, and not on immaterial objects be they concrete or abstract. Without some reason from Dolezal for accepting his permissive understanding of parts, the neoclassical theist is free to adopt whatever mereological view she wishes and can safely deny premise (1). In other words, the neoclassical theist sees no good reason why she should think that the abstract property creator is literally a part.

The neoclassical theist can also reject premises (4) and thus the inference to (5). (4) is not some obvious philosophical principle, nor has it historically been considered to be obvious. Within Western philosophical theology, premise (4) was hotly contested in the Middle Ages, as was the doctrine of divine simplicity (Harvey 2021, chapter 4). There is no good reason to think that the accidental property creator is somehow prior to God’s essence or existence. Again, the neoclassical theist says that God comes to have this accidental property by freely exercising His essential power. God’s existence and essence are quite obviously prior, or more fundamental, than the accidental property creator. As noted above, most classical and non-classical theists affirm that abstracta are either divine ideas, or emanate from God, or necessarily exist independent of God.

The neoclassical theist can also reject premise (6). There is no good reason to think that all instances of composition require a source that is external to a thing and its parts. Consider the case of the incarnation. By far, the most popular models of the incarnation within classical Christian thought are compositional models wherein Jesus Christ is composed of God the Son, a human soul, and a human body (cf. Marmodoro and Hill 2010). When God comes to be incarnate, does He require some external source to unite Him to this particular soul and body? That seems implausible. Surely a maximally powerful being can do the uniting Himself.

As it stands, the neoclassical theist can reject several of the premises in this argument. I suggest that what Dolezal has offered is not a legitimate form of the creator/creature violation objection that I have been looking for.

4. Kevin J. Vanhoozer on the Creator/Creature Distinction

Kevin J. Vanhoozer offers a critique of Thomas Jay Oord’s panentheistic model of God. Oord refers to his panentheistic model as a kind of essential kenosis. One of the primary objections that Vanhoozer offers is that Oord’s model of God blurs the creator/creature distinction by placing God on the same causal level as creatures (Vanhoozer 2020, pp. 17, 21–23). I find it difficult to identify what Vanhoozer’s argument is, but Vanhoozer takes the crux of the issue to be whether or not God is in a real relation to the universe. The results of this would be significant since most models of God affirm that God is in a real relation to the universe. In fact, there are even some classical theists who affirm that God is really related to the universe because they find it unintuitive to deny that God is really related to the universe (e.g., Rogers 2009). The claim that God is not really related to the universe is primarily found within classical theism. Hence, if Vanhoozer’s argument is successful, it would seem to cast doubt on everything but classical theism.

Following John Webster and Thomas Aquinas, Vanhoozer says that God is not in a real relation to the created universe, but is instead in a relation of reason (Vanhoozer 2020, p. 21). Creatures are said to be really related to God, but God is not really related to creatures. This is called a mixed relation because the relation is only real on one end, and not real on the other. Before articulating the actual concept of real relations, relations of reason, and mixed relations, I will start with Vanhoozer’s articulation.

As Vanhoozer understands it, a real relation runs in both directions, thus making both terms of the relations co-dependent. Since creatures depend upon God, they are really related to God. Since God is not dependent upon creatures, God is not really related to creatures. Instead, God stands in a relation of reason to creatures. Vanhoozer says that this relation of reason is not fictional nor insignificant. Instead, relations of reason are ontologically non-constitutive and causally asymmetrical (Vanhoozer 2020, p. 22). The essence of God is not constituted by His relationship with creation, thus Vanhoozer says that God is not really related to the universe. Because God is perfect in Himself, and does not need the created universe, God cannot be really related to the universe (Vanhoozer 2020, p. 23). Following Webster, Vanhoozer says that the denial of real relations gives a proper specification to the true nature of God’s relation to the world, and His love for the world. Of course, readers might find it confusing to hear one saying that the true nature of God’s relation to the world is that God is not related to the world, but I digress. Vanhoozer writes,

The goal is a properly theological understanding of the God-world relation rather than an extension of a univocal concept of relation learned from observing the created order. God is not one term in a dyad, in a zero-sum metaphysical arena, which is what ‘mutuality’ and ‘reciprocity’ suggest. Webster retrieves Thomas’s distinction, denying God’s ‘real relation’ to the world, in order to better establish the true nature of God’s loving relation to the world. It is precisely because God lacks nothing and needs nothing from the world that he can communicate his own good fullness to it.

(Vanhoozer 2020, p. 23)

As far as I can tell, Vanhoozer takes the following things to be wrapped up in the concept of a real relation.

(RR1) A real relation constitutes the essence of a being that stands in a real relation.

(RR2) A real relation implies the co-dependency of the beings in the relation.

(RR3) A real relation puts the beings in the relation on the same causal level, in a zero-sum metaphysical arena.

(RR4) A real relation is a univocal concept that is learned from observing the created order.

Given this, one might see why Vanhoozer will want to deny that God is really related to the universe. Contra (RR1), God is not essentially related to the universe. God can exist without the created order given the doctrine of creation ex nihilo that Vanhoozer affirms. Contra (RR2), God and creatures are not co-dependent on each other. All created things asymmetrically depend upon God given divine foundationalism. Contra (RR3), God is not on the same causal level as creatures because, according to Vanhoozer, God causes all creaturely movements (Vanhoozer 2020, p. 24; cf. Webster 2009, pp. 164, 167; Helseth 2011, p. 31). No creatures cause any divine movements. Contra (RR4), Vanhoozer wants a properly theologically theological concept of relation for God that I gather must be analogical, and not learned from observing the created order.

As far as I can tell, Vanhoozer seems to think the following are involved in saying that God stands in a relation of reason to creatures.

(NR1) A relation of reason does not constitute the essence of a being that stands in a relation of reason.

(NR2) A relation of reason does not imply a co-dependency of the beings that stand in the relation.

(NR3) A relation of reason puts the beings on different, asymmetrical causal levels.

(NR4) A relation of reason is not a univocal concept learned from observing the created order.

(NR5) God’s perfection entails that God is in a relation of reason to creatures.

Now that we have Vanhoozer’s understanding of real relations and relations of reason on the table, I can offer what I guess would be a creator/creature argument. Vanhoozer primarily offers assertions, so I can only guess what the argument is meant to be. One argument can go like this. Call this the Essential Dependence Argument.

(ED1) If God is really related to creatures, then God is dependent upon creatures for His essence.

(ED2) God is not dependent upon creatures for His essence.

(ED3) Thus, God is not really related to creatures.

Here is another kind of argument that Vanhoozer can run. Call this the Causal Level Argument.

(CL1) If God is really related to creatures, then God is on the same causal level as creatures.

(CL2) If God is on the same causal level as creatures, then God is a creature.

(CL3) God is not on the same causal level as creatures.

(CL4) Thus, God is not really related to creatures.

How might someone like Oord respond to Vanhoozer’s arguments? To start, one can point out that Vanhoozer has mischaracterized the concept of mixed relations by building far too much into the concept. The actual concepts of real relation and relation of reason contain significantly less theological baggage than Vanhoozer lets on. Return to the concept of a real relation.

Vanhoozer’s (RR1) says that a real relation constitutes the essence of a being that stands in a real relation. This is a demonstrably false understanding of the classical theistic notion of a real relation. When classical Christian theists from the Middle Ages developed the claim that God is not really related to the universe, they relied on the Aristotelian understanding of relations. Relations are one of Aristotle’s nine categories of accidents (Kerr 2018, p. 82). The medieval doctrine of real relations says that a relation is real when both terms in the relation have a relevant relational property. A common medieval example offered by Augustine is a master and slave relation (Augustine 1991, V.17). The master has the property master of the slave and the slave as the property slave to the master. These properties signify the relation by pointing to the other item in the relation. The relation is real because each object in the relation has a relevant relational property. Does this relation constitute the essence of the beings in the relation? Quite obviously not. The medieval doctrine of real relations states that real relations can be essential or accidental to the objects in the relation. For example, the medieval theologians would say that the real relation between the divine persons is essential, whereas the real relation between a master and a slave is accidental. The master might not have always been a master, and the slave might not have always been a slave. Once they enter into this real relation, they acquire the accidental properties of master of the slave and slave to the master. Another common example from the Christian tradition is the accidental properties of creator, lord, and redeemer. These were taken to be accidental properties that God would have if God were in a real relation with creation. Yet, thinkers like Augustine, Peter Lombard, and Aquinas say that God is not really related to the universe in order to avoid saying that God has these accidental properties (Augustine 1991, V; Lombard 2007, XXX.1; Aquinas 1934, II.12). They deny that God is really related to the universe directly in response to objections against timelessness, immutability, and simplicity. They are not engaged in some theological theology, or anything of the sort. Instead, they are making an ad hoc move in light of serious objections to classical theism.

Return to (RR2) which says that a real relation implies the co-dependency of the beings in the relation. This is true in a very minimal sense that has virtually no theological significance. If two beings are really related, they are dependent on each other in order to be really related. Consider the master and slave relation again. The two people in this relationship are dependent upon each other, among other factors, in order to have the relevant relational properties. If the master loses his slave, he will no longer have the property master of the slave. If the slave becomes free, he will no longer have the property slave to the master. Nothing about this real relation implies any deeper kind of co-dependency. For example, I am really related to the chair that I am currently sitting on. I can most certainly exist without the chair, and the chair can exist without me. Nothing about my existence or my essence depends upon this chair. The same is true of God if God is really related to creatures.

Consider (RR3) which says that a real relation puts the beings in the relation on the same causal level, in a zero-sum metaphysical arena. The concept of real relation has absolutely nothing to do with this, nor does the concept of a relation of reason have anything to do with being in a different causal level, as I shall explain later (Ward 2010, p. 292). Instead, I believe that Vanhoozer and others are importing theological concepts into the discussion that the classical tradition did not. As far as I can tell, the claim that God is on a different causal level simply means that (i) God is the cause of all things including creaturely actions in such a way that creatures still have free will, and (ii) creatures do not have any causal influence on God.6 This has absolutely nothing to do with real relations. In fact, according to the medieval scholar Thomas M. Ward, this is consistent with God being really related to the universe (Ward 2010, pp. 293–301). Condition (i) is simply theological determinism plus compatibilism about human freedom. Condition (ii) is the doctrine of divine impassibility. One can affirm both of these and say that God is really related to creatures. How? One can do this in two steps. First step: God has the accidental property of creator and creatures have the property of creatures. Second step: affirm theological determinism, compatibilism, and impassibility. The claim that God is not really related to the universe is doing absolutely no work here. (RR3) is question begging because it is smuggling in the model of God that it wishes to justify on the basis of (RR3).

Finally, consider (RR4) which says that a real relation is a univocal concept that is learned from observing the created order. I suppose that I can grant that this is true. However, this leads to my next point. Vanhoozer has misunderstood the concept of a relation of reason. Recall (NR4) which says that a relation of reason is not a univocal concept learned from observing the created order. This is demonstrably false. According to Ward, ‘Aquinas’s discussion of relations in these theological contexts are clearly not intended to be applicable to theological cases only’ (Ward 2010, pp. 279–80). The same is true of other major thinkers within the classical tradition. When Augustine and other medieval theologians appeal to a relation of reason to describe God’s relation to the universe, they were appealing to a widely accepted Aristotelian concept that they took to be applicable to ordinary, everyday objects (King 2003, p. 36; Dodds 2008, pp. 165–69; cf. McWhorter 2013). In other words, it is a univocal concept learned from observing the created order. The popular example that the medieval theologians appeal to is someone thinking about Socrates (Brower 2001). The medieval theologians are borrowing this example from non-theological contexts and putting it to use in a theological context. The claim is that when Augustine is thinking about Socrates, Augustine is in a real relation to Socrates. This is because Augustine has the relevant accidental relational property thinking about Socrates. The claim is also that Socrates is not in a real relation to Augustine because Socrates does not have any accidental relational property like being thought about by Augustine. Why? The medieval claim is that this is merely a relation of reason. A relation of reason is said to have no extramental existence. In other words, it has no existence outside of Augustine’s head. Further, the relation between Socrates and Augustine is said to be a mixed relation because the relation is real for Augustine but not real for Socrates. Medieval theologians took this everyday example that they learned from the observable world, and then univocally applied it to God to describe God’s relationship to the world. To further drive this point home, it should be noted that Augustine, Peter Lombard, and others employ these notions long before Aquinas’s doctrine of analogy comes into existence. So, the notions of real relations and relations of reason have no obvious conceptual linkage to the doctrine of analogy. Vanhoozer is simply mistaken about the concept of a relation of reason on multiple fronts.

This leads to the next mistake in Vanhoozer’s account. Recall that (NR3) says that a relation of reason puts the beings on different, asymmetrical causal levels. This is demonstrably false. Return to the example of Augustine thinking about Socrates. Socrates is in a relation of reason to Augustine. Socrates and Augustine are not on different, asymmetrical causal levels, whatever that means. Nothing about a relation of reason implies anything about causation.

Vanhoozer can affirm (NR1) and (NR2). These say that a relation of reason does not constitute the essence of a being that stands in a relation of reason, and do not imply a co-dependency of the beings that stand in the relation. However, as should be clear from the previous discussion, nothing about this is unique to relations of reason. A real relation does not necessarily constitute the essence of a being, nor does it imply any sort of co-dependency in terms of essence or existence.

What about (NR5)? This says that God’s perfection entails that God is in a relation of reason to creatures. I say that perfection has absolutely nothing to do with whether or not a being stands in a relation of reason to other beings. Socrates stands in a relation of reason to Augustine. It seems safe to say that Socrates is far from perfect. Yet, nothing about Socrates being perfect or imperfect has anything to do with him standing in a relation of reason to Augustine. The only thing that determines if this is a relation of reason is whether or not Socrates has an accidental relational property from the relationship. Medieval philosophers said that Socrates does not have such a property. Their intuition about this has nothing to do with the perfection or imperfection of Socrates. It is entirely about the relation itself. Vanhoozer is unjustifiably importing theological baggage into the concept of a relation of reason.

With the notions of relations cleared up, I can return to Vanhoozer’s creator/creature arguments. I will start with the Essential Dependence Argument. Recall that the argument says this.

(ED1) If God is really related to creatures, then God is dependent upon creatures for His essence.

(ED2) God is not dependent upon creatures for His essence.

(ED3) Thus, God is not really related to creatures.

Anyone who wishes to affirm that God is really related to creation can safely deny (ED1). Nothing about the concept of a real relation implies anything about dependence of essence. Real relations can be essential or accidental. Most Christian theologians will say that God’s real relation to creation is accidental because God is free to create or refrain from creating the universe. Hence, this argument is a non-starter.

Recall the Causal Level Argument. It goes as follows.

(CL1) If God is really related to creatures, then God is on the same causal level as creatures.

(CL2) If God is on the same causal level as creatures, then God is a creature.

(CL3) God is not on the same causal level as creatures.

(CL4) Thus, God is not really related to creatures.

I believe that there are at least three options for anyone who wishes to avoid this argument. First, one can say that the notion of different causal levels is unintelligible, and thus deny (CL1). She can even extend this to denying (CL2) by saying that the notion of different causal levels is unintelligible, and thus, cannot possibly have anything to do with God being a creature. However, as I indicated above, the notion of different causal levels seems to be nothing but the affirmation of impassibility, theological determinism, and compatibilism about human freedom. Perhaps, then, the concept is somewhat intelligible, but this does not help the Causal Level argument.

This leads to a second way to avoid the argument. One can reject (CL1) by saying that the notion of different causal levels is completely irrelevant to whether or not God is really related to creatures. To say that ‘God is on a different causal level’ is to say that one affirms theological determinism, impassibility, and compatibilism about human freedom. One can easily say that God is really related to creatures and affirm determinism, impassibility, and compatibilism. All one has to do is say that God has the accidental relational property of creator, and she will thus be affirming that God is really related to the created universe.

There is at least one more option for avoiding the Causal Level argument. One can deny (CL2) by saying that being on the same causal level is completely irrelevant to whether or not a being is a creature. In this strategy, one grants that God is on the same causal level of creatures. Say that a model of God affirms that God necessarily exists, is a se, self-sufficient, omnipotent, omniscient, perfectly good, and perfectly free. Further, say that the model affirms divine foundationalism. Moreover, that God is really related to creatures because He has the accidental property of creator. Given divine foundationalism, God is the ultimate causal source of all contingent reality. How does any of this imply that the necessarily existent, a se, self-sufficient cause of all of contingent reality is a creature? As far as I can tell, there is no plausible reason to think that God is a creature merely because He stands in a real relation to the universe that He freely created.

These reflections lead me to conclude that Vanhoozer has not identified a legitimate creator/creature argument.

5. The No Real Relations Doctrine Is Inconsistent with God Being the Creator

In this paper, I have surveyed a variety of attempts to develop a creator/creature argument, and I have found each of them wanting. This does not lead to the conclusion that there are no legitimate creator/creature arguments. Instead, my discussion should lead theologians to reflect on being more cautious with the accusation that a rival model of God violates the creator/creature distinction. As it stands, the accusation of blurring the creator/creature distinction is quite common in contemporary theology, but I suggest that it should become less common until legitimate forms of the argument can be identified.

Before concluding, however, I wish to make one further observation. I find it rather odd that Vanhoozer, Dolezal, and others make appeals to the no real relations doctrine when trying to develop creator/creature arguments. As I see it, the entire notion that God is not really related to the universe is inconsistent with the classical account of real relations, and inconsistent with the notion that God is the perfect knower and creator of the universe. Here is why. Within the classical tradition, standard examples of real relations are things like the relation between cause and effect, and the relation between the knower and what is known.

I shall start with the relation between the knower and what is known and argue that the no real relations doctrine undermines the notion that God is perfect in knowledge. Recall the earlier example of Augustine knowing Socrates. The knower, Augustine, is really related to Socrates, whereas Socrates is said to be in a relation of reason to Augustine (Ward 2010, p. 290). The knower to known relation is a real relation. With this in mind, consider the Knower Argument.

(KA1) The knower to known relation is a real relation.

(KA2) If God knows the universe, then God is really related to the universe.

(KA3) God knows the universe.

(KA4) Thus, God is really related to the universe.

(KA5) God is not really related to the universe.

(KA6) Thus, God does not know the universe.

(KA7) Thus, God knows the universe and God does not know the universe.

This is a contradiction, so the classical theist will need to reject one of the premises. (KA1) is the standard claim within the classical tradition about real relations. As discussed earlier, it is explicitly endorsed by classical thinkers like Augustine and Aquinas. (KA2) simply follows from (KA1). (KA3) follows from God’s omniscience, so no classical theist can deny it. This gets us to the conclusion in (KA4). Thus far, we have no good candidate premises for the classical theist to reject. (KA5) is explicitly endorsed by classical theists like Augustine, Peter Lombard, and Aquinas. Yet, from (KA2) and (KA5), we get the conclusion in (KA6). From (KA3) and (KA6), we get the contradiction in (KA7). This is not good news for the classical theist who wishes to say that God is not really related to the universe. This is good news for any model of God that affirms that God is really related to the universe since she can easily reject (KA5) and avoid the contradiction.

Next, consider the relationship between cause and effect. The causal relation is a classic example of a real relation, and not a relation of reason (Ward 2010, p. 297). The cause is really related to the effect. Saying that God is not really related to the universe undermines the claim that God is the creator of the universe. Call this the Causal Argument.

(CA1) Causes are really related to their effects.

(CA2) If God causes the universe to exist, then God is really related to the universe.

(CA3) God causes the universe to exist.

(CA4) Thus, God is really related to the universe.

(CA5) God is not really related to the universe.

(CA6) Thus, God does not cause the universe to exist.

(CA7) Thus, God causes the universe to exist and God does not cause the universe to exist.

As with the previous argument, (CA1) is a standard classical claim explicitly endorsed by thinkers like Aquinas (Ward 2010, p. 297). Moreover, it is immanently plausible (O’Connor 2013, p. 31). (CA2) simply follows from (CA1) (Ward 2010, p. 299). (CA3) seems obviously true since that is assumed in the classical doctrine of creation ex nihilo, and classical cosmological arguments for the existence of God. So, we are left with the conclusion in (CA4). (CA5) is explicitly endorsed by Augustine, Peter Lombard, and Aquinas. From (CA2) and (CA5), we get the conclusion in (CA6), which leads to the contradiction in (CA7). What is the classical theist to do? Surprisingly, the only option available seems to be to deny (CA3)!

Part of what it means to say that God is not really related to the universe is that God does not have the property being the creator of the universe. As Aquinas makes clear, we only predicate creator of God according to our way of understanding. The predicate creator of the universe does not have any extramental existence when we predicate it of God (Aquinas 1934, II.13–14). As proponents and critics of the no real relation doctrine agree, God cannot even stand in the real relation causing to exist with the universe (Kerr 2018, pp. 82–85; Craig 2001, pp. 61–78). I find it perplexing how such a God could possibly be the creator of the universe since the very definition of being the creator includes causing the universe to exist. Hence, why I say that the no real relations doctrine undermines the very notion of God as the creator.

These two arguments lead me to suggest that the no real relations doctrine undermines the notion that God is a perfect knower, and the notion that God is the creator of the universe. Hence, the doctrine of no real relations cannot be part of the creator/creature distinction.

6. Conclusions

To conclude, I find the following state of the dialect rather baffling. On the one hand, we have non-classical theists saying that God does in fact have the property creator of the universe. On the other hand, we have classical theists explicitly saying that God does not have the property creator of the universe. These very same classical theists who deny that God has the property creator, accuse non-classical theists of blurring the creator/creature distinction precisely because these non-classical theists affirm that God literally has the property creator. That seems very odd to say the least. One would think that having the property creator would be a necessary and sufficient condition for being the creator of the universe. Yet, if the classical theist is correct, having the property creator is a necessary and sufficient condition for being a creature. Surely that cannot be the right result, and I suggest that classical theists rethink things before accusing others of blurring the creator/creature distinction.