Necessary existent theology

Abstract

A meta-theology makes claims about the structure of theological claims: it identifies a single, fundamental claim about God, and shows how other theological claims are derivable from the fundamental claim. In his book Depicting Deity and other articles, Jon Kvanvig has identified three distinct meta-theologies: Creator Theology, Perfect Being Theology, and Worship-worthiness Theology. In this article, we argue that the medieval Islamic philosopher Avicenna's views about God have the structure of a meta-theology, and that it is distinct from the three projects Kvanvig identifies. This view is Necessary Existent Theology.

Preliminaries

In Depicting Deity, Jon Kvanvig outlines the project of ‘meta-theology’. A meta-theology, Kvanvig says, is a theological project which begins with a single fundamental claim about God, and derives other standard theological claims from the fundamental claim (Kvanvig (2021)). For example: a proponent of ‘Perfect Being Theology’ begins with the fundamental claim that God is the most perfect being, and then derives from this fundamental claim the further claim that God is all-powerful, on the grounds that anything that is less than all-powerful would fail to be the most perfect being. In general, while an ordinary theology articulates the claims that are true of God, a meta-theological project structures the theological truths, by identifying one claim as fundamental and others as non-fundamental, or derived, truths.

Kvanvig is well aware that endorsing a meta-theological project is not obligatory, since one could hold that there are theological truths, but no single, fundamental truth (Kvanvig (2021), 4). He does claim, however, that if one is to endorse a meta-theology, the there are three options that are ‘promising’:

The task I set is thus metatheological: an investigation into the kinds of approaches one might take in developing a theology, an investigation into the proper starting point for characterizing the nature of God . . . Among the most promising starting points for such a project are conceptions of God which begin from one of three initial assumptions. In brief, I label these three approaches ‘Creator Theology’, ‘Perfect Being Theology’, and ‘Worship-Worthiness Theology’. (Kvanvig (2021), 3)

It is clear, however, that Kvanvig is careful not to rule out the possibility of additional alternatives. In calling these three meta-theological approaches ‘the most promising starting points’ he does not insist that it is impossible to characterize alternative meta-theologies.¹

In this article, we will argue that there is indeed a fourth alternative, which is both a central player in Medieval Islamic philosophy, and has a serious philosophical pedigree. This meta-theological view is Necessary Existent Theology.² The primary claim of Necessary Existent Theology is that God is the Necessary Existent, and its most prominent proponent is Avicenna. In what follows, we will argue that Avicenna's view meets the conditions for a meta-theology that Kvanvig sets out and, moreover, that Necessary Existent Theology is distinct from the three meta-theological views that Kvanvig discusses. But first, a few clarifying remarks on the meta-theological project in general will be helpful.

What is a meta-theology?

Kvanvig makes several claims about what a meta-theology involves. It will be important to be clear on what these are, as in the next section we will argue that Avicenna has a view which counts as a meta-theology, in Kvanvig's sense. But there are some details which need to be cleared up, in order to make the concept of a meta-theology clearer, especially in light of some of the central claims in Avicenna's view of God.

A meta-theology identifies what is fundamental to the nature of God

Kvanvig puts this point as follows:

To assume one of these three standpoints (viz., Creator Theology, Perfect Being Theology, or Worship Worthiness Theology) is to assume that one of these three is fundamental to an adequate account of the nature of God, that moving from bare theism to a more substantive theology begins from one of these standpoints, with the additional claim that what is valuable in the other approaches can be derived from what is fundamental. (Kvanvig (2022), 139)

Let us assume that, in general, talk of what is metaphysically fundamental, and what is not, will be unproblematic. (Metaphysical notions related to fundamentality can be found in Lewis (1983), Fine (2001), Sider (2012), and elsewhere.) This may be a controversial assumption for some metaphysicians, but theological realists who are willing to hold that there are substantial truths about God are not likely to dispute it. Typically, theological realists will grant that it makes sense, for example, to ask whether God's commandment not to steal is more fundamental than our obligation not to steal. Fundamentality-scepticism is not, we will assume, a live possibility here.

However, even if talk of what is fundamental is in general well-disciplined and contentful, there are special issues that arise when applying it to claims about the nature of God. As is well known, Avicenna, along with a significant portion of the Islamic, Christian, and Jewish philosophical traditions, held that God is simple. The doctrine of simplicity is controversial in its details, but many of its proponents at the very least understand it to prohibit the existence of any parts or attributes in God. A meta-theological distinction between the fundamental and derived claims about God, however, risks a kind of division between the fundamental and non-fundamental in God. In other words, if perfection is fundamental to God's nature, as Perfect Being theologians hold, then it would seem to follow that God's eternity is not fundamental. Hence God's eternity is distinct from his perfection, because they differ in the following respect: perfection is fundamental, whereas eternity is not fundamental. This appears to conflict with the doctrine of simplicity.

This is a sketch of a general problem, which affects any meta-theological project which aims to be consistent with the doctrine of simplicity. But Necessary Existent Theology, if it is a legitimate candidate for a (distinct) meta-theology, faces an especially acute version of the problem. There is a straightforward argument that necessary existence entails simplicity. In later sections, after sketching the argument that Necessary Existence Theology entails simplicity, in addition to other claims about God, we revisit the tension between simplicity and the meta-theological.

Meta-theological claims are metaphysical, not epistemological

Kvanvig insists that we distinguish between the epistemology of theology and its metaphysics. In particular, how we come to know that God is timeless and omnipotent need not begin with God's fundamental nature, and involve a derivation of timelessness and omnipotence. As far as our knowledge is concerned, we might come to know the fundamental fact about God by first knowing some metaphysically non-fundamental claims, and then inferring the metaphysically fundamental.

For a particular meta-theology, the important question is whether the central theological claim can be shown to metaphysically explain other facts about God. In his discussion of Creator Theology (CT), which takes the central claim about God to be that God is the source of all that exists, Kvanvig says the following:

CT should not be characterized in terms of what can be supported by cosmological and teleological considerations. Instead, it should be characterized in metaphysical terms. The central feature involved is that of a creatorship or sourcehood, which I will characterize as follows: according to CT, God is the asymmetric source of all else. The idea is to start with the characterization of God that is central to CT and see what can be learned about God from that fundamental starting point. The common mistake noted above is to conflate the metaphysical project with an epistemological one. The latter project has us focus on what we can learn about a creator through cosmological and teleological considerations; our project, however, involves the question of what can be derived from the assumption that God is, fundamentally, the asymmetrical source of all else. (Kvanvig (2021), 10–11)³

Kvanvig has in mind critics who insist that the inference is not very strong: a somewhat powerful, or somewhat devious creator could have created the universe. But as an epistemological question concerning how we know that the universe was not created by a somewhat powerful, less-than-omnipotent being, the issue is not one Creator Theology, as a meta-theological position, has to address. This is because a meta-theologian is making metaphysical claims. What is required, if Creator Theology is true, is that God's being a creator is more fundamental than, and metaphysically explains, the facts that God is all-powerful and that God is omnipotent. God's being a creator does not need to serve the epistemological function of serving as a basis for knowledge that God is all-powerful.

The distinction between metaphysics and epistemology will turn out to be crucial in what follows. We will rely on this distinction in making the case that Necessary Existent Theology is a distinct meta-theology. This is because there is a proof of God's existence in Avicenna that God must exist as a cause of everything else. The proof begins with the premise that things exist, then goes on to show that, even if these things are merely contingent,⁴ it follows that there must be a necessary existent, namely, God. Thus Avicenna seems to hold that we come to know that God exists, by means of a proof, because of God's causal relationship to other existents. But this on its own is an epistemological claim, and does not show that Avicenna agrees with the central meta-theological claim of Creator Theology. That would require for Avicenna to hold that the central, fundamental claim is that God is the cause, or source, of all else that exists. As we will argue below, Avicenna does not hold this. While he agrees that it is true that God is the cause of all that exists, he does not think that this is the central, fundamental claim about God.

Avicenna's ontology and proof of God's existence

Here we turn to arguing that Avicenna provides us with a view that meets the conditions for counting as a meta-theology, subject to the above provisos. This view is Necessary Existent Theology. In this section, we outline the central notions and core claims of Avicenna's view. In doing so, we make no claim to originality, as the centrality of necessary existence to Avicenna's theology will be no surprise to readers of Avicenna.⁵ In the next section, we turn to the central thesis of this article, as we argue that these claims constitute a unique and distinctive meta-theology.⁶

Avicenna's modal ontology: necessary and possible existence

Avicenna's basic ontology divides everything into one of four categories, which involve two pairwise distinctions. The first is the distinction between necessary and possible existence. Avicenna says in Salvation VIII.2.1.1, ‘the necessary existent is the existent which, when posited as not existing, an absurdity results. The possibly existent is the one that, when posited as either existing or not existing, no absurdity results’ (Avicenna (2007), 211).

Avicenna also distinguishes between those things which have their modal status in themselves from those which have this status through another. In Salvation VIII.2.1.2, he says, ‘the necessarily existent through itself is that which is owing to itself not to any other thing . . . the necessarily existent not through itself is that which becomes necessarily existent if something other than it is set down’ (ibid.).

Conjoining these distinctions, the conceptual space for existents can be divided into four exhaustive categories: the necessary in itself, the necessary through another, the possible in itself, and the possible through another.⁷ What we would call ordinary contingent objects occupy, for Avicenna, two of these categories. Take Giles the cat: Giles is possible in itself since there is both no absurdity in positing that Giles exists, and no absurdity in positing that Giles does not exist. Giles is also necessary through another, on Avicenna's view: if we do posit the (perhaps very complex) causes of Giles then an absurdity does result if we posit the causes without Giles. In this sense, Giles is necessary through another, where the ‘other’ is the complex set of causes of Giles.

Note that, for Avicenna, the concept of necessity is not the modern concept of existence in all possible worlds. A necessary existent in itself cannot fail to exist in any possible world, because it does not depend on anything outside itself for its existence. But Giles is also necessary, in a sense, because an absurdity results from positing the existence of the causes through which Giles exists, without also positing the existence of Giles himself. It does not follow as a matter of conceptual necessity, however, that Giles exists in all possible worlds; it is conceivable that there are worlds where the causes of Giles don't exist, and so Giles doesn't either. This is the sense in which Giles is necessary, but through another.⁸

Avicenna identifies God with the necessary existent through itself. We will save discussion of Avicenna's proof for the necessary existent through itself for later. Here, we simply note that Avicenna thinks that there is a necessary existent through itself, and that this is God. The relationship between God and necessary existence through itself is very strict, as some interpret him as going so far as to hold that necessary existence through itself is not God's essence; strictly speaking, God has no essence. Rather, necessary existence through itself is God's very self (dhāt).⁹ The category of possible existence through another is empty: for Avicenna, things are possible (or not) in virtue of their own essences. What make Giles possible is that, as a cat, there is no absurdity in supposing that Giles exists, or in denying that Giles exists. Giles does not need another for his possible existence; others are needed, informally, to make his existence actual.

Although we use compound English expressions (or, in Avicenna's case, Arabic) to name these categories – necessary existence through itself, necessary existence through another, etc. – it is best not to think of these as metaphysically derived, or complex, categories on Avicenna's view. Instead, there are four simple, basic categories, which are not composites of distinct, more basic categories.¹⁰ These can be thought of as fundamental metaphysical categories; there is no sense, for Avicenna, in which the essence of Giles is composed of possibility and through anotherness, and likewise there is no sense in which God is composed of necessity and through itselfness. This will become evident as we discuss God's simplicity below. For now, we simply note that conceptual structure does not map on to metaphysical structure.

Avicenna's proof of the necessary existent

Avicenna does not rely on a simplistic assumption that an infinite regress of causes that are merely necessary through another is impossible in order to establish the necessary existent. Rather, he allows that each possible existent may be such that it is necessary through another – that is, it has a cause – and in each case the cause is another possible in itself. If this is the case, then there is an infinite series of existents that are made necessary (‘actual’) through another that is only possible in itself. Of course, Avicenna does not outright assert that the series of possible-in-themselves is infinite; rather, his argument in Salvation II.1.2 is independent of the truth on this matter, since the same questions about the necessity or possibility of the series can be asked independently: ‘As for their [i.e. the contingent, possible-in-themselves, existents] existing all together, and none is a necessarily existing being, then either the totality, insofar as it is that totality, whether finite or infinite, exists necessarily through itself or possibly in itself’ (Avicenna (2007), 215; emphasis added). The central claim in Avicenna's proof is that there are three possibilities concerning this (possibly infinite) totality of existents that are possible in themselves. One is that the totality is necessary in itself, which Avicenna says is absurd because ex hypothesi it is composed only of existents that are possible in themselves. The second is that the totality is possible in itself, and is made necessary through another cause which is internal to the totality. However, Avicenna says, the totality can only be caused by a cause which is responsible for the existence of each member of the totality. Thus, if a member of the totality causes the existence of the totality, then that member would be causing its own existence. This is a contradiction, since it entails that a member of the totality is necessary in itself, but ex hypothesi that totality consists only in things which are necessary through another. Finally, Avicenna says in Salvation II.1.2 that the cause could be external to the totality. In this case, the cause must be necessary through itself because the totality, we have assumed, includes all of the existents that are merely possible in themselves. So the only coherent option entails that something which is necessary through itself exists (Avicenna (2007), 215).

A few preliminary notes regarding the relationship between Avicenna's proof of a necessary existent, and its relation to the concept of a meta-theology, are in order.

The proof clearly begins with a premise about the existence of things.¹¹ Its conclusion is the claim that there is a necessary existent, which is established on the grounds that God, as the necessary existent in itself, is the cause of other existents. The relationship between God and what God causes, on Avicenna's view, is clearly central to epistemology: when we follow his proof, we come to know that God exists by first knowing that some things exist, and then inferring God as a cause. As Kvanvig has warned, however, we must distinguish between the epistemological and the metaphysical claims of a meta-theological project. Simply because God's status as first cause is central to Avicenna's proof, it does not follow that God should, metaphysically, be identified primarily as the cause of things that are possible in themselves. As we will argue below, Avicenna's metaphysics does not treat the claim that God is a cause as fundamental.

A second point is that the proof makes use of a notion of causation that is not what we would call material causation, which in a modern sense is the kind of causation that generally falls under the laws of physics. Avicenna is not arguing that any totality of material causes of things that are possible in themselves must have a necessary existent as their material cause, where the causal relationship is either governed by laws of physics or is related to physical causation in some analogous sense that is appropriate to God's causing creation to come into existence ex nihilo. In fact, he rejects the view that there must be such a first cause in the order of material causes, since he denies that the world was created ex nihilo. Rather, Avicenna's premises apply to the cause of an existent in the sense of that which explains, and sustains the fact that the existent has existence rather than non-existence. For Avicenna, the essence of something which is possible in itself cannot explain why it exists rather than not. A possible in itself can be coherently supposed not to exist. Adding a preceding material cause doesn't provide the needed explanation: it might explain why its effect comes into existence, but the fact that there was a material cause at time t does not explain why its effect exists at a later time, t*. Since the essence of the thing hasn't changed, there must be another existent now, at this very time, which explains why it exists rather than not.

As noted in (Zarepour (2022), 17, 21–22) Avicenna's notion of a cause is much more closely related to what a contemporary metaphysician would call a ground – see, for example, Fine (2001) and Schaffer (2009). This is why in Salvation II.1.3 Avicenna thinks the totality of possible-in-themselves cannot contain a circular explanation of the existence of each (emphasis added):

To set down a finite number of possible existents, each one of which is a cause of the others in a circle, is as absurd and obvious as the first problem. Particular to it, however, is that each one of them would be a cause and an effect of its own existence, where x comes into existence from y only after y itself comes into existence, but anything whose existence depends on the existence of what exists only after its own later existence cannot exist. (Avicenna (2007), 215)

Avicenna's notion of dependence is central to the absurdity of a circular chain of causes. When a grounds b, where both a and b are merely possible in themselves, a explains why b exists (is ‘necessary’ in Avicenna's terminology) and is not among the non-existent. What would be absurd is for a to provide this explanation for b's existence in conjunction with the fact that b is a part of the same kind of explanation of a's existence. Avicenna's notion of dependence, like the contemporary notion of ground, is asymmetrical. But a circular chain of grounding-relations among the possible-in-themselves would require a symmetrical grounding relation, and is absurd. b cannot be part of the grounds of a if a must already be supposed to exist as the ground of b.¹² Note that the plausibility of this argument decreases significantly if, instead of dependence, the causal relation at issue is material causation. It is much more plausible that a circular chain of material causes is possible since, while odd, it is not absurd to suppose backward causation is possible.¹³

The metaphysics of Necessary Existent Theology

A meta-theology identifies a single fundamental claim about God. Any particular meta-theological picture is compatible with the existence of additional truths about God, but these are not fundamental. Instead, the additional truths are derived claims, as they are claims which (allegedly) follow from the fundamental claim. Competing meta-theologies differ over which claim about God is fundamental, and the manner of derivation of the derived claims.

According to Necessary Existent Theology, the fundamental theological claim is:

God is the necessary existent

There are many derived claims according to Necessary Existent Theology; in what follows we will focus on the derivation of only a few derived claims within Necessary Existent Theology, which include:

God is simple

There is only one God

God is the first cause

God is perfect

Each derived claim can be shown to be true in one of three ways. First, a derived claim might be shown to be identical to the fundamental claim, in the sense that the derived claim states the same fact as the fundamental claim: that God is the necessary existent. Second, the derived claim could follow from the fundamental claim. One way to show this is to show that the negation of the derived claim is inconsistent with the fundamental claim. Finally, a derived claim might state a fact about God's relation, as ground, to other existents that are not necessary in themselves. In this case, the derived claim does not follow from the fundamental claim alone, but instead follows from the fundamental claim in conjunction with facts about other existents.

Below we will begin by showing how Necessary Existent Theology accommodates the derived claims in one of these three ways. Or, more precisely, we will show how Necessary Existent Theology purports to accommodate these claims. While we will sketch Avicenna's arguments that each of the derived claims follows from God's necessary existence, we do not mean to take a stand on the soundness of these arguments. Our claim is that Avicenna's treatment of God as the necessary existent has the structure of a distinct meta-theology.

God's simplicity

Avicenna says in Healing I.7 that God's simplicity follows from the fundamental claim that God is the necessary existent. In particular, God cannot be composed of parts, cannot be a genus or have a species, and cannot have accidents. In each case the proof is similar. If God is the necessary existent in itself, then a part, which is not numerically identical to God, must be something distinct from, and additional to, the necessary existent in itself. The part is then only possible in itself, and so must have a cause, which secures its existence. This contradicts the fundamental claim, that God is the necessary existent in itself. God, by having a part, a specific difference, or attribute, would then have a cause, and so would only be necessary through another (Avicenna (2005), 34–38).

Here it is clear that the simplicity of God is a consequence of the fundamental claim, namely that God is the necessary existent in itself. Avicenna has argued that the denial of God's simplicity is incompatible with the fundamental claim. So, Necessary Existent Theology includes the claim that God is simple, as a derived claim. In this case the derived claim follows from, but is not identical to, the fundamental claim of Necessary Existent Theology.

There is only one God

Avicenna argues in Healing I.6.11 that there is only one God, and also does so on the basis of the fundamental claim, that God is the necessary existent in itself. Here the simplicity of God, established above, serves as a lemma in the argument. Avicenna argues that the supposition that there are multiple necessary existents in themselves is incompatible with the simplicity of the necessary existent(s). The reason is that, if there were two necessary existents, then there must be something in virtue of which they are distinct. But then at least one of the necessary existents must have something that distinguishes it from the other necessary existent – if not, there would be nothing to make them distinct, and so there would not be two necessary existents. However, any distinguishing feature requires a cause, and hence the distinguished existent must be distinct from God, because God is simple. So there can be at most one necessary existent in itself (Avicenna (2005), 33).

The uniqueness of God is therefore a second derived claim that follows from the fundamental claim of Necessary Existent Theology. As with the derivation of the simplicity of God, it follows from the fundamental claim that God is the necessary existent.

God is the first cause

The priority of God as a first cause follows from the simplicity and unity of God, plus the status of every distinct existent as something that is merely possible in itself. Avicenna says in Healing VIII.4.1:

Since nothing other than Him is a necessary existent, He is the principle of the necessitation of the existence of everything, necessitating [each thing] either in a primary manner or through an intermediary. If the existence of everything other than Him derives from His existence, He is [the] first. (Avicenna (2005), 273)

It is noteworthy that Avicenna, in what immediately follows, is explicit that God's being the first cause is not part of the essence of God, but rather is a relation between God and other beings that are not necessary in themselves. God is the cause of, for instance, Giles the cat, and since there is no further cause of God, this relation between God and Giles is a part of God's being the first cause. (Avicenna allows that God, while being absolutely simple, can be related to other things – a point Maimonides would dispute later.¹⁴)

This means that the fact that God is the first cause is, for Avicenna, a distinct fact from the fact that God is the necessary existent. Its grounds include the fact that God is the necessary existent, but also include God's relation to other existents, as the reason why these possible-in-themselves are necessary through another. At the same time, God's status as the necessary existent is clearly the more basic fact: it is only in virtue of his status as the necessary existent that some things which are merely possible in themselves exist at all.

God is perfect

Avicenna holds in Healing VIII.6.1 that God's perfection consists in his necessary existence:

The Necessary Existent is thus perfect in existence because nothing belonging to His existence and the perfections of His existence is lacking in Him. Nothing of the genus of His existence is extraneous to His existence, existing in another in the way, for example, it exists extraneously in another in the case of a human being. (Avicenna (2005), 283)

In subsequent passages in Healing VIII.6.2 Avicenna adds that God is ‘above perfection’ because ‘not only does He have the existence that belongs only to Him (since God is necessary in itself), but every [other] existence also is an overflow of His existence and belongs to Him and emanates from Him’ (Avicenna (2005), 283).

Avicenna's account of God's perfection is particularly interesting in the present context, as he not only thinks that the perfection is a consequence of necessary existence, but he seems in addition to think that perfection (or being ‘above perfection’) is identical to the kind of self-sufficiency that only the necessary existent in itself can possess. That is, there is a contrast with Avicenna's proof of God's perfection and his proofs of other non-relational claims about God, including his uniqueness and simplicity. When giving the latter proofs, Avicenna relies on substantive claims about the relationship between simplicity or uniqueness, and necessary existence. For example, in proving uniqueness, Avicenna relies on the claim that, if there were two necessary existents, there would have to be some distinguishing feature that makes them distinct. But Avicenna relies on no auxiliary premises in proving perfection; all he relies on is the definitional claim that the self-sufficiency characteristic of the necessary existence suffices for perfection.

It is also worth noting that, when arguing that God is pure good, Avicenna adds in Healing, VIII.6.3 a relational conception of goodness. The good, Avicenna says, is ‘that which everything within its [own] bound desires and through which its existence is completed’ (Avicenna (2005), 284). Since everything desires existence, they desire God. This is a claim that partly depends on the previous argument that God is (beyond) perfection. God's perfection consists in the fact that God is the single, unique necessary existent; that is, God is the perfection of existence. The relational fact – that everything desires perfection – thus amounts to the relational fact that everything desires God.

We won't try to settle how Avicenna thinks of the relationship between these two arguments for God's perfection. Avicenna may have had two distinct concepts of perfection; or, alternatively, may have held God as perfect because he is desired is a consequence of God's intrinsic perfection. There are clear differences between these arguments: one identifies an intrinsic feature of God's necessary existence; the other rests on a relation between God and other creatures. However, they both share a distinguishing mark of a meta-theology: they derive God's perfection from a single fundamental claim about the nature of God. If there are auxiliary premises in the argument, they are not further claims about God's nature, but rather are claims about creatures, or are definitional claims about perfection.

Comparisons with other meta-theological approaches

Distinct meta-theologies

Kvanvig, in articulating what it takes to be a meta-theology, identifies three substantive meta-theologies: Creator Theology, Perfect Being Theology, and Worship-worthiness Theology. Although he does not explicitly say that there can be no fourth alternative, he does say that they represent ‘the most promising beginnings’, and are ‘the major metatheological competitors’.

We have already outlined the essentials for an argument that Necessary Existent Theology meets the criteria to count as a distinct meta-theology: it has a metaphysical framework, which carves out ontological space for God as the necessary existent in itself. Moreover, the nature of God on this picture can clearly be distinguished from the epistemological steps through which we know that God exists. Avicenna holds that some key claims about God, including his simplicity, uniqueness, status as the first cause, and perfection, follow from God's necessary existence. Since Avicenna is a prominent and influential figure in the history of philosophical theology, his view should count as a major player in the meta-theological debate.

One remaining possibility is that, while Avicenna clearly relies on a central fundamental claim about God's nature, Necessary Existent Theology is not a distinct alternative, since its fundamental claim is not substantially different from that of one of the standard meta-theological views. Thus, it is worth pausing to note the reasons why the fundamental claim of Necessary Existence Theology is not plausibly identified with any of the fundamental claims of the alternatives.

We take it to be clear that Worship-worthiness Theology is not the same is Necessary Existent Theology. Worship-worthiness is a relational property, as it related God to humans, or other potential worshippers. A relational property like this has no place in Avicenna's modal ontology, and so cannot feature in the fundamental claim of Necessary Existent Theology. Some derived claims are relational, in Avicenna's scheme. But the fundamental claim, that God is the necessary existent in itself, is not a relational claim, and is only used to derive relational claims in conjunction with facts about the possible-in-themselves.

The fundamental claim of Perfect Being Theology is that God is perfect. Avicenna agrees that God is perfect. But this claim, on Avicenna's view, is derived: Avicenna claims that it follows from God's status as the necessary existent in itself. As the discussion in the previous section shows, the derivation relies on a definition of what perfection consists in, and in particular defines perfection as consisting in the self-sufficiency that is characteristic of the necessary existent. For Perfect Being Theology, on the other hand, God's perfection is not a derived claim; moreover, the conception of perfection as self-sufficient existence is not one the perfect being theologian is likely to share – a point we return in the conclusion.

We can turn next to Creator Theology. Perhaps the closest resemblance between Necessary Existent Theology and Creator Theology lies in the explanatory priority assigned to God on each view. According to Creator Theology, the fundamental claim is that God is the source of everything else. Necessary Existent Theology, we have seen, includes the fundamental claim that God is the necessary existent in itself. This claim is partly a claim about what explains God's existence, as the explanation is ‘through itself’ rather than ‘through another’. The priority of God as first cause is closely related to the way in which God exists, according to Necessary Existent Theology.¹⁵

As Avicenna argues, the fact that God's existence is necessary in itself implies that God is the cause, or explanation, of the existence of all other things. His proof of God's existence proves that God exists because there must be a first cause, although in an unorthodox sense.

But Avicenna's view is not an unorthodox version of Creator Theology. A meta-theology is not an account of the epistemology whereby we come to know that God exists, and so Avicenna's proof is not a clear indication of the structure of his meta-theology. The metaphysically fundamental claim of a meta-theology can, in principle, be separated from the premises in an argument for the existence of God. In Avicenna's case, this is precisely what we need to do: the argument that God exists, even though it establishes God as the first cause, does not commit Avicenna to the metaphysical claim that God is, fundamentally, the first cause.

The metaphysics of God must instead be read off from Avicenna's modal ontology. This ontology includes the category of necessary existence in itself – that is, existence which is explained through the one existing, and not another. While we employ the notion of a first cause to establish the existence of God, God's essence does not involve any relation (including the relations of being first or prior to) to the possible-in-themselves which God causes. Instead, the fundamental claim about God, for Avicenna, locates God in the modal category of necessary existent through itself. Once we separate epistemological claims from metaphysical claims about what is fundamental to God, it is clear that Necessary Existent Theology is not a variant of Creator Theology.

Simplicity and meta-theology

As we have seen, the doctrine of simplicity plays a central role in Avicenna's derivation of the other theological claims about God, including those that are the fundamental claims according to competing meta-theologies. In the context of characterizing Avicenna's views as a distinctive meta-theology, this raises special problems. Avicenna's commitment to a strong version of simplicity entails that God is not composed of parts, is not composed in part by a form, does not belong to any genus, and does not possess attributes that are distinct from each other, or from God's self or essence. This is not an idiosyncratic commitment of Avicenna's view: divine simplicity is a doctrine that looms especially large in any discussion of Avicenna, but is also endorsed in some form by Aquinas, Duns Scotus, Averroes, Maimonides, and others.

A meta-theology involves a commitment to a distinction between what we have called fundamental and derived theological claims. Here we will briefly sketch how this distinction appears to conflict with divine simplicity. This is not a decisive consideration: we will also sketch several ways in which a meta-theological view can be reconciled with simplicity. There may be additional options for dealing with the difficulty; we will not try to canvass all of them here. The upshot of this discussion is that any reconciliation will require us to refine, to some extent, what a meta-theology amounts to.

Take a claim about God's nature that, according to one's favourite meta-theological view, is not fundamental. For example, assume that the claim ‘God is timeless’ is not a fundamental claim about God's nature. What does it mean to say that God is timeless is less fundamental than some other claim about God's nature? Here we have several options for understanding what less fundamental amounts to. On some ways of understanding what fundamentality amounts to, this follows from the fact that there is a further fact about God's nature which grounds the fact that God is timeless. These are ground-centric conceptions of fundamentality, as they explicate the notion of fundamentality in terms of the notion of ground.¹⁶

From the perspective of a theological view that includes a commitment to divine simplicity, these distinctions are potentially important. Take the idea that the distinction in fundamentality follows from a grounding relation between facts that concern God's nature. This view entails that there are distinct facts about God's nature, as the fact that God is timeless is distinct from some other, more fundamental, fact about God. That God is timeless can't be identical to the fact that God is the necessary existent (or whatever the fundamental fact is), because the fact that God is timeless is less fundamental than, and so distinct from, the fundamental fact that God is the necessary existent. This appears to conflict with a strict view of divine simplicity.

The conflict becomes clearer if we adopt a standard formulation of facts as structured entities involving real-world constituents and properties.¹⁷ If the fact that God is timeless involves an entity – timelessness – this entity must be distinct from entities contained in other facts about God. Timelessness is a constituent of a fact that is grounded by the fundamental fact, whereas the fundamental fact contains no such entity. So God's timelessness is not the same thing as God's necessary existence, a straightforward conflict with simplicity.¹⁸

One way out is implicit in the discussion of relational claims about God's nature in the last section. For example, we pointed out that God's status as the first cause is, for Avicenna, a relational fact which holds between God and distinct existents that are merely possible in themselves. God is the first cause because, as the necessary existent in itself, God is the reason other things which are merely possible in themselves are necessary through another. There is no conflict with simplicity when the derived claims are relations between God and other entities, since the derived fact, which concerns God's status as the first cause, is partly grounded in things outside of God's nature. Here there is no pressure to reject simplicity in order to account for the distinctness of the derived facts. One option is to adopt a similar strategy for other derived claims, by construing them as relational facts.

Another option, which is also compatible with simplicity, is to abandon the ground-centric conception of fundamentality. It is not obligatory to explicate the fundamental and derived facts in terms of the grounding-relations which hold between distinct entities. Instead, we can hold that there is one, single fundamental fact about God, and that this fact can be represented in different ways. This is a representational approach to fundamentality, which is to be contrasted with the ground-centric approach. (See Fine (2012), Sider (2012), and Dunaway (2015) for different versions.) A non-relational derived fact about God, on this approach, is another way of representing the fundamental fact. For instance, Sider (2012) holds that the fundamental facts are those that are expressed with ‘perfectly’ structural vocabulary, while statements of non-fundamental (i.e. derived) fact represent the same fact, with ‘non-structural’ vocabulary.

Applied to the theological claims involved in a meta-theology, the representational strategy goes something like this. If God is the necessary existent through itself is the fundamental claim, then the term ‘necessary existent through itself’ is a perfectly structural term. Derived claims, such as the claim that God is unique, represent the same fact that God is the necessary existent through itself represents. But this is not a statement of fundamental fact because ‘unique’ is not perfectly structural. Thus it is not the case that there are distinct facts in play: that is, there are not two facts, one involving necessary existence, and another involving uniqueness. Instead, what is going on is there is one fact – God's necessary existence – and two ways of representing it.

We do not need to adopt Sider's terminology to make this point: all it requires is that some representations better capture God's essence (self) than others. If we have a distinction along these lines in hand, we can use it to make the fundamentality-distinctions that are central to a meta-theology, without contradicting divine simplicity. Our point here is that any meta-theological project that countenances non-relational derived theological facts must use a strategy along these lines, in order to avoid ruling out major theological competitors.

Conclusion

We have argued that Avicenna's Necessary Existent Theology (i) constitutes a meta-theological position, and (ii) is distinct from common meta-theological positions discussed in the literature. Of course this does not mean that it is the most plausible candidate. In fact, there are a number of grounds on which it might be criticized. It would be of significant interest to a meta-theologian to investigate the comparative plausibility of the view in relation to other options in meta-theology. For example, Necessary Existent Theology according to Avicenna accommodates the claim that God is perfect. Perfection, on this view consists in God's complete self-sufficiency: as the necessary existent in itself, God does not depend on any other being. But a Perfect Being Theologian might object that perfection consists in more than self-sufficiency. In this case, the merits of Necessary Existent Theology involve not just which theological claims it can accommodate, but also how it accommodates them. In conclusion, while we think it is clear that Necessary Existent Theology is a distinct option in meta-theology, we also think its interest consists not primarily in its plausibility, but rather in the questions it raises for how to carry out the meta-theological project.