1.1 Divine Simplicity in Jewish Theology and Philosophy

Commitment to divine simplicity is deeply imbedded – although not entirely uncontroversial – in Jewish philosophical theology. As far back as the first century, Philo of Alexandria (d. 50 C.E.) insists that God is, as Carlos Fraenkel puts it, “both numerically and internally one.”Footnote ² Philo is, of course, a committed monotheist, but he goes beyond numerical unity to affirm internal unity, and he cashes out this latter notion as divine simplicity. God is not “mixed” with anything; God is not composed of anything more fundamental than divinity and not only does not but indeed cannot be mixed with anything either “above” or “beneath” God. Combining scriptural exegesis with insights largely drawn from Plato, and presenting Moses as a profoundly philosophical thinker (albeit one who wrote for broadly political and religious rather than philosophical purposes), Philo works to provide a philosophical basis for Jewish theology.Footnote ³ In doing so, he rejects any notion that there are distinctions within the divine being.

Engagement with Islamic thinkers deeply impacted some sectors of medieval Jewish thought.Footnote ⁴ For instance, Saadia Ben Joseph Al-Fayyumi (d. 942) is in fundamental agreement with some important commitments of kalam in holding that genuine knowledge comes from various sources: it comes from revelation and tradition, but of course it also comes from sense perception, intellect, and proper inferences drawn from perception and truths known a priori.Footnote ⁵ Saadia is also in concord with kalam in his insistence that revealed truths are more important than those known from reason.Footnote ⁶ Fraenkel observes a sharp contrast to Philo: Saadia “does not interpret the biblical revelation of God insofar as it conflicts with reason as part of a pedagogical-political program for non-philosophers”; instead he holds that the only literal predication that can be made about God is “the fact of his existence.”Footnote ⁷ What then is one to do with the biblical depictions of God? Saadia makes the case that these refer to the created effects of divine action rather than to God per se.

Saadia argues, largely on the basis of natural theology, that the divine unity is both numerical and internal. He makes the case that the world must be created (rather than existing necessarily or as somehow self-created), and then he argues from this conclusion that the creator must be free of all composition. Considering divine power, life, and wisdom (which are not only depicted in the Bible but also entailed by the argument from creation), Saadia concludes that these are “coextensive”; and he argues that in point of fact these terms are only distinctly human and thus limited ways of referring to the same entity.Footnote ⁸ As he says, we “are compelled to employ in designating them three expressions,” but “it is not to be imagined that the Eternal, blessed be he, possesses several distinct attributes.”Footnote ⁹ In his polemics against the Christian doctrine of the Trinity, Saadia says that distinction implies not only composition but also corporeality. For “their allegation of the existence within his essence of distinction, with the result that one attribute is not identical with the other, is equivalent to an allegation on their part that he is really a physical being.”Footnote ¹⁰ Saadia is certain: He unflinchingly denies “the existence of any distinction in the essence of God.”Footnote ¹¹

Maimonides (d. 1204), in distinction from Saadia’s grounding in kalam, represents the philosophical tradition that saw “a number of the main doctrines of kalam as philosophically feeble and inadequate.”Footnote ¹² For the proponents of kalam, the role of philosophy is primarily instrumental; it is to be used in the explication and defense of what has been divinely revealed. For Maimonides (and, more broadly, the falsafa), on the other hand, philosophical insight is a vitally important source of theological knowledge, and philosophically precise expressions of theology are of a higher order than common and popular expressions of piety.Footnote ¹³ His Guide of the Perplexed is sometimes judged to be “the single most important Jewish philosophical work ever written,” and in it he takes leave of the direct dependence on divine revelation that is characteristic of kalam and Saadia.Footnote ¹⁴ Despite these important methodological differences, however, Maimonides joins Saadia in commitment to a strong doctrine of divine simplicity.

Following Avicenna, Maimonides presents an argument for the existence of God from contingency. He argues that anything that exists contingently depends upon something or someone else for its existence, and he argues further that, on pain of infinite regress, at some point what is contingent must depend upon what is necessary – which is God.Footnote ¹⁵ From a necessary being (God), he then argues for a strong account of internal unity, “because if the necessary existent were composed, it would be dependent on its components and their composition for its existence.”Footnote ¹⁶ Thus, from necessity, he deduces not only incorporeality and monotheism but also simplicity. Indeed, it is only through concession to weakness of the human intellectual capacity that we speak of “divine attributes” at all – and we most certainly should not be misled by such statements into thinking that God actually has attributes.Footnote ¹⁷

Maimonides is often interpreted as insisting upon a radically apophatic theology. We can know what God is not, and he argues that our speech about God is only reliable in so far as it is making negative claims. This means that any positive claims about God can only be understood as utterly equivocal. Thus, according to Howard Kriesel, “he treats the essential attributes of life, will, power, and knowledge as identical with God’s essence and as forming a single notion.”Footnote ¹⁸ Alexander Broadie notes that both Maimonides and Thomas Aquinas (d. 1274) “stress the concept of the absolute oneness of God, and they describe this oneness in very similar terms”; both Maimonides and Aquinas hold that God’s perfections “do not differ from each other or differ from God,” and both “believe that it is more appropriate to say that God is, than that he has, goodness, and that he is, rather than that he has, wisdom, and so on for all perfections” that are finally to be understood as “identical in God and with God.”Footnote ¹⁹ But where Aquinas opts for an analogical account of theological predication, Maimonides insists that all human terms applied to God are equivocal.

Many interpreters of Maimonides observe an unresolved tension in his thought. On one hand, he denies that our language has any real purchase on divinity (at least with respect to positive predications); on the other hand, he does not want to deny the scriptural witness to the actions of God with persons in and across human history.Footnote ²⁰ Can a deity so simple – and thus so radically immutable – even truly know the particulars of human existence? As Fraenkel explains, Maimonides wants to resolve this dilemma by saying that God knows the particulars of human experience by knowing his own essence and by knowing that his own essence is the cause of everything. But this proffered solution only raises more questions: How can such a God be really distinct from the world if the world is entailed by – or even included in – the divine essence? Are these the only options: either God is unrelated to the world and so does not know the particulars of creaturely existence and action or this world is, strictly speaking, entailed by the divine essence? Fraenkel observes that “we seem to have come full circle,” because while “we started with a God who had no connection whatsoever to existents outside himself,” nonetheless we “end up with a God into whose essence all existents are folded in undifferentiated unity.”Footnote ²¹

Later Jewish philosophers seek to resolve the tension in various ways. One option is to take the route of Levi Ben Gershom or Gersonides (d. 1344).Footnote ²² Gersonides, who has been described as “the most prolific and versatile medieval Jewish scholar,” rejects outright the apophaticism and accompanying equivocation of Maimonides.Footnote ²³ Gersonides understands that a God who is utterly simple and immutable cannot be said to know the changing particulars of temporal creatures. Faced with this dilemma, Gersonides rejects the traditional view of God in favor of a doctrine that “loosens” the concepts of immutability and simplicity and that denies divine foreknowledge.Footnote ²⁴ But another way of resolving the tension is exemplified by the early modern biblical scholar and polymath Baruch Spinoza (d. 1677), who wholeheartedly accepts the entailment that everything is necessary and who concludes that God and the world are one.Footnote ²⁵

1.2 Divine Simplicity in Islamic Theology and Philosophy

Commitment to the oneness (tawhid) of God is absolute bedrock for Islamic theology and piety. As Mohammed S. Zarepour explains, “the cornerstone of Islamic faith, regardless of its interpretation by different branches of Islam, is tawhid; there exists one and only one God Who is the ultimate ground of everything in the universe.”Footnote ²⁶ Tawhid is foundational for Islamic theology, but it is also, according to Khalil Andani, “the most contested Islamic doctrine.”Footnote ²⁷ For there is nothing close to consensus on the best way to understand the claim that God is radically one. And while many important Muslim philosophers and theologians do not hesitate to affirm and defend robust conceptions of divine simplicity, other prominent figures and traditions criticize and reject the doctrine.

Among early Muslim theologians, the mu’tazallites and some other proponents of kalam were committed to the denial of genuine distinctions within God. They believed that divine omniscience, omnipotence, will, and other attributes are “all reducible to the divine essence.”Footnote ²⁸ These divine attributes are “identical to God’s Self or God’s Essence,” and thus “God is metaphysically simple without any distinctions.”Footnote ²⁹ Among other schools of kalam theologians, on the other hand, many thinkers demur from a strong conception of divine simplicity and instead maintain that there are genuine distinctions between the divine attributes and between the attributes and the divine essence. As Andani puts it, “most Sunni theologians in the kalam tradition” affirm “real-distinct entitative attributes for God” that are “mutually distinct.”Footnote ³⁰ Each divine attribute is an “entitative reality (haqiqa ma’ani)” that is a “real and distinct entity” – something that is “eternal, uncreated, and numerically distinct from God’s essence or self.”Footnote ³¹

Islamic philosophers identified as falsafa largely defend the position held earlier by the Mu’tazallites; they resolutely insist upon a strong and uncompromising doctrine of simplicity.Footnote ³² For instance, al-Kindi (d. ca. 873) insists that God is devoid not only of form and matter but also quantity, quality, and relation; instead, God is “pure and simple unity.”Footnote ³³ Avicenna or Ibn Sina (d. 1037) offers an extended and spirited defense of a very strong version of the doctrine.Footnote ³⁴ He defends the view that “there is nothing in him external to his essence,” for God is “total simplicity.”Footnote ³⁵ In a manner that bears some similarities to Anselm’s famous ontological argument, Avicenna argues that God not only exists but also exists necessarily.Footnote ³⁶ From this conclusion he mounts two additional arguments: first, he argues that there can only be one necessarily existing entity (numerical unity); and, second, he argues that any such necessarily existing entity must be completely simple (internal unity). As Davlat Dadikhuda summarizes Avicenna’s position, it involves two important claims. The first is a negative thesis: “God has no metaphysical, physical, or logical parts;” while the second is a positive thesis: “The divine attributes are identical to the divine essence and to each other.”Footnote ³⁷ Avicenna says that the “Necessary Existent does not become multiple in any respect whatsoever;” to the contrary, God’s essence is “utterly unitary, pure truth.”Footnote ³⁸ The “Necessary Existent cannot be of a characterization that entails composition so that there would be some quiddity (that quiddity being necessary to [its] existence) such that that quiddity would have a meaning other than its reality.”Footnote ³⁹ God has “no quiddity, no quantity, no ‘where,’ no ‘when,’ no equal, no partner, and no contrary – may he be exalted and magnified – [and] he has no definition…”Footnote ⁴⁰ Dadikhuda offers a helpful clarification of Avicenna’s theory; the salient point is not that the divine attributes are merely coextensive, it is that they are nothing less than identical. Indeed, nothing less than “conceptual sameness” will suffice, for divine knowledge and divine power and divine volition are all the same.Footnote ⁴¹ So even though finite creatures may look at these elements as if they are distinct attributes, in point of fact such divine predicates are not so much as even rationally or conceptually distinct. To the contrary, what we refer to as distinct attributes are exactly the same thing.Footnote ⁴² Accordingly, human language for God does not truly define God; instead, it works best when it offers meaningful negations and points toward relations between creatures and God.Footnote ⁴³

The insistence that the divine essence is identical to the divine attributes seems to have serious theological consequences “downstream” and at any rate raises many questions. Avicenna understands his doctrine of simplicity to carry serious implications, and he does not hesitate to affirm them. As Andani explains, “since God’s will is identical to his knowledge, power, and Essence,” then the divine will to create this particular world “is also a necessary action as opposed to a libertarian choice,” and thus Avicenna holds that “God’s act of creation is necessitated by God’s essence.”Footnote ⁴⁴

Moreover, if God’s knowledge is the same as God’s essence, then it would seem that a simple and immutable God could not have knowledge of the constantly changing particulars of creaturely existence. If the divine essence just is the same thing as the divine knowledge, then is not God’s knowledge that, say, Tom will eat ice cream at time t1 identical to God’s own essence? If so, then is not the divine essence inextricably tied to Tom’s-act-of-eating-ice-cream-at-t1 – and thus in some sense even dependent upon Tom actually eating ice cream? If the divine essence exists necessarily, then is Tom’s act of eating ice cream also, strictly speaking, necessary? (Or, taken the other way, is the divine essence and existence thus contingent upon Tom’s act of eating ice cream? If Tom were to refrain from eating ice cream, would God fail to exist or fail to exist as God?) If the divine essence is identical to the divine knowledge of any particular event, then it seems obvious that the modal status of “both” is the same – either both are contingent, or both are necessary. Many of the falsafa were willing to bite the bullet here and simply deny that God has knowledge of particulars. Yes, God’s knowledge is identical to God’s essence, but God’s knowledge does not include contingent creaturely activities. Avicenna, however, is commonly interpreted as someone who departs from the common opinion of the philosophers at this juncture; on this view, he maintains that God knows the details of creaturely existence, but he insists as well that God has such knowledge in virtue of God’s own exhaustive knowledge of God’s own essence and God’s understanding of the causal efficacy of the divine essence and the discrete particulars that it necessitates.Footnote ⁴⁵

Among the medieval proponents of kalam, al-Ghazali attacks the views of the falsafa on several fronts. He is especially exercised to criticize the philosophers for denying that God knows particulars, for their advocacy of the eternity and necessity of the world, and for their apparent rejection of the resurrection of the body in the eschaton.Footnote ⁴⁶ Al-Ghazali criticizes the work of the philosophers with their own logical and metaphysical weapons, and he works to show that their conclusions are internally incoherent as well as inconsistent with historic Muslim faith and piety.Footnote ⁴⁷ He denies that God’s essence and attributes are identical; he also insists that they are coeternal and uncaused. He says that when we refer to “God, the Exalted, then we have referred to the divine essence together with the divine attributes, and not to the essence alone.”Footnote ⁴⁸ To account for this, he offers a mereological illustration: “Zayd’s hand is not Zayd and is not other than Zayd; rather both expressions are absurd.”Footnote ⁴⁹ He says that “every part is not other than the whole, nor is it the same as the whole,” and he concludes that “it is possible that an attribute is other than the essence in which the attribute subsists.”Footnote ⁵⁰ The “attributes are not the essence;” instead they are “additional to the essence.”Footnote ⁵¹ For “the attribute is other than the essence, and the essence is other than the attribute…”Footnote ⁵²

In summary, it should be obvious that while there is a deep and fundamental commitment to the absolute oneness or unity of God within classical Islam, there is nothing resembling a consensus with respect to the best way to understand such a commitment. Everyone agrees on numerical unity with respect to God – after all, monotheism is nonnegotiable. But when it comes to internal unity, there is no such agreement. Some theologians (including some Sunnis and some Shi’a) as well as the medieval falsafa insist upon nothing less than a strong account of divine simplicity – and then differ among themselves with respect to the entailments of the doctrine. For instance, some of the falsafa think that commitment to the doctrine of divine simplicity carries with it the implication that God does not and indeed cannot know particulars. Other falsafa retain the unyielding commitment to divine simplicity but seem quite willing to accept the necessitarian implications of that doctrine.Footnote ⁵³ Other theologians argue against the falsafa; they protest that the views of the philosophers produce not only error and impiety but even blasphemy, and they are willing to reject strong doctrines of simplicity. Still others, including some of the Shi’a “Ismaili” theologians and later the Sufis, take a “radically apophatic approach to the theology of tawhid”; these thinkers affirm a very strong account of simplicity but then go on to argue that knowledge of the divine essence and speech about it absolutely transcend all creaturely categories.Footnote ⁵⁴ Whereas thinkers such as Avicenna limited proper speech about God to what is said by negations and relations (along with the affirmation that God’s existence is necessary), the Sufi strategy is to deny that any of our statements can apply to God.Footnote ⁵⁵ These debates continue to the present day, with some theologians and philosophers arguing for older theories and others working to forge new routes.Footnote ⁵⁶

1.3 Divine Simplicity in Christian Theology and Philosophy

Richard A. Muller says that “the doctrine of simplicity is among the normative assumptions of theology from the time of the church fathers, to the age of the great medieval scholastic systems, to the era of Reformation and post-Reformation theology, and indeed, on into the succeeding era of late orthodoxy and rationalism.”Footnote ⁵⁷ It is believed and taught by Christian theologians across a broad range of times, places, and ecclesial locations, and it is held by not only Roman Catholic but also Orthodox and Protestant divines of various confessions as well.

1.4 Summary: The Doctrine of Divine Simplicity?

Just as it should be obvious that commitment to divine simplicity is widespread in Christian, Jewish, and Muslim traditions, so also it should be obvious that there are important differences in the various articulations and defenses of the doctrine. For instance, in Christian theology there are important formulations in the Greek tradition that differ rather sharply from many influential Latin statements.Footnote ⁸¹ As Scott M. Williams explains, “the Greek side affirms the simplicity of the divine essence” while it also “accepts contingency and numerical distinction” between various divine energies, while much Latin theology does not recognize a distinction between the divine essence and the divine energies.Footnote ⁸² Beyond this, as we have seen (and shall see in more detail), there are very important differences within the Latin scholastic tradition. Given the variegation in formulations of the doctrine of divine simplicity, one might wonder if it even makes sense to refer to “the doctrine.”Footnote ⁸³ Is it even meaningful to refer to so many views under one umbrella term? The question alerts us to an important reality that might easily be missed: There is no single “standard” or “official” statement of the doctrine. Indeed, efforts to subsume all accounts under one tightly defined theory are mistaken and misleading.Footnote ⁸⁴ At the same time, however, it is also important not to overreact and emphasize the diversity so much that the significant common ground is lost. Just as there are very real and significant differences between, say, Anglicanism and Lutheranism and Presbyterianism, nonetheless it is also true that all assent to (some version of) what is recognizably Protestant doctrine. Or again: Just as there are real and significant differences between, say, Banezian Thomists and Molinists, so also it is true that there remains such a thing as a general doctrine of divine omniscience. Disagreement – even significant and very forceful disagreement – on issues of higher-resolution detail neither undercuts nor overrides genuine agreement on other points. The situation is similar with respect to the doctrine of divine simplicity.

But there can be no doubt that there is considerable variegation with respect to the formulations of the doctrine. To get an initial grasp of this, compare:

with

It is important to see that most Christian theologians have held something close to M-DDS and have rejected or held at arm’s length S-DDS.Footnote ⁸⁵ In other words, Christians have almost universally maintained that the doctrine of divine simplicity rules out the possibility that there are “parts” of God or that God is reliant upon anything other than God’s own self for God’s existence and identity. But they have not ruled out the possibility that there are genuine distinctions in God’s own essence or Triune life. For instance, the Lutheran scholastic theologian Johann Gerhard says that “the question is not whether there are in God some real distinctions that effect a real multitude or plurality.”Footnote ⁸⁶ Given the doctrine of the Trinity, of course there are real distinctions between the divine persons. Instead, the question at stake is this: Are there “in God some real components that effect real composition”?Footnote ⁸⁷ And the answer to this question is a consistent and resounding “No.” Christian theologians are broadly (indeed, almost universally) committed to a notion of numerical sameness without identity; indeed, this commitment is so widespread in medieval theology that Andrew Arlig cautions against any presumption that “the principle or core notion of sameness for a medieval is Leibnizian identity…”Footnote ⁸⁸ Christian theology (in its premodern modes) is deeply committed to the doctrine, but many scholastic Christian theologians were also quite willing to be “flexible” as they developed the doctrine in various ways.Footnote ⁸⁹ In other words, they are certain that God is “absolutely simple” (thus Scotus) but ready to dispute among themselves about exactly what such a claim means and entails. Many Muslim and Jewish theologians, on the other hand, seem to be committed to (S-DDS); they are much less willing to adapt or develop the doctrine – but some appear considerably more willing to reject it altogether.

Despite such historic commitments in Judaism, Islam, and perhaps especially Christianity, however, the doctrine has attracted – and continues to attract – criticisms. To these we now turn.

5.1 The Challenge

According to creedally orthodox accounts of the Trinity, the divine persons are genuinely distinct (or what some scholastics referred to as distinction ex natura rei) from one another.Footnote ¹⁵⁴ Indeed, for much of the tradition, it is entirely safe to say that the distinctions are nothing short of real distinctions (in the technical sense). This yields what we can refer to as:

At the same time, however, the traditional formulations of the doctrine of divine simplicity also exert pressure on Christian theology, and one prominent version of that doctrine (ratified at the Fourth Lateran Council) delivers the verdict that the divine persons just are the one ultimate divine reality that is the single divine essence. This yields what can be referred to as:

Some contemporary theologians, as well as many theologians in the tradition do not hesitate to affirm both (TRIN) and (SIMP). In our day, Adonis Vidu insists that “one must speak of numerical identity between person and essence” because “the distinction between the persons and the essence is not real but only conceptual” even though the “distinction between the persons is, against Sabellianism, a real distinction.”Footnote ¹⁵⁵ Vidu’s statements arguably have precedent. As Thomas Marschler observes, after Lateran IV “every subsequent theory of the distinction between the divine essence and the persons of the Trinity had to avoid a real distinction.”Footnote ¹⁵⁶ But every subsequent theory also needed some kind of distinction.

Here is why. According to classical logic, identity is reflexive, symmetrical, and transitive, and the identity relation yields the indiscernibility of identicals. Accordingly, claims that the divine persons are really distinct from one another but really and numerically identical with the divine essence run the risk of incoherence. Obviously, Christians committed to creedal orthodoxy affirm (TRIN); they reject the notion that the Father and Son are distinct only in our way of thinking about things and instead affirm that the Father and Son (and Holy Spirit, of course) are really distinct from one another. But if they also affirm (SIMP), then they are faced with a serious problem. For if the Father is identical to the divine essence, and the Son (and Spirit) is identical to the divine essence, then, given the transitivity of identity, it follows logically that the Father is identical to the Son (and also to the Spirit). But such a conclusion directly contradicts (TRIN) – according to (TRIN), the Father is not identical to the Son, but, on the entailment that comes with (SIMP), the Father is identical to the Son. And since, according to (classical) logic, contradictions are necessarily false, any (non-glutty) theology that includes both (TRIN) and (SIMP) (either by affirmation or by entailment) is not only false but even necessarily false.

This conclusion follows logically. According to classical logic (or, indeed, any system of logic hospitable to the transitivity of identity), the conclusions – problematic as they are for orthodox Christian theology – are inevitable. Consider the following argument for (TRINITY FAIL):

1. (1) If sentence $(\alpha )$ is true, then identity indeed is transitive;

   Sentence $(\alpha )$: $(\forall \text{x})(\forall \text{y})\left[\text{x}=\text{y}=>(\text{P}(\text{x})\iff \text{P}(\text{y}))\right]$

2. (2) Sentence $(\alpha )$ is true;

3. (3) Therefore, identity is transitive (from 1, 2);

4. (4) Sentence $(\beta )$ is true;

Sentence $(\beta )$: $\left[(f=\text{E})\&(s=\text{E})\right]$. The Father is identical to the divine essence, and the Son is identical to the divine essence.

1. (5) Therefore, $f=s$; The Father is identical to the Son (from 3, 4).

Unfortunately, (5) has no shot at compliance with creedal orthodoxy. And matters get even worse. For if the Father and the divine essence are identical, then any property possessed by the Father must also be possessed by the divine essence. Moreover, if the persons’s identity with the divine essence entails the persons’s identity with one another, then whatever property is possessed by one divine person must be possessed by the other person. According to the mainstream Latin tradition, however, the Father just is the divine essence plus the personal property of paternity that makes him distinct. And, mutatis mutandis, so is the Son, and so is the Spirit. According to mainstream (Latin) Trinitarian formulations, the Father is not merely the divine essence; the Father is the divine essence plus his personal property (paternitas). Similarly, the Son is not merely the divine essence; the Son is the divine essence plus his personal property (filatio). The same point applies more generally. Even James Dolezal’s defense of a (broadly) Thomist doctrine appeals to what can only appear to be differences in properties between the essence and the persons: The essence is absolute and communicable, while the persons are relative and incommunicable.Footnote ¹⁵⁷ But if the divine essence has the properties of being communicable and being absolute, while the divine persons have properties of being incommunicable and being relative, then the persons and the essence clearly have different properties. They are not, and on pain of heterodoxy cannot be, numerically identical. Thus neither Father, Son, nor Holy Spirit is numerically identical to the divine essence.

So it should be clear that if the Father were identical to the divine essence and the Son were also identical to the divine essence, then they would be identical to one another. But consider further:

1. (2) Sentence $(\alpha )$ is true;

2. (5) $f=s$;

3. (6) $\text{P(}f\text{)}$, the Father has the personal property Paternity;

4. (7) Therefore, $\text{(}f\text{)}\text{(}s\text{)}\left[f=s=>(\text{P}(f)\iff \text{P}(s))\right]$; if the Father and Son are identical, then, if the Father has the personal property Paternity and is identical to the Son, the Son has Paternity too.

So if the persons are really identical, then, because any property possessed by one must be possessed by the other, and the Father and Son would have all their properties in common. Things get even worse. Not only would both the Father and the Son have the property paternitas, it turns out that both Father and Son would have the property filatio too. But the mainstream Latin tradition’s insistence on the irreducibility of the “relations of opposition” and the “personal properties” that distinguish the divine persons would make this impossible. And surely it would be pointless to reject Sabellianism if the personal properties are confused or shared. Numerical sameness? For the mainstream Latin tradition, yes. Identity? No – absolutely not. So, if we were to affirm with Vidu that the Father and Son are each identical to the divine essence, then we would be committed to the affirmation of claims that are undeniably false according to orthodox doctrine.

Second, given classical logic, any contradiction is not only false but indeed is necessarily false. Affirming that the Father and Son are each identical to the divine essence entails conclusions that are false (for orthodoxy). But if we were to go further and affirm that the Father and Son both are and are not identical, then we would be affirming a claim that is not only false but even necessarily false. Surely that is not a viable route forward for Trinitarian theology (at least one that wishes to retain continuity with the tradition). But – just as surely – something must be done. Scholastic theologians recognized this point, and they gave much effort to articulating accounts that affirm numerical sameness but that do not affirm numerical identity.

5.2 Rational Distinctions: Aquinas and Thomism

Aquinas himself considers and addresses the basic concern that I have highlighted here; he too is aware that simultaneous affirmation of (TRIN) and (SIMP) generates serious concerns. Unlike some of his twenty-first century defenders, he takes the following objection seriously: “Simultaneous affirmation and negation of the same things in the same respect cannot be true. But affirmation and negation are true of essence and of person. For person is distinct, whereas essence is not. Therefore, person and essence are not the same.”Footnote ¹⁵⁸ He addresses the objection head-on, and he apparently thinks that he has an adequate response to it.

Let us first be clear about what Aquinas does not do: He does not deny that contradictions are necessarily false. Nor does he deny the transitivity of identity. He recognizes the force of the criticism, for elsewhere he says this: “It would seem that the divine relations are not really distinguished from each other. For things which are identified with the same, are identified with each other. But every relation in God is really the same as the divine essence. Therefore the relations are not really distinguished from each other.”Footnote ¹⁵⁹ So, clearly, Aquinas is aware of the concern; given the transitivity of identity, if two relations (in this case, subsistent relations, i.e., persons) are really identical to the divine essence, then it would seem that they are identical to one another. Aquinas does not deny the transitivity of identity, and he admits that the standard (Aristotelian) position is strong with respect to entities that are both really and logically identical. But he denies that in this case there is logical identity, and he insists that the subsistent relations are not identical because they are relations of opposition.

So Aquinas sees the challenge, and he takes it seriously. But what are we to make of his response? It is not clear that it is successful as it stands. It looks like he tries to avoid the problematic entailment by appeal to what can only be (for his account) another strictly conceptual distinction. As he puts it, “as essence and person differ in our way of thinking, it follows that something can be denied of the one and affirmed of the other…”Footnote ¹⁶⁰ Surely Aquinas is right in his observation that person and essence are distinct in human ratiocination. But is that all? Is the relevant distinction only in human ways of thinking? And if so, does this work? Does a merely conceptual distinction block the troubling inference? Can a merely conceptual distinction secure a real distinction within God’s own life? It is far from clear that Aquinas’s response resolves the dilemma. For if the distinction is not mind-independent but instead relies upon (or perhaps is grounded in) human cognition, then the distinction is either not really in the being of God or God is dependent on human ratiocination for the constitution of the divine being. Either way, it is hard to be optimistic about this strategy. For merely conceptual distinctions do not close the door to anti-realism – and surely anti-realism cannot secure a real distinction. As Bruce Marshall says, “surely… Aquinas is just covering a mental hiatus with verbal plaster.”Footnote ¹⁶¹

Apparently recognizing the inadequacy of Aquinas’s response, later Thomists offer further refinements, and they deploy these to address the challenge. The early modern scholastics draw further distinctions here that were not explicit in medieval theology; they distinguish between the distinctio rationis ratiocinantis and the distinctio rationis ratiocinatae quae habit fundamentum in re. The first of these is easy enough to grasp; this is the “distinction by reason reasoning” or distinction strictly on the part of the reasoner. This is, as Richard A. Muller puts it, “a merely rational distinction resting only on the operation of reasoning and not on the thing.”Footnote ¹⁶² To revisit our illustration, when we talk about “the morning star” and “the evening star,” we surely are not talking about two different and separable entities. To the contrary, we are referring to one planetary body (Venus), and while our varied references to it (as the “morning star” and “evening star”) are meaningful, they are meaningful simply as a matter of convention or because they tell us something about our relation to the single planetary object.

The “distinction of reason belonging to the object of reasoning with a foundation in reality” (distinctio rationis ratiocinatae quae habit fundamentum in re), along with what is sometimes called the “virtual distinction,” is meant to occupy logical space between the merely rational or conceptual distinction and the real distinction. That much is clear. But exactly what it is meant to do is rather less clear. And whether or not the deployment of it is successful is less clear yet. The basic idea is that it is a distinction whose actuality depends entirely on the operation of the intellect of the cognizers. Yet it is also supposed to be, in the words of Muller, “a distinction expressive of extramental reality since it is grounded on the thing and therefore preserved from being merely a product of mind.”Footnote ¹⁶³ Muller further says that it “represents no distinction in the thing but a truth of reason concerning the thing.”Footnote ¹⁶⁴ Richard Cross quotes John of St. Thomas (John Poinsot) in saying that “the distinction is not merely that the object is signified variously by cognate concrete and abstract nouns with the same semantic content, as in the rational distinction belonging to the reasoner” – yet “the relevant predications have one and the same simple truth-maker”; for as “Poinsot puts it, ‘The foundation for the rational distinction belonging to the object of reasoning, on the side of the object, is some virtual distinction [in the thing], or the thing’s eminence, which, as a singular existent, contains, in some esse, many notions or perfections.’”Footnote ¹⁶⁵

But exactly what this amounts to remains unclear. Suppose we take the theory to deny that there are genuine distinctions within the entity itself but instead hold that the entity in question happens to (more or less reliably, or perhaps even necessarily) “trigger” a particular cognitive response whereby the one thinking about it (more or less reliably, or perhaps even necessarily) conceives of distinctions. In this case, it is hard to see how this isn’t just a fancier version of a merely rational or conceptual distinction. If there really is no corresponding distinction within the entity itself but only a “trigger” in me so that I think that there is one (whether more or less reliably, or even necessarily), then the distinction might tell me something about me but nothing about the entity. Or, on the other hand, if there is actually something “in” the entity itself that corresponds to the cognizer’s triggered reaction, then this is hardly a distinction of reason only.

The virtual distinction is closely associated with the distinctio rationis ratiocinatae. Sometimes they are conflated, but some defenders of Thomism may take the virtual distinction to exist prior to and independent of the operation of the intellect – but nonetheless only in potency rather than actuality.Footnote ¹⁶⁶ The rational distinction with a foundation in reality, on the other hand, is actual, but it is actual precisely because it is posterior to the operation of the intellect that cognizes it. The relation of all this to Aquinas’s own teaching remains contested and still somewhat obscure.Footnote ¹⁶⁷ Given the remaining questions, it is not yet obvious that such strategies are successful. But this is a venerable tradition, and it may yet be shown to be viable.Footnote ¹⁶⁸

5.3 The Modal Distinction: From Durandus to Duby

The distinctio modalis is yet another attempt to account for the logical and metaphysical space between real and merely rational distinctions. The modal distinction in its mature form is widely associated with Francisco Suarez.Footnote ¹⁶⁹ The term has a background at least as far back as Henry of Ghent and Scotus, but they did not develop it as it was later used, and it does not seem to have been widely understood or employed.Footnote ¹⁷⁰ Durandus of St. Pourcain employs a version of it, but, again, the definition of it awaits clarity. Prior to Suarez, as Robert Pasnau points out, “there was no fixed technical sense of mode.”Footnote ¹⁷¹ Pasnau judges Suarez to be “breaking new ground” and “doing something new” when he develops and clarifies the distinction.Footnote ¹⁷²

So just what is the modal distinction? It is not a replacement for either the real or rational distinctions; Suarez remains committed to these and indeed finds them obvious and the arguments for them compelling. But is there room for anything else – can there be anything other than real and rational distinctions? On one hand, Suarez feels the force of the arguments that “there can be no middle ground between being dependent on mind and being independent of mind, and no middle ground between being the same and being different.”Footnote ¹⁷³ On the other hand, however, he also sees the need for something between these poles. Thus he posits the modal distinction as different than the real distinction; whereas the latter holds “between res and res,” the former obtains “between res and mode.”Footnote ¹⁷⁴ As Jorge J. E. Gracia explains, “modes are not realities independent of the things they modify, but nevertheless are intensionally distinct from them.”Footnote ¹⁷⁵

So Suarez does not hesitate to endorse the real and rational distinctions. And he distinguishes further between the rational distinctions that have a foundation in reality (fundamentum in re) and those that do not. He is aware of the scholastic theologians who reject any possibility of further distinctions beyond the real and the rational (e.g., Ockham, Hervaeus, Cajetan, and DeSoto).Footnote ¹⁷⁶ He is critical of Scotus’s development and defense of the formal distinction as a third option.Footnote ¹⁷⁷ Instead, he presents his modal distinction as the best and indeed the only possible third distinction. Explicitly rejecting the formal distinction, he says that “besides real, modal, and mental distinctions there is no other which is not common to these, or comprised in them.”Footnote ¹⁷⁸

Modal distinctions can pertain to quality, quantity, and intensity. As Gracia says, “thus a mode of walking, say quickly, is nothing other in fact than walking, but it is something intensionally distinct from just walking, since walking can also be done slowly.”Footnote ¹⁷⁹ Brighter red is modally distinct from darker red, and both the brighter red and the darker red are modally distinct from the redness itself. Importantly, the entity in question (in these cases walking and redness) can exist without the mode, but the mode cannot exist without the entity. For neither the brighter red nor the darker red can exist without the redness, but the redness can exist without either this mode of brightness or this mode of darkness.

One-way separability is widely understood to be the key identifying characteristic of the modal distinction.Footnote ¹⁸⁰ Real distinctions are independent of mind and are marked by logical and ontological separability. Real distinctions, in other words (commonly and standardly), yield two-way separability.Footnote ¹⁸¹ Rational distinctions, at the other extreme, are mind dependent and conceive of “things which are not distinct as though they were distinct” – and thus are of course strictly inseparable.Footnote ¹⁸² Modal distinctions, as the third (and, for Suarez, only additional) category between rational and real distinctions, have a kind of objective reality in the sense that there is one-way separability: The entity can exist without the mode under consideration, but the mode cannot exist without the entity. Suarez says that if two entities can remain in existence when separated, then we are dealing with a real rather than modal distinction: “Whenever one thing can be separated from another and preserved in its separated state, an evident sign is at hand that it is not the mode of that other.”Footnote ¹⁸³ On the other hand, he says that “if two things are actually separated in such a way that one continues in existence but the other does not, they must be at least modally distinct.”Footnote ¹⁸⁴

Steven J. Duby employs the modal distinction. Appealing to Aquinas and the Reformed scholastics as important proponents of the theory, he says that “proponents like Thomas, Zanchi, Mastricht, Turretin, and others carefully uphold that God’s being is free from real and formal distinctions and yet, under the relative aspect, includes modal and even real relative distinctions.”Footnote ¹⁸⁵ In response to the argument of (TRINITY FAIL), he says that

Armed with the modal distinction, he insists that “it also becomes clear that the identity of each of the persons with the one God does not entail that ‘each person of the Trinity is identical to the whole Trinity,’” for “the modal distinction between a given person and the divine essence allows that other persons who are distinct modus subsistendi also should be really identical with God considered essentially.”Footnote ¹⁸⁷ And thus he is convinced that we can circumvent “the idea that if one person is identical to the essence, then he must be identical to the other persons.”Footnote ¹⁸⁸ Indeed, he is certain that the modal distinction obtains not only between the divine essence and the divine persons but also between the divine persons.

Duby is confident that the modal distinction allows one to affirm both (TRIN) and (SIMP) without risking incoherence. But it is less than obvious that this strategy delivers. When we consider again the standard examples of the modal distinction, it is not hard to see that they do not map neatly onto the doctrine of the Trinity. For instance, quantity is hardly appropriate or helpful as an example of the modal distinction when considering the doctrine of God. Quality and intensiveness seem like even less obvious candidates to help us make sense of the doctrine; surely that way lies the Arian heresy. And consider what happens if we try to apply one-way separability (which, again, is the hallmark sign or characteristic of the modal distinction). Suppose, following Duby, we think of the divine persons as modally distinct from the divine essence. Given one-way separability, we would be left with the conclusion that while the divine persons could not exist apart from the divine essence, the divine essence surely could exist apart from the divine persons. But according to Trinitarian orthodoxy, neither the divine essence nor the divine persons are more fundamental than the other. And, traditionally, the divine persons are understood to exist necessarily rather than contingently. All this seems to be at risk if we think of the divine essence to be modally distinct from the divine persons.

If possible, things get even worse for the erstwhile orthodox Christian who holds to both (TRIN) and (SIMP) and who also maintains that the divine persons are modally distinct from one another. Again, if we have one-way separability between the divine persons, then we would be left with the entailment that some divine person could exist apart from the other person(s) – for instance, while the Son could not exist apart from the Father, the Father could exist apart from the Son. But, again, surely this is to court heterodoxy. For if the Father exists necessarily while the Son only exists contingently, then it is hard indeed to conceive of any meaningful sense in which the Father and Son are of one essence or substance (homoousios).

Even Suarez, who is arguably the foremost scholastic champion of the modal distinction, tosses it aside when considering the doctrine of the Trinity. After reflecting on it at length, Suarez is adamant in his rejection of it as unsuitable for the doctrine of the Triune God: “I deny that the divine essence is separable from the paternity.”Footnote ¹⁸⁹ For “it is one thing for the divine essence to be communicable to filiation or to procession, but quite another for it to be separable from the paternity.”Footnote ¹⁹⁰ On the contrary, he insists that “of its very nature the divine essence must be so communicated to the Son in filiation and to the Holy Spirit in procession that it is not separated from the Father in his paternity.”Footnote ¹⁹¹ His conclusion is starkly clear: The modal distinction, at least as it is commonly understood and vigorously defended by him in other contexts, “has no application here,” for it fails as an account of the distinction between person and essence.Footnote ¹⁹² Indeed, it “is greatly opposed to the divine simplicity and perfection.”Footnote ¹⁹³

In light of these concerns, it seems safe to conclude that appeal to the modal distinction to avoid the conclusion of (TRINITY FAIL) is not a good option.

5.4 The Formal Distinction: The Scotist Alternative

There is another important option within the tradition of Christian theorizing. This is the appeal to the formal distinction between the divine essence and the divine persons. Despite its neglect in late modern theology, it was a much-debated and widely held position in late medieval and early modern scholastic theology. The distinctio formalis represents a further attempt to make sense of reality by finding space between the real and merely rational distinctions.Footnote ¹⁹⁴ The “formal distinction,” championed by Scotus and his followers, posits genuine distinction that is inherent in the entity in question as well as inseparability.Footnote ¹⁹⁵ It occupies the logical space “between” a real distinction and a merely conceptual distinction. As Bradshaw puts it, any two entities are “formally distinct when they cannot exist separately (and thus are not ‘really distinct’)” but nonetheless are genuinely distinct.

Despite its neglect in contemporary theology, there is ample support for the formal distinction in scholastic theology. And it promises to have the resources needed. Cross’s summary is helpful:

What this yields, then, is numerical sameness without identity. If this is right – and, crucially, if there are no other damning problems or fatal objections with respect to the “formal distinction” – then it seems that we have enough to block the inference. For if the Father is formally distinct from the divine essence, then, even though he is not separable from it and it is not separable from him, nonetheless there is a genuine distinction between the Father and the divine essence. And so also with the Son and Spirit. As Bradshaw concludes, each of the persons is “thus also ‘really identical’ with the essence, in the limited sense of being unable to exist separately from it; but real identity in this sense (which is very different from Aristotle’s) clearly is not transitive, and so does not threaten the distinction of the persons.”Footnote ¹⁹⁷ Is not this what the Christian is looking for?

5.5 More Radical Responses

None of these options can claim anything approaching consensus within the tradition, and theologians and philosophers of religion continue to raise concerns about them. But perhaps there are other options available.Footnote ¹⁹⁸ Christians might simply decide to do Trinitarian theology without the doctrine of simplicity. To do so, we might think in terms of composition as identity. Some contemporary Trinitarian theorizers refer positively to “parts” in God. These affirmations happen across the spectrum of theories or models. From the “Latin Trinity” side of things, Brian Leftow uses the language of parts: “God eternally has three parts of his life going on at once, without succession between them. One part is the Father’s life, one the Son’s, one the Spirit’s.”Footnote ¹⁹⁹ On Leftow’s account, these parts “add up to the life of one God.”Footnote ²⁰⁰ Meanwhile, on the “Social Trinity” side, William Lane Craig and J. P. Moreland suggest that “we could think of the persons of the Trinity as divine because they are parts of the Trinity, that is, parts of God,” and they conclude that “it seems clear that there is some sort of part-whole relation obtaining between the persons of the Trinity and the entire Godhead.”Footnote ²⁰¹ So indeed there are various mereological proposals before us.

As I am using the term, “Composition as Identity” (CAI) does not mean that the composition relation is analogous to identity.Footnote ²⁰² Nor does it mean that composition is what is sometimes referred to as nonnumerical identity (whatever exactly that is).Footnote ²⁰³ Instead, what I mean is that the composition relation is nothing short of numerical identity.Footnote ²⁰⁴ CAI is, of course, deeply controversial, and I shall not address the common arguments for and against it here.Footnote ²⁰⁵ Instead, I shall focus only on its suitability (or lack thereof) for the Christian doctrine of God.

The promise for a doctrine of the Trinity that wants to avoid the common challenge should be obvious. If the divine persons are parts – or, at least, relevantly like parts – of the Trinity, then neither the Father, the Son, nor the Holy Spirit is identical to the divine essence. Thus there is no reason to think that the problematic entailment generated by transitivity will even get off the ground. It seems that there are (at least) two possible ways of moving forward here; one could hold that the persons each instantiate the divine essence and then compose the Trinity and are thus collectively identical to the Trinity. Or one might – with at least a nod toward divine simplicity – posit that the divine persons collectively compose the divine essence that (somehow) just is the Trinity.

But the shadows are deep, and dangers may lurk there. It might seem as though the indiscernibility of identicals and transitivity will still cause problems. If the personal parts are identical with the Triune whole, then of course the personal parts won’t have properties that are distinct from the Triune whole. But of course the personal parts have properties – that is, paternity and filiation – that the Triune whole does not have. And if the personal parts that each instantiate the divine essence are identical to the Triune whole, then the personal parts are identical to one another. Or so it might seem.

But these objections are based upon misunderstandings of CAI. As Meg Wallace says, the CAI

Thus, as Wallace says, there is “no violation” – at least none that is obvious – and thus nothing to worry about in this regard.Footnote ²⁰⁷

It is worth noting that some of the standard criticisms of CAI do not seem particularly troublesome for the Trinitarian theological application (whatever one makes of them in other contexts). For instance, Wallace mentions “persistence worries” that come in both temporal and modal varieties, but these need not seem decisive when considering the Triune God who exists eternally (or even everlastingly) and necessarily.Footnote ²⁰⁸ For while it may be true that a table and its constitutive parts exist in overlapping but distinct temporal spheres and have different modal properties, the Triune God is importantly different from the table in this way. So even if identity problems plague tables, such problems do not transfer to God (or at least not obviously).

But other concerns do seem more pressing. Shieva Kleinschmidt argues that the use of CAI takes one no further than accounts of the doctrine of the Trinity known as “Social Trinitarianism,” and she is concerned that such doctrines entail polytheism.Footnote ²⁰⁹ One might try to escape polytheism by appeal to counting by sortals (as is common in cases of CAI), but if one is willing to embrace relative identity, then it is not clear that composition is needed at all.Footnote ²¹⁰ And it should be obvious – at least on the face of it – that such moves are a fairly drastic departure from the mainstream traditions of Trinitarian orthodoxy.

5.6 Summary

There are, of course, other concerns related to the conjunction of the doctrines of the Trinity and simplicity, but it seems that the most pressing is the (TRINITY FAIL) argument. As we have seen, the (TRINITY FAIL) argument assumes the conjunction of (TRIN) and (SIMP). (TRIN) requires that there be genuine and robust distinctions between the divine persons, but (SIMP) denies that there can be real distinctions between the divine persons and the divine essence. As we have also seen, some response to the argument was seen as necessary and important, and the responses to it are quite varied. Some scholastic theologians, including not only many Dominicans but also some Jesuits and many Protestants, followed Aquinas in his deployment of the rational distinction. Indeed, some of the late medieval and early modern scholastics drew further distinctions within the rational distinction (e.g., the rationis rationcinantis and the rationis ratiocinate with a fundamentum in re or virtual distinction). Alas, it is not clear that such distinctions are robust enough to avoid the conclusion of (TRINITY FAIL).

Other scholastic theologians defend and deploy the modal distinction. But there is reason to think that this distinction is rather too robust; the concern is that allowing even one-way separability compromises both the doctrine of simplicity and the orthodox doctrine of the Trinity. Content with neither merely rational distinctions nor modal distinctions, other late medieval and early modern theologians embrace the formal distinction. Sitting between merely conceptual or rational distinctions, on one side, and real and modal distinctions (with two-way and one-way separability, respectively), on the other side, the defenders of the formal distinction insist that this account secures both inseparability (even logically) and genuine distinction. Taking stock, these points seem clear: first, the modal distinction is unsuited for Trinitarian theology; second, despite centuries of hard work by brilliant thinkers, as it stands, the appeal to rational distinctions appears unable to block the dreaded inference and needs further defense if it is to be viable; and, third, although the formal distinction has languished in modern theology, it promises to deliver what is needed to avoid the conclusion of (TRINITY FAIL).

We should be clear that (SIMP) is demanded by Lateran IV and thus seems nonnegotiable for Roman Catholic Christians. But it is not obligatory for Protestant or Orthodox Christians (although it seems that most confessional Protestants in the tradition have worked hard to avoid transgressing it). Bradshaw, however, notes that the formula of Lateran IV is something “that is found in none of the Fathers, East or West, and is far removed from the subtlety and complexity of their thought”; in his judgment “there is nothing necessary or even very plausible about the idea” when “viewed from a patristic perspective.”Footnote ²¹¹ The upshot of Bradshaw’s point is clear: There is nothing about the doctrine of simplicity, as historically conceived within patristic and subsequent Palamite thought, that demands commitment to (SIMP). And so, if it turns out that none of the extant Latin strategies work to reconcile (TRIN) and (SIMP), Orthodox (and, for that matter, Protestant) Christians need not surrender either the doctrine of the Trinity or the doctrine of simplicity.

Finally, we would do well to remember that not only has the doctrine of simplicity been affirmed along with the doctrine of the Trinity, but indeed that for much of the tradition it has been seen as important in the development and defense of orthodox and confessional Trinitarian theology. What Muller says with reference to post-Reformation Protestant scholasticism surely applies more broadly as well: “The doctrine of divine simplicity is a corollary and support of the doctrine of the Trinity.”Footnote ²¹² For as early as the mid-fourth century, we see Gregory of Nyssa endorsing and weaponizing the doctrine of simplicity against the arguments of Eunomius.Footnote ²¹³

In conclusion, we can see that there is more than one way to block the argument of (TRINITY FAIL). Commitment to the doctrine of divine simplicity does not – or at least, depending upon which strategy one takes, need not – rule out belief in the doctrine of the Trinity. Nor does affirmation of the orthodox doctrine of the creeds and confessions militate against the doctrine of simplicity. To the contrary, there are reasons to affirm with the mainstream Christian tradition that the doctrine of simplicity is helpful in the articulation of Trinitarian theology.

6.1 The Threat of Modal Collapse

But what, more exactly, is the problem? One common way of understanding the argument can be summarized as (CONTINGENCY FAIL):Footnote ²¹⁷

1. (1) Necessarily, God exists;

2. (2) God is identical to God’s act of creation;

3. (3) Necessarily, God’s act of creation exists.

Such a conclusion appears to lead directly and inexorably to modal collapse.

However, as Christopher Tomaszewski has shown, this argument can readily be shown to be invalid. Tomaszewski notes that (CONTINGENCY FAIL) commits “the famous formal fallacy of substituting a contingently co-referential term into the scope of a modal operator.”Footnote ²¹⁸ Compare:

1. (1*) Necessarily, 8 is greater than 7;

2. (2*) The number of planets is identical to 8;

3. (3*) Necessarily, the number of planets is greater than 7.Footnote ²¹⁹

From this counterexample, we can see that something has gone badly wrong with the argument. But even though this argument is unsuccessful as it stands, it does raise important issues, and similar arguments (that reformulate the argument in terms of rigid designators) can be made that do not make the obvious mistake that ails (CONTINGENCY FAIL).Footnote ²²⁰ At any rate, there is widespread agreement that defenders of traditional doctrines of simplicity face a genuine and serious challenge. The major responses fall into several broad categories: Some theologians simply admit modal collapse, other theologians try to develop various interpretations of Aquinas’s doctrine, and still others develop alternative accounts that draw inspiration from Palamas, Scotus, or Suarez.

6.2 Fatalistic Responses: Admitting Modal Collapse

Brower observes that arguments against divine simplicity that are based on concerns about modal collapse strike against “the plausibility not the coherence of the doctrine,” for there is nothing inconsistent about saying that “God could not have known or willed otherwise than he actually knows or wills.”Footnote ²²¹ Indeed, various theologians of Muslim, Jewish, and (rarely) Christian traditions appear to have been willing to accept modal collapse. Some contemporary theologians and philosophers continue to do so.Footnote ²²²

But such a response immediately encounters stiff challenges. In addition to the obvious and very serious philosophical problems and the fact that modal collapse violates very basic and very common modal intuitions, we are also faced with damning theological problems. If modal collapse goes through and this world that is actual is the only possible world, then we are faced with the loss of divine sovereignty and aseity. For in such a case it is not so much as logically possible for God to exist without this creation and these creatures. Thus, as Craig puts it, the doctrine entails the denial of “God’s sovereignty over creation and his freedom to exist without creation.”Footnote ²²³ In other words, the doctrine entails the denial of the intuition identified by Plantinga as a key motivation for the doctrine: the “sovereignty-aseity intuition” – and thus what the doctrine is meant to safeguard and protect it now destroys. With the loss of divine freedom comes the complete loss of creaturely freedom – and with that freedom, moral responsibility. This leads to further theological problems; as Craig observes, this entails “disastrous consequences for Christian dogmatics with respect to the problem of evil, grace, God’s goodness, and so forth.”Footnote ²²⁴

6.3 Thomist Responses: Affirming Contingent and Extrinsic Predication

Many critics of modal collapse arguments observe that such arguments rely upon a key assumption – namely, that divine intellective and volitional actions are intrinsic to God rather than extrinsic and contingent.Footnote ²²⁵ Some defenders of the doctrine protest that this key assumption is unwarranted; Brower, for example, thinks that it is not only “plausible” to take these as extrinsic and contingent but also that this indeed is the way that Aquinas and other luminaries from the tradition understood matters.Footnote ²²⁶ But the views of Aquinas (and subsequent generations of Thomists) are controverted and branch off into several theories.

W. Matthews Grant and Mark Spencer employ the notion of “object essentialism” (OE) to help clarify matters. As they summarize the position, on (OE) “acts have their objects essentially: it is not possible for an act with a specified object to exist without that same object.”Footnote ²²⁷ Aquinas clearly and repeatedly (and quite forcefully) insists that God has freedom to do otherwise than God does; God is free to create one world rather than another or to refrain from creating altogether.Footnote ²²⁸ But his identification of divine essence with the divine act threatens such freedom – at least if Aquinas is taken to endorse (OE). As Grant and Spencer explain: “Suppose that God’s act of knowing that Francis is Pope during a particular time period is identical to God.” Given (OE), “to be this act of knowing is to have that particular object,” and thus it would not be possible “for that act of knowing to exist with a different object, and so not possible for that act to exist unless Francis actually is Pope during that time period.” But, if this is true, “it seems that it will not be possible to be God without being (among other things) the act of knowing that Francis is Pope” – and thus “it will not be possible for God to exist unless Francis is actually Pope during that period.”Footnote ²²⁹

Grant and Spencer distinguish between no fewer than five interpretative (and constructive) options in the Thomist tradition:

• Position 1: God’s acts have their objects essentially, and all of God’s acts are identical to God.

• Position 2: God’s acts have their objects essentially, and all of God’s immanent acts of knowing and willing are identical to God, while God’s transitive acts of creating, sustaining, and moving are not identical to God.

• Position 3: God’s acts have their objects essentially, and all of God’s acts are identical to God, except those that take actual creatures as their objects.

• Position 4: God’s acts do not have all their objects essentially, and all of God’s acts are identical to God.

• Position 5: God’s acts do not have all their objects essentially, and all of God’s immanent acts of knowing and willing are identical to God, while God’s transitive acts of creating, sustaining, and moving are not identical to God.Footnote ²³⁰

Although Position 2 is more modest than Position 1, it is hard to see how either of these strategies might offer refuge to the defender of simplicity who is seeking safety from the threat of modal collapse. As Grant and Spencer recognize, “these positions seem contrary to divine freedom.”Footnote ²³¹ For (tracking an objection from Suarez), they say that this seems to entail that God “could not exist without the actual world existing, and, if the actual world could fail to exist, then God could fail to exist.”Footnote ²³² The defenders of these views sometimes appeal to conceptual or rational distinctions to block the transitivity of identity; the payoff of which is meant to deliver the conclusion that “since God and the divine acts are conceptually distinct, it would not then follow from the divine acts’s being really identical with God that God couldn’t exist without these acts.”Footnote ²³³ But this strategy immediately encounters objections. Indeed, it seems desperate: Even sympathetic interlocutors such as Grant and Spencer admit that “revision of the relevant principle concerning the logic of identity is a bold move.”Footnote ²³⁴ The central problem echoes one that we saw with respect to the appeal to rational or conceptual distinctions when considering the doctrine of the Trinity – again, it is very hard to see how a merely conceptual distinction between “two” entities that are really identical can do the work needed to block the inference.

Positions 4 and 5 “emphasize that God is perfect in himself, apart from any “reference to creatures” and that “God gains nothing by creating.”Footnote ²³⁵ Accordingly, these positions are friendly toward not only divine aseity and transcendence but also divine benevolence and gratuity. But such positions still leave unexplained the nature of the distinction between the divine being and act. In any case, however, these positions commit the defender of simplicity to the affirmation of extrinsic properties. Even though Position 5 is significantly more modest than Position 4, nonetheless both such strategies involve commitments that are deeply counterintuitive. Indeed, to say that it is counterintuitive to reject (OE) is a massive understatement. It is wildly implausible to think that

• (A-N) Alvin’s act of writing that results in the production of The Nature of Necessity is identical to, or not really distinct from,

• (A-W) Alvin’s act of writing that results in the production of Warranted Christian Belief.

Granting the creator-creature distinction and all that, it nonetheless also seems highly implausible to think that

• (G-C) God’s act that results in the production of the actual world is identical to, or not really distinct from,

• (G-~C) God’s act that results in the production of no creation whatsoever.

Some critics view such strategies as nothing more than a last-ditch effort to purchase “bare coherence” for the doctrine – but at the “expense of desperate expedients.”Footnote ²³⁶ Even if they admit bare coherence, such critics think that these strategies are so far-fetched and counterintuitive that they demonstrate that the doctrine is “not plausibly true.”Footnote ²³⁷

Position 3 may be more promising. It requires no major revisions or rejections of classical logic. By maintaining (OE) it retains an intuitively motivated conviction, and by holding that divine acts are identical to God, it retains a central Thomist conviction. But, precisely by denying that God’s immanent acts of knowing and willing that apply to the actual creatures of the actual world are identical to God, it “avoids the objections altogether.”Footnote ²³⁸ So, because it denies that divine actions “intending creatures” are “identical to the divine substance,” there is no good reason to conclude that “the position threatens God’s freedom or omnipotence.”Footnote ²³⁹ This strategy leaves much unexplained, and it requires the defender to be “comfortable with a degree of mystery and incomprehensibility in one’s view of God.”Footnote ²⁴⁰ Some critics will likely judge that “degree of mystery” to be unbearably high. However, the strategy promises an account that is not obviously incoherent and that can satisfy the major desiderata. For as Pruss concludes, “once we recognize that a property might be intrinsic in the case of one entity and non-intrinsic in the case of another, we see that there is no immediate contradiction in a simple God having accidental properties such as willing that p or believing that p, as long as these are not intrinsic properties in the case of God.”Footnote ²⁴¹

6.4 Non-Thomist Responses: Looking for Alternatives

The foregoing responses are common among theists who are committed to broadly Anselmian and Thomistic versions of the doctrine. However, as we have seen, these are not the only options on the market. The Scotist doctrine has been under-explored in much of the contemporary discussion, but the proponents of it maintain that it allows the defender of divine simplicity to escape modal collapse.Footnote ²⁴² The Scotist can employ the formal distinction at this juncture, and by this strategy she might find the proverbial “goldilocks zone” between real distinctions, on the one hand, and merely rational or conceptual distinctions, on the other hand.

Or the defender of divine simplicity might go in a somewhat different direction. The proponent of simplicity might formulate a broadly “Suarezian” account; here she might posit a modal distinction between the divine essence and the divine act(s) of knowing and willing creation as something more than a merely conceptual distinction but less than a real distinction. Accordingly, we might think of this in terms of one-way separability; the divine act of creation cannot be separated from the divine essence, but the divine essence indeed is separable from the divine act of creation.

Orthodox Christian theologians typically take a different approach. Bradshaw judges Aquinas’s “view of divine simplicity” to be “inconsistent with his position on free choice.”Footnote ²⁴³ He thinks that “to suggest that creating the world makes no difference at all to the activity God performs is wildly counter-intuitive.”Footnote ²⁴⁴ Rejecting Thomism as internally inconsistent and unable to avoid the loss of divine freedom, he defends the traditional Orthodox commitment to the distinction between the divine essence and the divine energies.Footnote ²⁴⁵ In line with this broadly “Palamite” approach, one might seek refuge from the modal collapse objection in this distinction; the act of creation is an uncreated divine energy rather than the divine essence, and while such uncreated divine acts are, of course, consistent with the divine essence, they are not identical to it.Footnote ²⁴⁶

6.5 Conclusion

Considerations of the relation of divine simplicity to divine freedom – and, more specifically, to the problem of modal collapse – have been at the forefront of recent debates over the doctrine. And for good reason; there are serious concerns here that go to the heart of some important philosophical and theological issues. Surely these debates will continue, but at this point it seems safe to say the following. First, those broadly Thomist strategies that rely on merely conceptual or rational distinctions will continue to struggle, as it is far from apparent that they can block the inference to modal collapse. Second, it is clear that the broadly Thomist strategies that rely on the rejection of (OE) face stiff challenges with respect to plausibility. Third, there are versions of the doctrine that do not either obviously entail modal collapse or flaunt our very basic intuitions about (OE); in particular, broadly Palamite, Scotist, or “Suarezian” proposals might be further developed.Footnote ²⁴⁷

7.1 The Biblical Argument Revisited

We have already noted (in 2) that appeal can be made to Holy Scripture in both attacks on the traditional doctrine and in defense of it. Pruss appeals to the Johannine dictum that “God is love” (1 John 4:8, 16) as he argues for the doctrine. He takes this dictum to most plausibly entail that “God is identical with his attribute instance of love.”Footnote ²⁴⁹ Minimally, he says, this demonstrates that “there is no contradiction in God’s being identical with an attribute instance”; more positively, he also concludes that, “because there is some reason to think that all of a being’s attributes are metaphysically on par,” then “if one is identical with the being, so are they all.”Footnote ²⁵⁰ From this, Pruss mounts an argument that a denial of simplicity amounts to (IDOLATRY):

1. (1) It is permissible to center one’s life on God’s love;

2. (2) To center one’s life on anything other than God is idolatry;

3. (3) Idolatry is impermissible;

4. (4) If God is other than God’s love, then to center one’s life on God’s love is idolatry (from 2);

5. (5) If God is other than God’s love, then to center one’s life on God’s love is impermissible (from 3, 4);

6. (6) Therefore, God is God’s love (from 1, 5).Footnote ²⁵¹

The first three premises are supported not only by biblical theology but also deep intuitions held by all major Western monotheistic traditions. But from them, Pruss argues, we derive the conclusion that God is identical to God’s love. And, since what is true of God’s love is most plausibly true of God’s other property instances, then we should affirm divine simplicity.

7.2 The Argument from Divine Perfection

Some of the arguments are drawn from the affirmation of divine perfection. We may summarize a basic version of this argument as (PERFECTION):Footnote ²⁵²

1. (1) If God is perfect, then God cannot be composed of parts or entities that are either less than God or somehow actually equal to or greater than God or somehow potentially equal to or greater than God;

2. (2) God is perfect;

3. (3) Therefore, God cannot be composed of parts or entities that are either less than God or that are somehow actually equal to or greater than God or somehow potentially equal to or greater than God. Thus, God is simple.

The second premise enjoys ample support. Affirmations of this claim are found throughout many religious and theological traditions. And for good reason; (2) finds ample support in natural theology. Consider ontological arguments.Footnote ²⁵³ Both Anselmian ontological arguments and modal ontological arguments yield divine perfection.Footnote ²⁵⁴ In addition to arguments from natural theology, (2) arguably enjoys a basis in biblical theology; not only does Hebrew and Christian Scripture testify to the greatness and utter incomparability of God (e.g., Ps 48:1; Ps 89:6), it even comes very close to an outright affirmation that God is the greatest conceivable being: “Great is the Lord and most worthy of praise; his greatness no one can fathom” (Ps 145:3).

Support for the first premise, on the other hand, is less obvious and surely more controversial. Craig judges this premise to reflect the “Greek” assumption that “anything simpler is thereby more perfect, so that a being of maximum perfection must be absolutely simple.”Footnote ²⁵⁵ He says that “the argument is clearly question-begging without some reason for the assumption that to be simpler is to be better,” and he asserts instead that a being with a robust set of “superlative properties… is clearly greater than one lacking any such properties.”Footnote ²⁵⁶ Craig dismisses arguments from perfection as “remarkably weak” and deserving of “little comment.”Footnote ²⁵⁷ He makes a good point when he says that something further should be done to motivate the premise, but both Craig’s dismissal and his own assertion that complexity is “clearly greater” also seem hasty and in need of further support.

Is it so obvious that composition and/or complexity are better than simplicity? We are not thinking about perfection abstractly at this point; the salient issue concerns the perfection of unity. And surely some forms of unity are stronger and thus better than others; the unity of a pile composed of grains of sand is surely weaker than, say, the unity of an integrated human whose body and soul form a hylomorphic compound, and the human person composed of body and soul arguably enjoys a weaker unity than, say, that shared by the distinct but inseparable faculties of the soul.Footnote ²⁵⁸ When we are talking about divine perfection, we are talking about divine unity (although not only about that), and here it seems that stronger forms of unity are better candidates for the office of great-making property.

The intuitions behind (PERFECTION) may gain strength as we reflect further on divine aseity. The link between divine aseity and divine simplicity is crucial in many historic and contemporary discussions of the doctrine. Even such a critic as Plantinga recognizes the importance of aseity in these debates. Approaching the issue in terms of properties (rather than truthmakers), he says that “if God were distinct from such properties as wisdom, goodness, and power but nonetheless had these properties, then he would be dependent on them in a dual way.”Footnote ²⁵⁹ Plantinga explains the first way: If such properties are essential to God, then it is not possible that God exists without them; in all possible worlds that God exists, God has them. “He would have not existed if they had not.”Footnote ²⁶⁰ This yields the result that God “would be dependent upon something else for his existence” – and this “runs counter to aseity.”Footnote ²⁶¹ The second way, according to Plantinga, is this: “God would be dependent upon these properties for his character.”Footnote ²⁶² And this is all bad news for the theologian who believes in divine aseity. As Plantinga puts it, “the intuition – call it the sovereignty-aseity intuition – underlying the doctrine of divine simplicity must be taken with real seriousness.”Footnote ²⁶³ For if we “suppose God has essentially the property of being omnipotent and suppose that property is an object distinct from him, is uncreated by him and exists necessarily,” then in an important sense God indeed “does depend on that property.”Footnote ²⁶⁴ It is hard to take seriously the notion that anything with aseity depends upon anything for existence, fulfillment, or wholeness.

7.3 The Argument from Divine Ultimacy

Defenders of the doctrine have made several additional arguments in the neighborhood of aseity. Consider the argument from (ULTIMACY):

1. (1) If God is the First Being, then God is not and cannot be composed of, and thus posterior to, parts or other entities;

2. (2) God is the First Being;

3. (3) Therefore, God is not composed of, and thus posterior to, parts or other entities.

The idea here is not merely that God is first in a temporal sequence (although if God is temporally located, then this much would be true); it is that God has an ontological primacy or ultimacy. In other words, God does not come along “after” (either temporally or logically) something or someone else composes God from preexisting bits and pieces that have their own independent reality. Moreover, as Leftow puts it, “prior parts limit and determine” what something is and can do, and whatever is composed of those parts owes its existence to the parts.Footnote ²⁶⁵ Zarepour says that “the parts of a whole… have some sort of explanatory priority over the whole” because “these are the parts of a whole that make it what it is.”Footnote ²⁶⁶ This means that “the whole is at least partially dependent on every single part of itself,” and if “a whole is not simple, it would have at least two proper parts (i.e., parts that are not identical to the whole).”Footnote ²⁶⁷ This would mean that “the dependency of a whole on its proper parts is tantamount to dependency on things other than itself.”Footnote ²⁶⁸ But if so, then the whole is not – and cannot be – self-independent – and thus “not a necessary existent either.”Footnote ²⁶⁹ In Leftow’s expression, it is “better not to have other things limit and determine what one is and can be,” and thus it is “better not to have prior parts.”Footnote ²⁷⁰ For only in this way “can one qualify for the title ‘ultimate reality.’”Footnote ²⁷¹

7.4 The Independence Argument

A related argument at the heart of concerns about aseity is (INDEPENDENCE):

1. (1) If God exists a se and thus is not a dependent being, then God cannot be dependent upon parts or entities of which God is composed;

2. (2) God exists a se and thus is not a dependent being;

3. (3) Therefore, God cannot be dependent upon parts or entities of which God is composed.

Again, the second premise can safely be regarded as not only true but even obviously true (at least for the vast majority of theists of various traditions, and certainly including the major Christian traditions). It also seems hard to dispute the first premise, and the challenge is before those who would argue that it is false. We may grant that discussions of dependence can be complicated, and we may grant further that such complications introduce some ambiguity into the first premise. Such ambiguity may allow some wriggle room for the opponent of the doctrine. Even granting this much, however, the basic idea is both easy to see and intuitively powerful: As Leftow says, “where one’s very existence is at stake, it is better not to be at the mercy of others – to be independent and self-sufficient instead.”Footnote ²⁷²

7.5 The Causation Argument

Yet another argument closely connected to the doctrine of aseity is (CAUSATION):

1. (1) If God is composed of parts or other entities which are more fundamental than God is or equally fundamental with God, then God owes God’s existence to some external cause;

2. (2) God does not owe God’s existence to some external cause; to the contrary, God is the first efficient cause of all that exists that is not God;

3. (3) Therefore, God is not composed of parts or other entities which are more fundamental than God is or equally fundamental with God.

Leftow notes that while there are “weaker and stronger understandings of aseity,” it means at least this much: “existing without ever being caused to exist or to persist from without.”Footnote ²⁷³ Of course it is true that consideration of causation raises many vexed issues in metaphysics, both medieval and modern. But the basic point is not hard to see. Arguably, at least, composition would expose God to external causation in at least two ways: first, the discrete parts would be causes of God’s existence; and, second, something external to the parts and the divine whole would have to be the composer who puts the whole thing together. But, according to theism, God just is the uncaused cause of all things that are not God, and it is better to be this than it is to be anything else.

7.6 The Argument from Creation

Alexander Pruss makes a similar argument. As he lays it out, (CREATION) holds that:

1. (1) God is the cause of everything other than himself;

2. (2) Nothing that has proper parts is the cause of all of its proper parts;

3. (3) All of God’s proper parts are other than God;

4. (4) If God has proper parts, then he is the cause of all his proper parts (from 1, 3);

5. (5) Therefore, God has no proper parts (from 2, 4).Footnote ²⁷⁴

Note that (3) is demanded by the definition of proper parts and thus should not be controversial. (1), however, is more controversial, for it is rejected by theists who are Platonists. Such theists may take either of two routes; they may endorse

• (1*) God is the cause of everything other than himself that is contingent;

or

• (1**) God is the cause of everything other than himself that is concrete.Footnote ²⁷⁵

Pruss argues that (1*) is unsatisfactory for at least two reasons: first, “it could be that some Platonic entities are contingent,” and, second, it would be possible for someone to consistently hold to (1*) while also holding that “neither we nor anything else is created by God.”Footnote ²⁷⁶ The fortunes of (1**) look no better as a way out of the argument. For, plausibly, if God has parts, then all such parts are themselves concreta, and thus a restriction of (1) to (1**) “still yields an argument for parthood simplicity.”Footnote ²⁷⁷

7.7 The Argument from the Threat of Decomposition

Central to many of these discussions is a type of argument we can refer to as (DECOMPOSITION):

1. (1) If God is composed of various discrete parts or other elements equally fundamental or more fundamental than God, then God is subject to decomposition;

2. (2) God is not subject to decomposition;

3. (3) Therefore, God is not composed of various discrete parts or other elements equally fundamental or more fundamental than God.

The concern here should be obvious: Anything composed seems subject to decomposition; whatever is put together can, at least logically, come apart. According to the doctrine of divine aseity, this is not possible. The theological and existential problems should be obvious too – if God, who is supposed to be holding the universe together by his powerful word (Heb 1:3) could come unglued, then everything else is in big trouble. This argument is strengthened by further modal considerations. Importantly, God does not just happen to be non-decomposable; to the contrary, God is necessarily non-decomposable. This brings us to our next set of arguments.

7.8 Some Arguments from Divine Necessity

There is another family of arguments that is typically traced back to Avicenna but that also finds expression within Judaism and Christianity (a version of it is developed by Maimonides, and another is defended by Scotus). The Avicennian version relies upon a couple of important principles. The first is called the Composition Principle by Zarepour, who explains it as the thesis that every mereological sum is either necessary in itself or possible in itself but necessitated through another.Footnote ²⁷⁸ The second is referred to as the Decomposition Principle; this is the thesis that “every part of an existent is itself an existent,” which is either necessary in itself or possible in itself and necessitated by another.Footnote ²⁷⁹ Armed with these principles, Zarepour mounts a two-stage argument from divine necessity to monotheism and then to simplicity. The first stage can be summarized as (NECESSITY-1):Footnote ²⁸⁰

1. (1) There is a necessary existent. Call it “G.”

2. (2) Either the individuation of G is entirely due to its essence or it is at least partially due to accidents.

3. (3) If the individuation of G is entirely due to its essence, then every necessary existent (having the same essence as G) is identical to G.

4. (4) If the individuation of G is at least partially due to its accidents, then G must have accidental properties that are conferred upon it through something other than itself.

5. (5) If something has accidental properties that are conferred upon it through something other than itself, then that being is not a necessary existent (it is supposed to be self-independent in all respects).

6. (6) If the individuation of G is at least partially due to its accidents, then G is not a necessary existent (from 4 and 5).

7. (7) The individuation of G is not even partially due to its accidents (from 1 and 6).

8. (8) The individuation of G is entirely due to its essence (from 2 and 7).

9. (9) Therefore, every necessary existent is identical to G (from 3 and 5).

10. (10) Therefore, there is only one necessary existent.

So the first stage of the argument gives us the conclusion that there can be only one necessarily existent being with the divine essence: divine necessity entails monotheism. The second stage of the argument moves on to simplicity. Consider (NECESSITY-2):Footnote ²⁸¹

1. (1) There is a necessary existent G that has a proper part X (reductio assumption).

2. (2) Either X is necessary in itself or X is possible in itself (from 1 and the Decomposition Principle).

3. (3) If X is necessary in itself, then there is more than one necessary existent.

4. (4) There is only one necessary existent (from NECESSITY-1).

5. (5) X is not necessary in itself (from 3 and 4).

6. (6) X is possible in itself (from 2 and 5).

7. (7) Either X is essential to G or X is accidental to G.

8. (8) If X is essential to G, it could not fail to exist and would be necessary in itself.

9. (9) X is not essential to G (from 5 and 8).

10. (10) X is accidental to G (from 7 and 9).

11. (11) If X is accidental to G, it is necessitated through something other than G itself.

12. (12) If X is necessitated through something other than G itself, then G depends on that thing.

13. (13) G, as a necessary existent, does not depend on anything.

14. (14) X is not accidental (from 11 to 13).

15. (15) Contradiction: X is accidental to G, and X is not accidental to G (from 10 and 14).

16. (16) Therefore, there is no necessary existent that has a proper part (from 1 to 15).

17. (17) Therefore, the only necessary existent – that is, the Necessary Existent – is simple (from 4 and 16).

Zarepour’s arguments, which are clearly inspired by and indebted to Avicenna, require a distinct set of metaphysical commitments. Notably, the arguments rely heavily on the modal distinction between necessity and possibility. But the distinction between necessity and contingency, which is also vitally important, does not feature prominently in the argument. What is necessary is, of course, possible, but not all that is possible is necessary. What is contingent is possible but not necessary – and this latter notion is clarified in recent work on modality. Taking contingency with full seriousness, it is possible to make arguments that are similar to Zarepour’s Avicennian arguments but that allow access to insights from more recent and contemporary work in the metaphysics of modality.

Here is a very brief summary of the salient insights drawn from more recent work on modality.Footnote ²⁸² Let us begin by endorsing S5 as the best available account of modal logic.Footnote ²⁸³ We can take possible worlds to be either maximally consistent states of affairs or “maximally specific consistent ideas in the mind of God” (for present purposes, nothing hinges on the difference).Footnote ²⁸⁴ Accessibility relations between possible worlds are reflexive, symmetric, and transitive.Footnote ²⁸⁵ One and only one world is actual, and the only actually existing individuals are the individuals who exist in the actual world. While such individuals actually exist in only one world, they also exist in more than one world; in other words, they enjoy trans-world identity. These individuals have some of their properties essentially and other properties accidentally. An individual has a property essentially if and only if that individual has it in all the possible worlds in which that individual exists, while a property is possessed accidentally if the individual possesses it in some possible world(s) but not others. So some properties are possessed necessarily (and the proper understanding of necessity de re is essentially), while other properties are possessed contingently. Moreover, as Pruss and Rasmussen point out, “any plausible account of metaphysical necessity will yield the Distribution Axiom” which “basically states that necessity transfers across entailment: so, if p is necessary, and if p entails q, then q is necessary.”Footnote ²⁸⁶

Finally – and now somewhat less conventionally – we add what we can call the Contingency Transfer Axiom, which says that the modal status of a contingent entity that is an essential proper part of a whole is transferred to the whole. To see this, consider some entity W. W is a whole that is composed of proper parts. Suppose that W has some proper part P that is essential to W. Accordingly, W has P in all possible worlds in which W exists; it is not possible that W exists without it. Suppose further that P is contingent; P enjoys existence in only some possible worlds but does not exist in others. If P is contingent and only exists in some possible worlds but not others and is essential to W, then it cannot be the case that W exists in all possible worlds. In other words, P’s contingency is distributed to W – and so if P exists in some possible worlds but not others, then, because W can only travel as far as it can take P with it, W also exists only in those possible worlds where P puts in an appearance. Worlds in which P does not exist are worlds in which W cannot exist. If P exists contingently, then W cannot exist necessarily.

With this brief summary in hand, let us consider (MONOTHEISM):

1. (1) If God has some essential proper part P, then either P exists necessarily or P exists contingently.

2. (2) If P exists necessarily, then P exists as a necessary divine entity or being that is not numerically the same as God.

3. (3) If P exists contingently, then God is partly composed by some divine entity or being that is not numerically the same as God and that exists contingently.

4. (4) If God exists necessarily, then God cannot be composed of any essential parts that exist contingently.

5. (5) God exists necessarily.

6. (6) Therefore, God cannot be composed of a proper part that exists contingently (from 4 and 5).

7. (7) Therefore, if God has some essential proper part P, then P exists necessarily (from 1 and 6).

8. (8) Therefore, if God has some essential proper part P, then P exists as a necessary divine entity or being that is not numerically the same as God (from 2 and 7).

Note that (1) is a conceptual truth: There are only two modes of existence; for just anything that exists, that thing either exists necessarily or it exists contingently. (2) and (3) are very plausible given what we know about proper parts – if something is a proper part, then it is not numerically the same as the whole. (4) follows from the Contingency Transfer Axiom. If one accepts (5) (on the basis of, say, ontological arguments), then (6) and (7) follow logically. This brings us to (8), which leaves us with a God who exists necessarily – and which also leaves us with another necessarily existing divine entity. Plausibly, any necessarily existent proper parts would be instantiations of concreta – such parts are, that is, capable of causation (after all, in this case they make a causal contribution to God’s existence and essence).Footnote ²⁸⁷ But this looks suspiciously like polytheism. If something is divine, exists necessarily, and has causal powers but is also not God, then it looks like a really good candidate for the category “god.” Two Gods is obviously one God too many for a monotheist, and two independent deities (say, one God and one god, or one personal God and one impersonal deity) also gives us one deity too many for monotheism. The upshot should be clear: If God is composed of any parts that exist contingently, then God cannot be said to exist necessarily; but if God is composed of parts that exist necessarily, then God cannot exist without divine entities with causal powers that are distinct from God and that exist of necessity. If so, then arguably we have not only necessarily existent divine entities in addition to God but also a necessarily existent God who depends upon those additional necessarily existent entities. Again, as Zarepour puts it, “the parts of a whole… have some sort of explanatory priority over the whole” because “these are the parts of a whole that make it what it is.”Footnote ²⁸⁸ This means that “the whole is at least partially dependent on every single part of itself,” and if “a whole is not simple, it would have at least two proper parts (i.e., parts that are not identical to the whole).”Footnote ²⁸⁹ This would mean that “the dependency of a whole on its proper parts is tantamount to dependency on things other than itself.”Footnote ²⁹⁰ But if so, then the whole is not, and cannot be, self-independent – and thus is “not a necessary existent either.”Footnote ²⁹¹ And thus the monotheist who believes that God exists necessarily has good reasons to reject the notion that God has parts at all. Such a monotheist has good reason to believe in divine simplicity.

The foregoing argument assumes that necessary existence is possible for something that is composed of parts that exist necessarily. But it is less than obvious that such an assumption is safe. Lacking such an assumption, we can also consider an argument from (MODALITY):

1. (1) If God is composed of any proper parts (or indeed composed of any concreta that are not numerically the same as God), then God exists contingently;

2. (2) If God exists contingently, then God does not exist necessarily;

3. (3) Therefore, if God is composed of any proper parts (or indeed composed of any concreta that are not numerically the same as God), then God does not exist necessarily (from 1, 2);

4. (4) If it is possible that God exists necessarily, then God exists necessarily;

5. (5) If God exists necessarily, then God does not exist contingently;

6. (6) It is possible that God exists necessarily;

7. (7) Therefore, God exists necessarily (from 4, 6);

8. (8) Therefore, God does not exist contingently (from 5, 7);

9. (9) Therefore, God is not composed of proper parts (or indeed composed of any concreta that are not numerically the same as God) (from 1, 8).

Note that (2) and (5) are De Morgan equivalences and thus should be completely uncontroversial.Footnote ²⁹² (4) is perhaps somewhat less certain – but anyone who accepts S5 as the best available account of modal logic should be inclined to accept the premise. (6) is, of course, much more controversial; it is a major flashpoint of disagreements about ontological arguments. But when it comes to intramural disputes among theists (rather than, say, debates over the success or failure of ontological arguments as theistic proofs), surely there is widespread acceptance of at least the possibility that God exists necessarily. In other words, (6) should be at least plausible as a thesis of philosophical theology (whatever one makes of it as a piece of apologetics). And if we accept (6), then the conclusion follows logically and we have good reason to reject the notion that God has parts. Again, monotheists who believe that God exists necessarily have good reason to hold to the doctrine of divine simplicity.

In conclusion, it should be obvious that we have a plethora of arguments in favor of divine simplicity. Of course this is only a very brief sketch of the arguments, and each deserves much closer attention and analysis. Beyond these, there are others. For instance, Pruss says that the doctrine should be accepted by theists because “it is clearly strongly favored by Ockham’s Razor over other theories of the divine nature, precisely because it is ontologically (though maybe not conceptually) simpler.”Footnote ²⁹³ Any such arguments presuppose various metaphysical and theological commitments and have premises which can be questioned and debated. Thus, it is unlikely that any particular argument will be compelling to everyone. Nonetheless, it should also be clear that there are good reasons why religious believers might affirm not only monotheism but also divine simplicity.

Note, however, that these get us to (M-DDS) but not obviously all the way to (S-DDS); the arguments, if successful, rule out the possibility that God has parts – but they do not rule out the possibility that there are genuine distinctions within God. Such arguments also leave open for further analysis adjacent questions about how the proponent of divine simplicity should think about such distinctions: Are these distinctions to be understood as rational distinctions (perhaps with a fundamentum in re), as modal distinctions, as formal distinctions, or as something else?

7.9 Benefits of the Doctrine of Divine Simplicity

Much of the recent conversation and debate over the doctrine is concerned with criticisms; the detractors detract by making arguments against the doctrine, and the defenders defend by posing defeaters to the arguments. Such focus, however, should not distract us from consideration of the purported benefits of the doctrine. Traditionally, proponents of divine simplicity argue that it has theological payoff in several areas. First, the doctrine is said to have benefits in moral theology and metaethics. Stump notes that when thinking about the issue of God’s relation to morality, two main sorts of answers have tended to dominate the discussion. On one hand, “God will is sometimes taken to create morality in the sense that whatever God wills is good just because he wills it;” on the other hand, sometimes “morality is taken to be grounded on principles transmitted by God but independent of him, so that it a perfectly good God frames his will in accordance with those independent standards of goodness.”Footnote ²⁹⁴ As Stump points out, either sort of answer runs into difficulties. The first “constitutes a theological subjectivism in which, apparently, anything at all could be established as morally good by divine fiat.”Footnote ²⁹⁵ And thus it appears that we are left with a form of moral anti-realism according to which whoever has the most power simply gets to decide what counts as “good” and “evil.” The other option, however, “suggests no essential connection between God and the standards for evaluating actions.”Footnote ²⁹⁶ Moreover, it threatens “to impugn God’s sovereignty” by making God accountable to objective standards that are external to God and to which God simply finds himself accountable.Footnote ²⁹⁷ These options, as Stump sees matters, seem to either “be repugnant to common moral intuitions” or “presuppose moral standards apart from God, which God may promulgate but does not produce.”Footnote ²⁹⁸ Stump argues that the doctrine of divine simplicity “entails a third alternative”: because “God is simple, he is goodness; that is, the divine nature itself is perfect goodness.”Footnote ²⁹⁹ Accordingly, “there is an essential relationship between God and the standard by which he judges, the goodness for the sake of which and in accordance with which he acts, in accordance with which he wills only certain things to be morally good.”Footnote ³⁰⁰ This standard is nothing less than and nothing other than the divine nature; it is not an “arbitrary decision,” for “only things consonant with God’s nature could be morally good.”Footnote ³⁰¹ The moral standard is objective rather than subjective, and it is grounded in God’s own essential goodness.

Second, the doctrine of simplicity is important for the development of systematic theology. Sometimes it provides guardrails; at other points it offers resources for constructive proposals. Consider, as an example, the doctrine of the atonement in Christian theology. According to some articulations of the doctrine, the primary problem that must be addressed by the work of Christ is the “strife of attributes” that human sin has produced within God’s own life.Footnote ³⁰² On one hand, God’s mercy and love desire the salvation of sinners; on the other hand, divine justice and wrath demand retributive punishment. So the problem is “within God’s own being,” and the work of the incarnate Son is directed toward a repair of God’s own internal problem.Footnote ³⁰³ But if the doctrine of divine simplicity is true, then such theological speculation cannot even get off the ground. If God does not have parts – if God is not composed of various things called “divine attributes” – then it is not so much as possible that these attributes are opposed to one another. Thus, taking the doctrine of simplicity seriously guides us away from some formulations of the doctrine of the atonement.

Consider another illustration of how the doctrine of divine simplicity is relevant to the evaluation of contested options in Christian theology.Footnote ³⁰⁴ Reflection on the will of God has led some theologians in the tradition to employ a distinction between the “will of the sign” (voluntas signi) and the “will of good pleasure” (voluntas beneplaciti) in defense of distinctly Reformed doctrines of providence and predestination.Footnote ³⁰⁵ The first refers to the revealed will of God as it is signified in the commands and promises of God. According to this will, we learn that it is against God’s will to commit sins, and according to this will, we also learn that it is in accordance with God’s will that all sinners be saved (e.g., Ezek 33:11; 2 Pet 3:9). The second, however, refers to the secret or hidden will of God; as Muller explains, this refers to the “absolute, effective, and absolutely unsearchable will of God” that is the “decisive, deciding will of God.”Footnote ³⁰⁶ This latter category is especially important in the doctrine of predestination: This is the “hidden will of God to bestow special saving grace irresistibly upon the elect” that is “more ultimate than the revealed will of God to offer salvation to all by means of a universal grace.”Footnote ³⁰⁷ Such an employment of this distinction is resisted by theologians of various traditions. For instance, the Lutheran theologian Johann Gerhard quotes the Reformed divine Theodore Beza as admitting a contradiction: “that which lies hidden in God at times disagrees with what he reveals; therefore the will of the sign or revealed will contradicts the other secret will.”Footnote ³⁰⁸ Seeing this, Gerhard judges such Reformed theologians who are “architects of the absolute decree” to undermine divine simplicity: “Those who attribute contrary wills to God undermine the simplicity of the divine essence, for wherever there are contradictions of will, there is no room for the supreme and most perfect simplicity.”Footnote ³⁰⁹ In a similar vein, Suarez castigates the “heretics” who “refer all humans sins to the divine will, saying that God wills that we sin…”Footnote ³¹⁰ Suarez wields convictions drawn from the doctrine of simplicity against such claims: “The divine will is said to be the rule of its own actions not because it is guided by wisdom and prudence but because it itself essentially is its infinite wisdom and prudence.”Footnote ³¹¹ From closer to home within the Reformed tradition, Arminius argues similarly when he insists that “it cannot happen that God should at the same time will contradictory things.”Footnote ³¹² Thus, God “cannot prescribe what is unjust, because he is justice, wisdom, and omnipotence itself.”Footnote ³¹³

Third, the doctrine has important implications for pastoral theology and doxology. This sense is expressed well by the Reformed theologian Petrus van Mastricht: “God is most simple, and he thus gives simply (James 1:5), that is, he gives himself, all that he is, and all his attributes, which, by simplicity, are inseparable – his wisdom, power, goodness, and grace – devoting them to us…”Footnote ³¹⁴ For the divine simplicity is “a pure goodness that profits us, in which there is no malice to harm us,” as well as “a universal goodness, which allows no exception (Ps 34:9–10)” and “an unlimited and illimitable goodness, which cannot be restrained or impeded by any creature, no matter how powerful, which thus can be all things for us, and can supply all things to us, indeed, beyond what we ask or think in our mind (Eph 3:20).”Footnote ³¹⁵

Much more recently, the importance of the doctrine is obvious in Peter van Inwagen’s reflection on Julian of Norwich’s famous vision of the hazelnut. To recall, the medieval mystic says this:

van Inwagen observes that Julian’s sense of this hazelnut’s – and, crucially, her own – insignificance in relation to the vastness of the universe would only be expanded if she had been able to access the insights of modern astronomy. But the importance of her theological insight would not change at all. She “answers the question ‘Why does God pay attention to us?’” with a simple and direct response: God “pays attention to us because he loves us. And he loves us because he loves everything. And he loves everything because he is Love (1 John 4:8).”Footnote ³¹⁷ Urging us not to read “love” or “God” as abstractions, van Inwagen says that “God is Love as he is Goodness and Knowledge and Power: All these things are perfectly realized in him.”Footnote ³¹⁸ He draws a contrast between God and creatures: “Insofar as all are present in a rational created being, their presence in that being is a sort of imperfect (which is not to say flawed) copy of his goodness and knowledge and power.”Footnote ³¹⁹ Clarifying further, van Inwagen says that “we may equally well say that God is Goodness or that God is Knowledge or that God is Power”; indeed, “in a way that is difficult and perhaps impossible for us to comprehend in this present life, all these ‘God is’ statements are really the same statement.”Footnote ³²⁰ In contrast to human creatures whose virtues are “separable… because a human being is composed of parts,” God has no parts.Footnote ³²¹ van Inwagen takes the Johannine dictum that “God is love” as absolute theological bedrock, and it undergirds all our understandings of creaturely existence and significance: “God, being love, of necessity loves everything that he has made.”Footnote ³²² Reversing the order of Julian’s summary from economy back to ontology, we can say: God keeps it because God made it; God made it because God loves it; God loves it because God is love.

Scotus illustrates the importance of divine simplicity as an expression of piety and as an aid to the life of faith in this prayer: