The perfect being

In this section, I outline the place of modal panentheism in the current literature. Several articles have offered arguments for pan(en)theism from perfect-being theology (Harwood Reference Harwood1999; Lancaster-Thomas Reference Lancaster-Thomas2020; Nagasawa Reference Nagasawa and Kraay2014, Reference Nagasawa, Buckareff and Nagasawa2016, Reference Nagasawa2024; Symes Reference Symes2025). According to perfect-being theology, God is the being who possesses all great-making properties to the highest possible degree. On modal perfect-being panentheism, these great-making properties are said to include power, knowledge, goodness, consciousness, and encompassment.

Encompassment is the divine property that distinguishes modal panentheism from traditional theism. It is understood in terms of an entity’s scope – that is, how much of existence it occupies and for how long an entity encompasses that aspect of existence (Symes Reference Symes2025, 368). All other considerations being equal, one’s size and longevity are said to contribute to an entity’s overall greatness.

In determining which properties are intrinsically great-making, perfect-being theologians typically consult their intuitions through thought experiments. To that end, I draw a comparison between two trees: Sycamore-1 and Sycamore-2. Sycamore-1 stands fifty feet tall and has existed for two hundred years. In contrast, Sycamore-2 is five feet tall and has existed for two years. Setting aside other possible attributes – such as beauty, consciousness, or power – I take Sycamore-1 to be, in itself, the greater of the two entities.Footnote ¹

It may be argued that encompassment is only great in relation to other great-making properties. For example, one may claim that encompassment is insufficient for greatness: encompassment is only great if an entity is good (or, at least, not evil). This is an important point of clarification. Perfect-being theologians claim that God must hold all intrinsic great-making properties to the highest possible degree. Intrinsic great-making properties are those that are valuable in themselves, regardless of whether they are great in relation to some other end, wider value, or aspect of an entity’s character. I take encompassment to meet these criteria. This, after all, reflects our ordinary use of language. For example, The Great Fire of London was not is great in virtue of its size and longevity. At the same time, of course, the fire lacked overall perfection due to its destructive force. In the case of God, I accept that an all-encompassing being must also be a perfectly good being – not because all great-making properties must be good-making, but because God holds all great-making properties to the highest possible degree.

According to modal panentheism, the greatest encompassing being that can be conceived would be identical with the maximal set of concrete possible worlds, as it is intrinsically greater to be concrete than it is to be abstract (ibid., 371). This definition leaves open the possibility as to whether all worlds are, in fact, concrete. Some worlds, such as those that are incompatible with God’s perfect goodness, might exist as abstract entities. However, a substantial number of concrete worlds would be compatible with God’s goodness and, therefore, would be encompassed by a being who is maximally encompassing.

The upshot: a type of modal realism may be derived from the perfect-being thesis. If God exists, and if God is a maximally encompassing being, then we have a theistic motivation for believing in a substantial number of concrete possible worlds. Admittedly, this argument rests on the assumption that a substantial number of concrete worlds are possible. However, given the popular consensus – see Lewis (Reference Lewis1986), Kraay (Reference Kraay2010), Tegmark (Reference Tegmark2014), Almeida (Reference Almeida2017a/Reference Almeida2017b), Carroll (Reference Carroll2019), and Nagasawa (Reference Nagasawa2024) – I take it to be a sensible assumption that, if theistic modal realism were true, then the number of concrete possible worlds would be substantial.Footnote ²

The resulting view is a concept of God with a host of benefits. For example, Nagasawa believes that modal panentheism solves the traditional problem of evil, the problem of no best possible world, and the fine-tuning problem (Reference Nagasawa, Buckareff and Nagasawa2016, 96–100). Similarly, Michael Almeida has claimed that modal panentheism offers solutions to the paradox of sufficient reason, the less-than-best problem, and the problem of arbitrariness (Reference Almeida2017a, Reference Almeida2017b, 8). This list includes several reasons for favouring modal panentheism over other types of theism; I add to the list throughout this article.

Until recently, modal panentheism was neglected on account of the modal problem of evil (Nagasawa Reference Nagasawa and Kraay2014, Reference Nagasawa, Buckareff and Nagasawa2016, Reference Nagasawa2024). As we have seen, modal panentheism maintains that God would encompass every possible world. However, across the totality of possible worlds, says Nagasawa, there will be a significant number of concrete worlds that are incompatible with God’s perfect goodness. (I refer to evils that are incompatible with God’s perfect goodness as gratuitous evils and to worlds containing gratuitous evils as evil worlds.)Footnote ³ Nagasawa’s argument aims to show that modal panentheism is internally incoherent, as it posits a God who is both perfectly good and encompasses a substantial number of evil worlds. In contrast, while modal panentheism must address the existence of all possible evil, (non-modal) traditional theism needs only to reconcile God with actual evil.Footnote ⁴ Thus, Nagasawa maintains that traditional theism faces a far less formidable challenge and should, therefore, be preferred over modal panentheism. The argument can be stated as follows (Symes Reference Symes2025, 372).

1. 1a. According to modal panentheism, God is perfectly good and identical with all possible worlds. (Assumed for reductio)

2. 2a. All possible worlds exist and across all possible worlds there exists a significant (or infinite) amount of concrete evil that contradicts God’s perfect goodness. (Premise)

3. 3a. If God is identical with a significant (or infinite) amount of concrete evil that contradicts God’s perfect goodness, then it is not the case that God is perfectly good. (Premise)

4. 4a. God is identical with a significant (or infinite) amount of concrete evil that contradicts God’s perfect goodness. (1a–2a)

5. 5a. It is not the case that God is perfectly good. (3a–4a)

6. 6a. It is not the case that modal panentheism is true. (Conclusion)

Contrary to Nagasawa, modal panentheism can be defended against the modal problem of evil (ibid.). There are two ways of responding to the problem, both of which are adapted from arguments in the literature on traditional theism (Symes Reference Symes2025). The first response is the maximal-panentheism solution. The argument states that the perfect-being thesis need not be interpreted in line with the omni-God thesis, where God must possess all individual great-making properties to the highest possible degree. Instead, the greatest conceivable being may possess the maximum consistent set of great-making properties. Therefore, God ‘might or might not be omniscient, omnipotent and omnibenevolent’ (Nagasawa Reference Nagasawa2017, 92). The consequence of adopting this view, on maximal-panentheism, is that the modal problem of evil shows at least one of God’s attributes to be limited. For example, God’s goodness may be limited in order to maintain total encompassment; alternatively, God’s encompassment may be limited to worlds that are compatible with their perfect goodness (Symes Reference Symes2025, 373–375). In either case, the modal problem of evil does not constitute a reason for rejecting modal panentheism. In essence, this response rejects premise 1a and offers an alternative definition of modal panentheism – where God possesses the maximum combination of goodness and encompassment (1a*). By substituting this revised premise (1a*) for the original (1a), the modal problem of evil fails to undermine the truth of modal panentheism (ibid.).

The second response is the no evil worlds solution (ibid., 375–378). As previously defined, I take an evil world to be any world that contains some evil that is incompatible with God’s perfect goodness. Per modal panentheism, concrete possible worlds are derived from the intrinsic greatness of encompassment. (This is the origin of the modal problem of evil.) However, concrete possible worlds should not be derived solely from divine encompassment but – following the perfect-being thesis – in line with God’s wider attributes, which includes perfect goodness. In other words, any worlds that are deduced from the perfect-being thesis must reflect God’s perfect goodness; therefore, if God exists, then concrete evil worlds are not possible concrete worlds (and might only exist abstractly, if at all). Just as Richard Swinburne denies the logical possibility of divine sin (Reference Swinburne and Symes2024, 14) and Philip Goff rejects the metaphysical possibility of imperfect divine action (Reference Goff2020, 11), modal panentheists claim that God’s perfect goodness will guarantee the non-existence of concrete evil worlds (Symes Reference Symes2025, 375–378).

Given the proposed solutions, I suggest that the modal problem of evil is no less formidable than the traditional problem of evil and, therefore, modal panentheism ought not to be rejected in favour of traditional theism.

To date, all work on modal panentheism has focused on divine encompassment and goodness. Therefore, in the sections that follow, I offer the first general framework for understanding the wider attributes of God on modal panentheism. I argue that modal panentheism offers a coherent and compelling account of God’s mind, knowledge, and power that is not at a disadvantage – but, in fact, holds notable advantages – relative to other forms of theism.

The divine mind

Consciousness is widely regarded as an intrinsic great-making property (Pike Reference Pike1970, 136; Morris Reference Morris1991, 40; Hill Reference Hill and Symes2024, 4) – or a necessary consequence of another divine property, such as personhood or omniscience. All else being equal, the fact there is something it is like to be a given entity, contributes to an entity’s overall greatness. Now, in one sense, consciousness does not seem to qualify as a scaling property: a being is either conscious (it possesses the great-making property) or it is not (it does not possess the great-making property). In another sense, however, consciousness can be understood as scaling in terms of its scope: how much of an entity is conscious and for how long consciousness exists within an entity. Taking consciousness to be an intrinsic great-making scaling property, as I do here, consciousness that is vaster in scope and duration holds greater value than consciousness with lesser scope and duration. As modal space (God) must possess all great-making properties to the highest possible degree – and given that God is all-encompassing – consciousness (as a great-making property) must pervade modal space to the highest possible degree. For this reason, modal panentheists ought to adopt a theory of mind where consciousness is taken to be a pervasive feature of modal space.

A natural ally of modal panentheism is, therefore, a form of panentheistic cosmopsychism, where God’s mind is the fundamental, unified conscious subject that grounds the totality of modal space by subsumption.Footnote ⁵ On this view, God’s mind is the one fundamental entity, and everything else – each world, particle, and person – is grounded not as an independent entity, but as an aspect of the divine totality. That is, each element of modal space exists as an irreducible but ontologically dependent feature of God’s mind: nothing over and above the whole and irreducible to anything else within it. The inner nature of all physical matter, on this view, is consciousness – and that consciousness is God’s mind. I label this view modal cosmopsychism.Footnote ⁶

Modal cosmopsychism is a type of priority monism, which claims that there is only one fundamental entity, God, who is fully identical with the totality of modal space. On this view, modal space is a unified subject that exhibits all great-making properties to the highest possible degree: the mind of God constitutes the essence of every possible world, including the actual world.

Across modal space there will be countless smaller minds whose experiences are phenomenologically distinct. This raises important questions about the relationship between God’s mind and the minds of individual subjects, who have thoughts that are distinct and contradictory. I outline and address the problems associated with these two features of experience below.

The first problem is if and how distinct subjects of experience arise out of a larger, already unified field of consciousness. There are two aspects to this problem. The first is how smaller minds are separated from one another. It will be self-evident to some smaller subject (A) that they have direct and privileged access to their own thoughts. However, at the same time, the same subject (A) will not have direct and privileged access to the thoughts of some other smaller subject (B): their thoughts are distinct to them. This raises the following question: if both A and B are separate minds, then how can A and B be proper parts of the same larger mind, C?Footnote ⁷

This is a one version of the decombination problem, which, at its core, challenges cosmopsychists to explain how larger minds decombine to form smaller minds. I offer the following reply. I concede that the decombination problem counts as evidence against modal panentheism. However, this does not count against my central thesis, that – contrary to Nagasawa – modal panentheism should be rejected in favour of some other type of theism. Each branch of theism faces some version of the hard problems of consciousness and, importantly, any answer to this problem faces challenges. For instance, theistic dualism has long been criticised for the problem of causal interaction, and theistic physicalism remains vulnerable to a variety of conceivability arguments. I take these problems to be no less dangerous for their respective worldviews than the decombination problem is for modal cosmopsychism and, therefore, concede the rebuttal.

The second aspect of the distinctness problem is unique to pan(en)theistic conceptions of cosmopsychism: if and how the thoughts of smaller minds are shared by the larger, unified mind of God. Here, I answer positively: smaller subjects have direct access to their thoughts and, as their thoughts are part of the larger cosmos, the cosmos shares their thoughts. In other words, A and B are part of C and, therefore, everything possessed or experienced by A and B is also possessed or experienced by C. The view I propose entails a form of weak emergence, where no radically new properties appear when an individual subject arises in God’s minds: our natures are entirely reducible to their underlying properties. An alternative approach might deny that our thoughts are shared by the cosmos. However, if one were to claim that smaller subjects (A and B) are somehow separate from God (C) – say, they were not weakly emergent but strongly emergent – then smaller subjects would be something over and above the properties of God. However, as God is said to be all-encompassing, subjects being something over and above God gives rise to a further problem: an all-encompassing God who fails to encompass everything.Footnote ⁸ For this reason, I claim that the smaller minds (A and B) are accessible to God as proper parts of the larger divine mind (C).

The second difficulty for modal cosmopsychism is how the larger mind of God can possess contradictory beliefs. The view I have outlined maintains that the experiences of smaller subjects are fundamentally experiences of the larger universe and, in turn, larger modal reality. Otherwise put, the mind of modal reality (C) is a unified whole and all other entities – including smaller minds (A and B) – are merely aspects of that whole. The problem of contradictions, however, highlights the fact that the smaller minds (A and B) within the larger modal reality (C) will possess contradictory beliefs. For example, A thinks the earth in world-1 is flat, whereas B thinks the earth in world-1 is not-flat. As the thoughts of A and B are also held by C, this means that God (C) simultaneously possesses two contradictory thoughts (P and not-P). This is, what Miri Albahari calls, the perspective problem (Reference Albahari and Seager2019): the problem of how differing smaller perspectives can co-exist as part of a larger unified perspective.Footnote ⁹ The argument has the following structure.

1. 1b. According to modal panentheism, all perspectives are part of God’s perspective. (Premise)

2. 2b. The set of all perspectives includes contradictory perspectives. (Premise)

3. 3b. An omniscient being cannot possess any contradictory perspectives. (Premise)

4. 4b. God cannot possess the set of all perspectives. (1b–3b)

5. 5b. Modal panentheism is false. (Conclusion)

Before addressing the problem of contradictory perspectives, I begin by highlighting three unsatisfactory replies, each of which has precedent in the wider literature. The first reply claims that modal panentheists can side-step the problem of contradictory beliefs. According to this reply, modal panentheism endorses a type of cosmopsychism and, importantly, cosmopsychism does not maintain that the larger, fundamental mind is a rational agent who possesses no contradictory beliefs. This view – which has been defended by cosmopsychists such as Goff (Reference Goff2017, 243) – is not available to modal panentheists who understand the larger, fundamental mind to be the mind of God. Such a solution may be available to other pan(en)theists – such as Lancaster-Thomas, who claims that pan(en)theists are not required to accommodate God’s classical attributes (Reference Lancaster-Thomas and Symes2024, 168) – but not the proponent of perfect-being theology. The perfect being must possess all great-making properties (being rational and free of contradictory beliefs) and no bad-making properties (being irrational and containing contradictory beliefs). Therefore, as it is greater to be omniscient, rational, and hold no contradictory perspectives than it is to be non-omniscient, non-rational, and hold contradictory beliefs, modal panentheists cannot accept that God has contradictory perspectives or beliefs.

The second reply states that, although we cannot imagine a being who possesses contradictory perspectives, it is entirely possible that God would contain contradictory perspectives (Goff Reference Goff2017, 243). This response holds that the coherence of the greatest being transcends our understanding. In doing so, it appeals to a strong form of mysticism that hopes to resolve difficulties about God’s nature by placing them beyond our comprehension. For our purposes, however, this response is too hasty. Contrary to Ockham and Descartes, the consensus amongst philosophers of religion is that contradictions are not possible – not even for God. As I aim to incorporate mainstream understandings of the divine attributes into modal panentheism, I maintain that, rather than embracing contradictions, we should focus on explaining how such tensions might be resolved. While transparent alternatives remain on the table, I take recourse to mysticism to be unwarranted.

The third reply understands the larger mind of God to be a perspectiveless field of consciousness, within which exist smaller private perspectives. Proponents of this view claim that – in the absence of cognisensory content – there exists a pure, content-less ocean of consciousness that constitutes the essence of the universe (Kastrup/Schopenhauer Reference Kastrup2020; Albahari and Symes Reference Albahari and Symes2022). Importing this approach into modal panentheism, the ocean of perspectiveless consciousness would pervade the totality of modal space. Here, the hope is that smaller minds (A and B) rely on the existence of a larger mind (C), but – importantly – C does not share the experiences of A and B and, therefore, C does not experience contradictory thoughts. The problem with this approach is that it faces a dilemma. First, if A and B are nothing over and above C, then C must share in their perspectives and, therefore, according to the perspectives problem – assuming that contradictory perspectives cannot co-exist within a single perspective – C remains impossible. Second, if A and B are something over and above the properties of C – namely, content-possessing perspectives – then there is something that is not encompassed by God: content-possessing perspectives (A and B). Given that each side of the dilemma renders the larger mind either impossible or not all-encompassing, the claim that the larger mind is content-less fails to overcome the problem of how an all-encompassing God can possess contradictory perspectives.

In place of the above replies, I offer the following solution to the problem of contradictory perspectives. At the heart of the solution is the following claim: the larger mind of God is a subject of experience (C) whose contents is not exhausted by the unified minds of other subjects (A and B). While there are aspects of God (such as A and B) that have competing perspectives, there is a larger mind of God (C) that is able to harmonise the competing perspectives they contain. God has a further, broader perspective that transcends the minds of other subjects (A and B), which is able to rationalise otherwise competing perspectives. The crucial point is that God does not contain contradictory beliefs in a naïve way but experiences them within a larger evaluative framework – situating each in the broader context of their source, salience, and epistemic role within the divine mind – which prevents contradiction at the level of divine rationality. To differentiate appearances, then, is to attribute each experience to a distinct first-person perspective, without collapsing or confusing them at the level of divine self-awareness.

Consider, for example, the following analogy. Suppose that a human experiences the sensation of being pricked by pins and needles, say, in their right leg. From the perspective of their right leg, it feels as if they’re being pricked by hundreds of sharp objects. However, there is another part of the agent who understands their experience to be, in a sense, illusory. The larger agent is able to harmonise their competing perspectives – the sensation that they’re being pricked by pins and needles, and the lack of pins and needles in their visual field. To cash out the analogy, God can understand their competing sensory experiences (the sensation of needles, and the lack of needles in their visual field) with their cognitive understanding (knowledge of how the sensation comes about) in virtue of their wider knowledge and rationality. Ultimately, God’s collective mind contains a range of illusory experiences and beliefs – as agents typically do – but is able to sort between them in virtue of their wider perspective.Footnote ¹⁰

Before proceeding, I offer two points of clarification on the divine attribute of consciousness. The first clarification concerns modal possibility. Modal panentheism posits a substantial number of concrete possible worlds that collectively reflect the divine attributes. However, critics might object that this imposes a significant restriction on modal possibility, since, on modal panentheism, worlds without consciousness are not possible worlds. The objection states that, contrary to modal panentheism, non-conscious worlds seem to be conceivable and, therefore, assuming modal rationalism, possible. I offer the following reply. When it comes to restricting modal space, one must accept that modal panentheism imposes significantly greater limits on possible worlds than traditional versions of modal realism. If some argument for modal realism and non-conscious worlds were successful, then this would constitute an argument against divine consciousness (God) encompassing every possible world. The problem, however, is that such an argument would need to be more persuasive than the argument against non-conscious worlds from perfect-being theology. To undermine modal panentheism, one must present an argument for their possibility of non-conscious worlds that outweighs the considerations provided by perfect-being theology for the impossibility of non-conscious worlds. Here, the impasse is similar to those pertaining to discussions on traditional theism. For example, the ontological argument claims that one cannot conceive of a world without a necessary being (God), whereas the reverse ontological argument claims that one can conceive of a world without a necessary being (without God) – see Findlay (Reference Findlay1948) and Grayling (Reference Grayling2014, 90). One of my central claims is that modal panentheism is not at a significant disadvantage relative to traditional theism; accordingly, I acknowledge that if a conscious God fails to exist in all possible worlds, this will undermine both modal panentheism and standard forms of traditional theism.

To recap, the argument for modal perfect-being panentheism runs as follows. By definition, God must possess all great-making properties to the highest possible degree. Consciousness and encompassment are great-making properties. Thus, if it is reasonable to believe that God exists, it is also reasonable to believe that consciousness pervades all possible worlds – that is, all possible worlds must be conscious worlds. If belief in the existence of concrete possible worlds is grounded in the perfect-being thesis, then non-conscious worlds are not possible, as they involve contradictions that are a priori identifiable: they are both the God of modal panentheism (conscious) and not the God of modal panentheism (not-conscious). Per modal panentheism, all worlds are fundamentally conscious and free from gratuitous evil and, therefore, modal panentheism is significantly more restrictive on possibility than traditional modal realism. This is a welcomed consequence of modal panentheism: if the God of modal panentheism exists, then all worlds must participate in God’s divine nature.

The second point of clarification concerns the unity of God’s mind. On Lewisian modal realism, all possible worlds are isolated worlds – that is to say, there is no spatial, temporal, or causal connection between them. This raises a problem for modal panentheism: if each world is isolated from all other worlds, then how can God possess a unified mind encompassing the totality of possible worlds? This problem, the problem of divine unity, suggests that modal panentheism implies a form of polytheism; after all, if each world is isolated but divine, then modal space contains a substantial number of gods. To borrow from Lewis, this would make a modal panentheist ‘the most extreme polytheist going’ (Reference Lewis1983, xi).

There are two types of polytheism that might follow. First, possible worlds will differ in the amount and types of divine attributes they exhibit (power, knowledge, goodness, and consciousness) and, therefore, some worlds may be more deserving of the label God (the greatest being) than others. A single world could reflect the divine attributes more than other worlds and, consequently, a single world will be more deserving of the label God than any other world. In this case, all of the worlds that exhibit great-making properties to a lesser degree will be lesser gods. Second, there may be several worlds with equal levels of maximum power, knowledge, goodness, and consciousness. Accordingly, many worlds will reflect the perfect-being thesis and, therefore, share the title of God. This type of polytheism is close to the polytheism of John Leslie, who has ‘no trouble with the idea of infinitely many infinite minds, or with calling each of them “divine”’ (Reference Leslie, Buckareff and Nagasawa2014, 61). Leslie asks, ‘Why in particular would any such mind, if benevolent and able (as theists so often insist) to create absolutely anything that was logically possible, produce anything except more minds, infinitely many minds, of the same type?’ (ibid., 57)

I argue that, for the following reason, both forms of polytheism are unwelcome. God’s nature on modal panentheism is the consequence of the perfect-being thesis, which posits a substantial number of possible worlds as part of God’s nature. As it is far greater to encompass all possible worlds than a single world, we have reason to favour modal monopanentheism over modal polypanentheism.

It may be that modal realism does not secure the divine unity required for maximal greatness. However – and this is an important point of clarification – modal panentheists need not rely on Lewisian modal realism. If God’s properties cannot be unified across the isolated worlds of Lewisian modal realism, then perfect-being panentheists may adopt an alternative metaphysical framework – some other form of modal realism or multiverse theory – that can accommodate divine unity.

Let us take multiverse theories, for example. There are numerous theories of the multiverse that involve a substantial number of universes that are not isolated from our actual universe (see Carroll Reference Carroll2019; Nagasawa Reference Nagasawa2024; Tegmark Reference Tegmark2014).Footnote ¹¹ The broadest of these approaches is Nagasawa’s multiverse, which is said to include all logical and metaphysical possibilities (Reference Nagasawa2024, 76). Central to perfect-being panentheism is the claim that worlds or universes ought to be posited due to the greatness of encompassment. Therefore, even if Nagasawa’s (or some other multiverse) lacked scientific support, the perfect-being thesis constitutes a theistic motivation for believing in a maximally large multiverse (or set of worlds), which may exhibit the unity necessary for unified divine consciousness. Modal panentheists, I maintain, should be prepared to incorporate one of these views in place of Lewisian modal realism. In the case of Nagasawa’s multiverse, as Nagasawa points out himself (Reference Nagasawa2024, 76–77), the switch is a straightforward one: modal panentheists turned multiverse panentheists may substitute worlds for universes and use the term actual to refer to the universe (rather than the world) that we occupy.

The defence goes further: even if no multiverse model allows for divine unity, then modal panentheists can opt for a non-Lewisian approach to modal realism in which possible worlds are not isolated worlds. The reason is as follows: modal panentheists posit worlds that are additional to our own – not because other philosophers or physicists believe in them but – because such worlds are a necessary consequence of the perfect-being thesis, which should be defined in accordance with God’s wider attributes (including consciousness). Therefore, I maintain that modal (or multiverse) panentheists ought to preserve the unity of the divine mind by, if required, adopting a metaphysical model distinct from Lewisian modal realism. All that is required is that the chosen theory includes a substantial number of concrete worlds or universes. The modal in modal panentheism is only a label: the view’s fundamental and necessary commitment is to a substantial number of concrete worlds or universes that are identical with God.

Divine knowledge

The belief that knowledge is intrinsically great-making is also held by the majority of perfect-being theologians. All else being equal, the amount and type of knowledge a being possesses – that is, how many things one knows and the importance of the things they know – contributes to their overall greatness. As with the other divine attributes, an omniscient being is defined as that which knows everything that is logically and metaphysically possible to know. Simply put, impossible knowledge is not knowledge that can be lacked, not even by God.

There are two ways in which the God of modal panentheism may be said to possess omniscience. The first (the deep-knowledge approach) aligns with traditional theism; second (the part–whole approach) offers notable advantages over traditional theism. The God of modal panentheism can possess both kinds of knowledge (deep and part–whole), whereas the God of traditional theism simpliciter cannot possess part–whole knowledge, therefore, lacks the proposed advantages.

First, the perfect being must essentially possess maximum knowledge and, therefore, the mind of God may possess knowledge that is not shared by smaller subjects. In other words, a property of the wider mind of God (C) may include all knowable propositions without the smaller subjects within God (A and B) possessing such knowledge. I call this the deep-knowledge approach. The deep-knowledge approach to God’s knowledge is on par with God’s knowledge on traditional theism, as both concepts of God hold a complete knowledge of all propositional statements.

The second way of understanding God’s knowledge, on modal panentheism, is through the part–whole approach. On this view, God shares all of the knowledge that is held by the individual parts of God. Across modal space everything that is possible to know will be known by some part of God; therefore, as knowledge held by the smaller minds of God (A and B) are shared by the larger mind of God (C), God knows everything that it is possible to know. In terms of propositional knowledge, modal panentheism holds no advantages over traditional theism (or vice-versa). However, in terms of experiential and indexical knowledge, part–whole modal panentheism holds notable advantages over traditional theism.

Experiential knowledge is direct awareness of a phenomenon, event, or state of being – this is knowledge of what it is like to undergo an experience. Whereas propositional knowledge concerns facts that can be expressed in third-person terms, experiential knowledge, refers to first-personal awareness – the subjective character of experience. There will be experiential knowledge of what it’s like to, for example, see a duck or surf a wave, which can only be known to an entity who has experienced such things. It is difficult to make sense of a timeless, spaceless God knowing what it’s like to undergo certain experiences – running from a duck, falling off a surfboard – from the propositional facts alone. However, across the totality of possible worlds, modal panentheism holds that all experiential perspectives – no matter how alien or familiar – are aspects of God’s knowledge. All possible subjective experiences (of A and B) are shared by the divine totality (C). In contrast, traditional theism faces the challenge of accounting for experiential knowledge, which requires positing mechanisms for God to access such knowledge – such as omnisubjectivity, illusory, or imaginative experience (see Buckareff Reference Buckareff2023; Zagzebski Reference Zagzebski and Kvanvig2013) – or accepting that experiential knowledge is unavailable to God. To its credit, modal panentheism avoids these problems entirely.

Modal panentheism is also able to accommodate complete indexical knowledge. Indexical knowledge refers to knowledge tied to a specific perspective or context, such as knowing My name is Jessica, and it is currently 18:30. This type of knowledge captures a type of context-dependent knowledge that depends on the knower believing something about their place in space, time, or some other state of affairs. Under modal panentheism, since God is identical with everything that is logically and metaphysically possible, every I, here, and now (A and B’s indexical knowledge) is shared by God (C). In regards to traditional theism, several arguments have attempted to show that indexical knowledge is unavailable to God (Coburn Reference Coburn1963; Kretzmann Reference Kretzmann1966; Helm Reference Helm1988): a timeless, spaceless being could not know My name is Jessica, and that it is currently 18:30. Again, modal panentheism avoids these arguments as all indexical knowledge is part of God’s knowledge.

Although modal panentheism holds some advantages over traditional theism, the view faces two unique problems: the problem of contradictory indexicals and the problem of evil knowledge. I outline and address each in turn.

The problem of contradictory indexicals is as follows: one unified divine mind cannot hold contradictory indexical claims to be true. For example, one subject cannot believe the following indexical statements in conjunction at the same point in time: I am Jessica but I am not John and I am John but I am not Jessica.

1. 1c. According to modal panentheism, all indexical beliefs are God’s beliefs. (Premise)

2. 2c. The set of all indexical beliefs includes contradictory indexical beliefs. (Premise)

3. 3c. An omniscient being cannot possess any contradictory indexical beliefs. (Premise)

4. 4c. God cannot possess the set of all indexical beliefs. (1c–3c)

5. 5c. Modal panentheism is false. (Conclusion)

The problem of contradictory indexicals (1c–5c) is notably similar to the problem of contradictory perspectives (1b–5b). In fact, I offer a parallel solution. Although A and B will possess relational, context-dependent beliefs (I am Jessica and I am John), the larger mind (C) will possess a broader knowledge and perspective that allows God to rationalise the numerous indexical beliefs. The larger divine mind will contain perspectives (I am Jessica) to which their context-dependent beliefs (indexical claims) are true. However, the divine mind will also have access to a wider range of knowledge and perspectives that allow them to contextualise the different aspects of their being. For example, the divine mind will know, One part of me is Jessica and another part of me is John.

Multiple indexical beliefs can co-exist in one divine mind, so long as the divine mind can understand these beliefs within a wider context. There are helpful analogies to be drawn with pathological cases, in which subjects report feelings of alienation from their thoughts, indexical or otherwise. For example, in cases of depersonalisation, subjects feel a lack of phenomenological ownership of their thoughts: they feel detached from their thoughts as if they were listening to an audio-broadcast or watching a film. From the perspective of Jessica, it will be true that I am Jessica; however, from God’s perspective, there is feeling that this thought is owned by someone else. Similarly, in cases of schizophrenia, subjects report experiencing the voices of other agents; these voices can present contradictory beliefs and perspectives that co-exist within a subject’s mind. Patients may report indexical claims to themselves or others – I am Jessica and I am John – that are true in the context of those individual perspectives. From the perspective of the larger subject (God), they know that their various perspectives are part of a wider state of affairs.

The critic might argue that other indexical claims violate logical consistency. For example, it cannot be true for God that I am only Jessica. Such a statement can only be true for Jessica. I offer the following reply. Such statements are mistaken: they reflect beliefs, but they do not reflect knowledge. The point is not that the divine mind affirms I am Jessica as a universal truth, but that God recognises this to be true within a particular perspective – namely, that part of themself. In this sense, egoism is fundamentally an illusion. On modal panentheism, we are all fundamentally part of one person: the greatest being.Footnote ¹²

The second problem for part–whole omniscience is the problem of evil knowledge. The problem begins by acknowledging that all knowledge, across all possible worlds, is said to constitute a proper part of a perfectly good God (part–whole knowledge). However, within the total set of knowledge exists knowledge of evil state of affairs, such as what it’s like to commit evil acts (experiential knowledge) and to be the one who commits evil acts (indexical knowledge). For example, on modal panentheism, God holds beliefs such as I am Jessica, all of my motivations are evil, and I am currently torturing an innocent person. Such knowledge, says our critic, are incompatible with God’s perfect goodness. I present the argument below.

1. 1d. According to modal panentheism, God is perfectly good and knowledge across all worlds is possessed by God. (Premise)

2. 2d. If some experiential or indexical knowledge is incompatible with God’s perfect goodness, then modal panentheism cannot be true. (Premise)

3. 3d. Some experiential or indexical knowledge requires a first-person perspective of moral wrongdoing. (Premise)

4. 4d. A perfectly good being cannot possess the first-person perspective of moral wrongdoing. (Premise)

5. 5d. Some experiential or indexical knowledge is incompatible with God’s perfect goodness. (3d–4d)

6. 6d. Modal panentheism is false. (Conclusion)

I offer the following reply to the problem of evil knowledge (1d–6d). Modal panentheists ought to claim that there is no knowledge of evil-doing that is gratuitous – namely, there is no knowledge that is incompatible with God’s goodness. In doing so, panentheists may adapt traditional theistic theodicies (and defences) in order to show (or show to be possible) that all instances of evil knowledge serve some greater good. Simply put, modal panentheists ought to claim that all lower-order evil knowledge serves some higher-order good. For example, consider a character-development defence where an agent is made greater in virtue of the evils that befall them. According to this defence, knowledge of what it’s like to harm somebody and what it’s like to be harmed by somebody allow for a smaller subject to grow and achieve a number of goods, such as the goods of seeking redemption, being forgiven, and being the forgiver. In such cases, indexical and experiential knowledge of evildoing is necessary to achieving such higher-order ends.

Our critic might reply that the problem of evil knowledge is not a problem of evil knowledge being gratuitous. Rather, this is a problem of God’s intimate relationship with evil. Whether or not evil knowledge can be cashed out in terms of some good, the fact of the matter – says our opponent – is that God is partly identical with a person (A) performing some evil action, and that this same person (A) possesses the knowledge that they are not performing such an action in pursuit of some greater good. I offer the following reply. In so far as God is experiencing what it is like to be the person torturing for fun, or possessing the indexical knowledge that God is the one torturing for fun, theodicies and defences do not solve the problem in this way. But the objection is based on a misunderstanding. Let us say that a person (A) has indexical and experiential knowledge of torturing a person (B) for fun. It does not follow that the knowledge of the larger mind (C) is exhausted by the knowledge of A. Part of C’s knowledge will be A’s knowledge, but there will be aspects of C’s knowledge that contextualises A’s knowledge. C knows that parts of them (A and B) will be ignorant of the goods their evil-knowledge will produce, but C – in virtue of their broader knowledge – is able to understand that the knowledge of A and B is necessary to that greater good.

I end this section with an example. Suppose that one part of a person’s body (the hand) causes pain to another part of their body (the leg), say, by injecting some needle. The hand sees itself as doing some wrong; similarly, the leg sees itself as the victim of some wrongdoing. However, from the perspective of the brain, the state of affairs is positive – they are administering a vaccination. Like the body, God has the ability to harmonise the competing experiences and perspectives of its parts. From the perspective of God, the unified mind that pervades modal space, every indexical or experiential evil serves a greater purpose.

Divine power

The consensus amongst perfect-being theologians is that power is a great-making property – that is, all else being equal, the intuitive value judgement is that power contributes to the intrinsic greatness of the possessor.Footnote ¹³ As with the other divine attributes, my purpose here is to understand this within the context of modal panentheism. After developing such an account, I overview the benefits of modal panentheistic power over other forms of theism, before proposing (and resolving) a final objection.

The conventional definition of power is that it corresponds to the number and variety of states of affairs that a being is able to bring about – that is, a more powerful being is one that is capable of bringing about a greater number or variety of states of affairs, while a less powerful being can bring about fewer. Most philosophers of religion also accept view that, post-Aquinas – and contrary to Ockham and Descartes – God lacks nothing in being unable to bring about impossible states of affairs. With these qualifications in mind, let us take omnipotence to include every power that is logically and metaphysically possible to possess.

As we have seen, modal panentheism takes God to be identical with the totality of possible worlds, which includes the existence of every possible state of affairs. As I have argued, at the heart of all modal space is a unified divine mind (modal cosmopsychism). Individual minds (A and B) exist within this larger, unified mind of God (C) and, therefore, A’s and B’s mind and knowledge are also a part of C. As subjects (such as A and B) are nothing over and above parts of God (C), all subjects’ powers are also shared by God. This means that across the totality of possible worlds, every logically and metaphysically possible power that a smaller subject can exercise is also exercised by God (this is another part–whole constitution). Furthermore, all natural phenomena, such as waves and thunderbolts, are also nothing over and above God. Importantly, as such entities are nothing over and above parts of God, all of their powers are fundamentally God’s powers. The resulting view is that every power is possessed by God and shared by the distinct entities (persons, plants, and protons) that are understood to be proper parts of God. As the totality of worlds (God) includes every possible state of affairs, God can be said to possess every power that is logically or metaphysically possible. For these reasons, I take the God of modal panentheism to be omnipotent.

Modal panentheistic omnipotence holds several advantages over other types of theism. Here, I draw three comparisons. First, modal panentheism is preferable to pantheisms that restrict God’s power to the actual world. For example, on standard pantheisms, God is said to be identical with the natural world, such that all of God’s powers are expressed through the unfolding of natural laws within the actual world. In contrast, modal panentheism does not restrict God’s powers to the powers within the actual world. Across modal space, there are worlds where God’s power involves ‘totally different laws govern the doings of alien particles with alien properties’ (Lewis Reference Lewis1986, 2). For pantheisms that restrict worldly powers to the actual laws of nature, such possibilities are excluded, which renders such conceptions of God more restrictive. Given that God, as traditionally conceived, is thought to hold all great-making properties to the highest possible degree, modal panentheism should be preferred over standard pantheisms as it allows for a more expansive understanding of divine power.

Second, divine power on modal panentheism is preferable to non-panentheistic versions of perfect-being theism. Consider, for example, the non-panentheistic maximal-God theism of Nagasawa (Reference Nagasawa2017). According to Nagasawa, the God of traditional theism possesses the maximum consistent set of great-making qualities. This means that – in the light of some evidence, such as the existence of evil – one might restrict one or more of God’s traditional attributes. Consider, for example, Nagasawa’s analysis of the problem of evil: ‘In response to it, perfect being theists can simply maintain, by appealing to the perfect being thesis and the maximal God thesis, that if the argument is cogent at all, it teaches us only that God, as the being than which no greater is metaphysically possible, is not a being that is simultaneously omniscient, omnipotent, and omnipotent’ (Reference Nagasawa2017, 101). Now, suppose that the traditional maximal-God theist limits God’s power to account for the existence of actual evil. (Note that a similar approach is taken by Philip Goff (Reference Goff2023) and John Stuart Mill (Reference Mill and Matz1874/2009), who are also open to limiting God’s power on account of the problem of evil.) In such cases, for non-modal theisms, our understanding of divine power is shaped by the specific contents of the actual world. In contrast, modal panentheism holds that God encompasses all possible worlds – not just the actual one, which includes the many worlds where evil fails to exist. As a result, God’s power is not limited by the presence of evil in our world: God exercises the power to bring about worlds without evil. This is not to deny that modal panentheists may adopt a Nagasawa-style response to the modal problem of evil, by restricting God’s power or goodness. Modal panentheists can still endorse the maximal-panentheism view, which allows for trade-offs between divine attributes. The key point here is that modal panentheism permits a conception of divine power that extends beyond the constraints of non-modal theisms, which restrict God’s power to what is realised in the actual world.

Third, the exercising of God’s power is greater on modal panentheism than it is on non-modal forms of traditional theism. According to non-modal traditional theism, God possesses every power that it is logically and metaphysically possible to possess; however, God only utilises a small subset of such powers. In contrast, in virtue of encompassing every possible world, the God of modal panentheism successfully exercises every possible power.Footnote ¹⁴ It is important to note that both concepts of God (the God of traditional theism and the God of modal panentheism) possess the same amount of power – that is to say, they are both able to meet the same number of input conditions. However, the limited use of divine powers on non-modal traditional theism may fail to realise some other valuable end, such as the manifestation of God’s benevolence through the maximisation of goodness and divine creativity. According to the principle of plenitude, the best state of affairs includes the existence of maximal variety and value. Therefore, as a substantial number of good and unique worlds are possible, God ought to express them in virtue of their perfect goodness (Almeida Reference Almeida2017a; Kraay Reference Kraay2010). On non-modal traditional theism, God’s decision to create one world raises an important question: why would God create one valuable world, if God can create a substantial number of valuable worlds? This is a problem that modal panentheism avoids, as worlds of every variety and possible value exist as part of God.

I turn now to a potential problem with panentheistic omnipotence: the problem of evil powers. Per modal panentheism, all powers are constituted by God and, therefore, the exercising of any power is inherently an exercising of God’s power. Since powers have been exercised toward evil, it follows that God’s powers have been exercised toward evil. However, it would be paradoxical if God’s powers were exercised towards evil, as a perfectly good being could never use their powers for evil. Unlike the traditional problem of evil for orthodox theism – which asks why God might allow evil to occur – the problem of evil powers is more direct. The problem is as follows: why does God directly exercise their powers to bring about evil states of affairs? The problem of evil powers is unique to pan(en)theisms that claim that all of powers – including those of smaller subjects and the wider world – are shared by God and, therefore, the argument may constitute a reason for favouring traditional theism over modal panentheism. The problem of evil powers has the following structure.

1. 1e. According to modal panentheism, God is perfectly good and all powers exercised in all worlds are God’s powers. (Premise)

2. 2e. Some powers within some worlds (including the actual world) have been exercised toward evil. (Premise)

3. 3e. Some of God’s powers have been exercised toward evil. (1e–2e)

4. 4e. If some of God’s powers have been exercised toward evil, then God cannot be perfectly good. (Premise)

5. 5e. God cannot be perfectly good. (3e–4e)

6. 6e. Modal panentheism is false. (Conclusion)

I offer two replies to the problem of evil powers (1e–6e). The first reply is a rejection of the claim that all powers exercised within the actual world are God’s powers (1e). One might reject the premise by claiming that modal space includes libertarian free wills who are not fully determined by God’s powers. Here, choices are made by agents whose parts are constituted by God, but their choices – in keeping with a traditional libertarian model of freedom – are made independently of prior causes or powers. If smaller subjects possess the free will to exercise their powers towards evil, so says the reply, then the exercising of powers towards evil is not God’s responsibility but the responsibility of smaller agents.Footnote ¹⁵ The problem with this approach is that it takes libertarian free wills to be something over and above God’s fundamental nature – that is to say, some other entity (but not God) possesses some power, which means that all powers are not, contrary to our understanding of modal panentheism, God’s powers. One solution might involve individual wills strongly emerging from the world’s fundamental, divine nature. However, consequently, God would fail to be all-encompassing and, in the case of power, would mean that God does not possess every possible power. For this reason, I opt for the second reply.

The second reply rejects the claim that some powers within some worlds have been exercised toward evils (2e) that are incompatible with God’s perfect goodness (4e). Granted, if everything – including all powers – are nothing over and above God, then panentheists must accept that some of God’s powers (which include all powers) have been exercised toward evil. However, modal panentheists can still maintain that none of God’s powers have been exercised toward gratuitous evil. In other words, every power that is exercised by God fulfils some higher-order end, by necessity of God’s perfect goodness. It might seem to a distinct subject within God (A or B) that their powers have been exercised toward some gratuitous evil. However, God knows from the perspective of their wider mind (C), that all powers are ultimately exercised towards higher-order goods. These goods may include moral-developmental goods, such as repentance, reconciliation, or the capacity for forgiveness. Such goods are impossible in worlds without wrongdoing. To recycle our earlier example, we can make use of a character-development defence in which the different parts of God are made greater in virtue of their previous wrongdoing – the power to self-harm allows God to grow and achieve the good of self-forgiveness. The benefit of this approach is that it maintains God’s total encompassment and perfect goodness. It is worth pointing out that this approach removes the possibility of – what we often take to be – independent free will. After all, if all powers are God’s powers, then all of our actions are fundamentally God’s actions.

In sum, each response to the problem of evil powers comes at a cost: the first preserves creaturely freedom at the expense of total encompassment, while the second safeguards divine unity by dispensing with libertarian free will. For those who regard egoism as illusory, the second approach is the more appealing. Crucially, both responses succeed in preserving divine perfect goodness and resolving the challenge posed by the problem of evil powers.