A central feature of illness is diminished agency (see Gipps, 2022, p. 15).¹ This is true of both somatic and mental illness in general, and of depression in particular. Thus, in the DSM-5, the most recent psychiatric diagnostic manual, symptoms of major depressive disorder include fatigue, insomnia or hypersomnia, and diminished concentration and decisiveness (APA, 2013, pp. 160–161). The centrality of diminished agency alongside distress and suffering to depression is further underscored by the fact that it is necessary for diagnosis of major depressive disorder that “The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning” (APA, 2013, p. 161) (I take it that impairment of functioning is one kind of diminished agency).

The centrality of diminished agency to depression can also be found in first-hand descriptions of depression, which often report the person having diminished agency due to feeling a sense of the impossibility of action. In accounts by people who suffer with depression they often note that “Everything requires massive effort and I’m not really able to do anything”; “Things seem almost impossible” (cited Ratcliffe, 2015, pp. 168, 169). Matthew Ratcliffe, whose work on the phenomenology of depression articulates the experience of depression drawing on a wide range of different first-hand accounts, notes that:

People with depression often report an impaired ability to act. This not only affects actions that involve forethought or effort. Even habitual and undemanding activities, like making a cup of tea or having a shower, can seem overwhelmingly difficult and beyond one’s abilities. (Ratcliffe, 2015, p. 155).

However, not everyone who has depression seems to report a loss of agency in the sense of an impaired ability to act. For example, David Hilfiker writes that his symptoms have always been unusual, not only in terms of having “no thoughts of suicide” but also in terms of having “only occasional losses of energy,” allowing him to be successful in his studies and medical practice (Hilfiker, 2002). Nevertheless, I think that unusual cases of depression like Hilfiker’s usually exhibit diminished agency in other ways. Hilfiker reports experiencing diminished relational agency, in the sense that he reports feeling an “almost constant emotional distance” from his wife and children and describes his depression as interfering with his emotional relationship with others (Hilfiker, 2002). In Hilfiker’s words:

I feel an almost constant emotional distance from others: from Marja, from my children, from my friends, and from God. My daughter, now twenty-seven, recalls a childhood Christmas when she presented me a handmade gift. I said I liked it, and I said I was grateful, but even at age eight she knew I was faking it. That would have been typical for me: hindered from the positive emotions of the moment, emotionally blocked from the love and togetherness offered me by others. (Hilfiker, 2002)

To be unable to relate to others is a form of diminished agency because (as Hilfiker’s case highlights) the person cannot do something that they wish to do and that is also a normal – indeed, objectively desirable – human thing to do (namely, to respond emotionally to loved ones) (see Timpe, 2022, p. 159).²

Diminished relational agency, defined in this way, is extremely common in depression. In first-hand accounts of depression, people will frequently report (for example): “I couldn’t feel anything for [my husband]. I couldn’t feel anything for my children” (Anon, cited Ratcliffe, 2015, p. 218). Ratcliffe notes that “A common theme in almost every first-person account is the felt loss of interpersonal connection” (Ratcliffe, 2015, p. 218). Likewise, David Karp writes that:

Each person’s tale of depression inevitably speaks to questions of isolation, withdrawal, and lack of connection. The pain of depression arises in part because of separation from others; from an inability to connect, even as one desperately yearns for such a connection. (Karp, 1996, pp. 26–27)

So far, I have shown that diminished agency, including diminished relational agency, is a central characteristic of depression. Depression is generally regarded as a mental illness or disorder, a form of psychopathology. The term “mental illness” highlights that there is something undesirable about depression for the person who has it – depression is an absence or loss of human flourishing and so a privation of the good.³

A distinctive feature of mental illnesses is that the symptoms of mental illness are typically identical with the illness’ constituents (to be depressed is to experience the symptoms of depression). Therefore, the symptoms (or constituents) of depression – such as diminished relational agency – are privative in humans. Conversely, then, in humans at least, relational agency is a good.

I now want to turn to the question of whether there are any implications of what I have argued for a conception of God. So far, I have argued that diminished relational agency, such as we find in depression, is a privation of the good in humans, and that relational agency is a good in humans. In the process of doing this I have touched on two kinds of agency. One kind of agency involves the ability to act (evinced, for example, in the ability to make a cup of tea or have a shower). The second kind of agency involves “relational agency,” that is, the ability to respond emotionally to others.

These two conceptions of agency relate in different ways to two very different and competing conceptions of God. According to the first conception of God, God is completely self-sufficient: nothing can affect or influence God outside of God. This sounds very close to traditional notions of autonomy. On this conception of God, God is completely self-governing and able to act, unconstrained by any external agents or influences.⁴ Concomitantly, nothing can “move” God and so God does not have emotions that are responsive to other people.⁵ This idea is called “impassibilism.” According to impassibilists, God cannot have emotions that are responsive to other people since to have emotions that are responsive to other people is to be susceptible to forces outside of oneself, and so not genuinely to be self-sufficient. If I am the sort of being who has emotions that are responsive to other people, then others can hurt me, make me angry, make me happy, and so on, and so I cannot be totally self-determined or self-governed.

According to impassibilists, then, God has an absolute form of the first kind of agency, since God can do whatever God wills without influence or coercion by others (no one and nothing could stand in the way of God if God wished to create a universe or to make a cup of tea). However, God does not have relational agency, since God does not have emotions that are responsive to others. God wills the good of those God loves, but without the emotions that are part of or typically go with love in non-depressed human experience. As impassibilist James Dolezal explains:

In God nothing is made or caused to be. This means that whatever the character of God’s relationship to the world, it is not one in which the creature produces anything in God – no knowledge, pain, or pleasure. The righteous man or woman gives him nothing, and the sinner does nothing to him. (Dolezal, 2023, p. 252)

Impassibilists will often point to St Thomas Aquinas’ creation theology for support: if God is the uncaused cause, then (according to this line of reasoning) nothing can affect or influence God: “God […] is solely agent, and in no way does any passion have a place in Him” (Aquinas Summa Contra Gentiles I c. 89).

According to a different conception of God, God is emotionally responsive to others. On this second conception of God, God has a significant degree of autonomy, traditionally construed (God has the freedom to do a number of things, unimpeded by outside influences) and yet God also has relational agency. God’s relational agency means that God has less traditional autonomy in one sense: God can be affected by God’s creatures so that God’s creatures can cause God to feel joy or sorrow. Notably for the first impassibilist conception of God, being the sort of being who has emotions is a susceptibility, whereas for the second conception of God it is an ability: not to have the capacity for emotions is a loss or a lack. This second position is often called “passibilism.”

Does the fact that (as I have argued) diminished agency, including diminished relational agency, is a privation in humans give us a reason to affirm the passibilist conception of God over the impassibilist one?

Those who answer affirmatively typically adopt a methodology in philosophical theology called “perfect being theology.” According to perfect being theology, God is (definitionally) the most perfect being or, in St Anselm of Canterbury’s words, “God is whatever it is better to be than not to be” (Anselm Proslogion 5). A modern way of putting this is to say that God must possess every “great-making property” (Rogers, 2000, p. 12). In order to know (to a limited extent) what God is like, in addition to drawing on biblical revelation, we can, according to perfect being theologians, use our reason – more specifically, we can come to know what God is like by reflecting on what a perfect being would be like, or what properties are great-making. Thus, God is thought to be omniscient because a being who is all-knowing is more perfect than a being who is partially-knowing or knowledge-less. Likewise, God is thought to be omnipotent because a being who is all-powerful is better (more perfect, more great) than a being who is only partially-powerful or impotent. And so on.⁶

If diminished relational agency is a privation in humans, and relational agency is a good, then we might think that a perfect being such as God must have relational agency (“X”). As far as I know no one has expressed an argument for passibilism in precisely this way. However, people have made analogous arguments. For example, Ryan Mullins suggests a structurally similar argument, but with respect to psychopathy (instead of depression) and empathy (instead of relational agency) (Mullins, 2022, p. 238).⁷ In addition to structurally similar arguments, X is sometimes expressed as an intuition that motivates passibilism (see e.g. Webb, 2017, discussed below).

There are significant but inconclusive arguments for and against passibilism and impassibilism. It is beyond the scope of this paper to consider all of them. Rather, I want to consider whether X, which seems intuitively plausible, can withstand significant objections against it. If it can, then it seems that X provides a further (non-sufficient and non-conclusive) argument for passibilism alongside other arguments for passibilism.

In order to test X, I will consider three objections to it.

One objection to X is that impassibilists could make an analogous argument to X but for impassibilism. Let’s call the parallel argument “Y.” According to Y, suffering is a central characteristic of depression, and a privation in humans. Therefore an absence of suffering is a good in humans, and so an absence of suffering is appropriate to a perfect being such as God. The idea that God does not suffer is typically something that impassibilists affirm and passibilists deny, since suffering is an aspect of emotional responsiveness to others. The passibilist proponent of X, therefore, needs to show why, if her argument works, the opposite, seemingly analogous, argument does not work, since the conclusion of X and of Y are mutually exclusive.

To respond to Objection 1, I suggest that there is a disanalogy between diminished relational agency (on the one hand) and suffering (on the other). This disanalogy, I suggest, lies in the fact that, prior to the eschaton, suffering can be entailed by love. Because of this, while suffering is in itself a privation, in some cases suffering is a lesser privation⁸ than not suffering (since, in the situations I have in mind, not suffering entails not loving either). An example of this kind of situation would be equanimity or apathy in the face of a person who is hungry and homeless, where the appropriate response is compassion, which is a kind of suffering out of love. Suffering will not be entailed by love at the eschaton since the circumstances that give rise to suffering entailing love (such as homelessness and hunger) will no longer exist. Suffering is entailed by love prior to the eschaton if the beloved is herself suffering (perhaps because she is hungry and homeless) or because she is not flourishing in some other way.

Suffering, then, can be entailed by love, and so can be a lesser privation than not suffering. The same does not seem to go for diminished relational agency: relational agency does not seem to be in tension with love, either now or eschatologically. X and Y both begin with what is a good in humans to infer what must be a good in God, but, while proponents of Y are correct that suffering is a privation, there are situations in which not suffering is a greater privation. The same does not go for diminished relational agency, and so X and Y are disanalogous.

X starts from a human and so creaturely good, and then moves to what we might think is a good in God. Someone might object that this approach leads to absurd consequences. For example, they might point out that tails are goods in haddock. Therefore, they might ask, should we say that God has a tail – perhaps a giant or infinite one? What this reductio ad absurdum highlights, they might say, is that perfections are kind-relative, and so human perfections don’t tell us anything about what perfections are appropriate to God.

In response to this, I want to modify X from referring to “goods” in humans to referring to “spiritual goods.” By a “spiritual good,” I mean that something is an aspect of spiritual well-being. Thomas Aquinas says we have four duties of love (STh IIii, q. 25, a. 12). First and foremost, we have a duty of love to God, and, second, to ourselves. Third, we have a duty of love to our neighbour. Finally, we must love our bodily life – for example, to ensure that we are safe, fed, at a reasonable temperature and so on. I think the first three of these could be described as spiritual goods (and so aspects of spiritual well-being), since they concern our relationship with God and questions about whether we live our lives virtuously and with integrity. The fourth, our commitment to our bodily life, includes goods such as “Tasia the human has a nutritious diet” and “Perpetua the haddock has a tail,” and, while good, these do not seem to be spiritual goods or aspects of spiritual well-being. We might expect there to be a closer relationship between spiritual goods in humans (and potentially other creatures) and the divine life than other aspects of creaturely good (and the divine life) because spiritual well-being is the participation of creatures in the divine life, fully realised at the beatific vision but realisable to some extent in the here and now. Given this, I want to modify X in the following way: X2:

Relational agency (defined as an ability to relate emotionally to others) is a spiritual good in humans, and this gives us a reason to affirm passibilism (that is, to attribute relational agency to God).

I now want to consider a third, nearby objection: namely, that the perfect being theology being utilised in support of X and X2 will result in the projection of human ideals on to God, and thus result in an idolatrous anthropomorphism (see Trakakis, 2010a; 2010b).⁹ In other words, if we start from what we think a perfect being is like – which, by virtue of our humanity, will usually be about what a perfect human is like – then we will almost certainly end up conceptualising God as a kind of super-human. This is essentially a form of idolatry, since we end up worshipping a God made in our own image, rather than being responsive to a God who is radically other.¹⁰

To this, the proponent of X2 might point out that, while perfect being theology is still in the background, X2 does not start from any old human good to infer what God is like, but from something that is also a spiritual good or an aspect of spiritual well-being. As already noted, spiritual well-being is properly regarded as a creaturely participation in the divine life. and we have some additional reason for thinking that, by looking at distinctively spiritual goods we might be able to see a reflection of the life of God.

To this, the impassibilist might reply: human spiritual well-being is correctly seen as a participation in the divine life, but this participation in the divine life is experienced by humans in a distinctively human and creaturely way. Accordingly, relational agency is indeed entailed by spiritual well-being in humans, since, for humans, emotional responsiveness is properly a part of love (and she who loves lives in God). However, according to the impassibilist, emotional responsiveness is not properly a part of love in the case of God, and so God can love without the relational agency that accompanies love in humans.

The passibilist might reply that it is impossible to make sense of God’s love while holding God to lack relational agency. However, what the impassibilist’s point highlights is that moving from X to X2 is not sufficient to overcome the impassibilist’s concerns about passibilism being anthropomorphic. The passibilist needs to find a different way of responding to this concern.

Here the passibilist can reply that the impassibilist is just as likely to project human ideals and so anthropomorphise about God as the passibilist. One person who suggests this is the psychologist and passibilist theologian Marcia Webb, who seeks to dispel problematic theologies of mental illness (Webb, 2017). In the course of her book, Webb points to the ways in which psychological “research indicates that we tend to view ourselves and God in similar ways” (Webb, 2017, p. 146). Webb argues that our understandings of God and our self-images are in continuous interaction with one another, such that our self-image affects our conception of God, and our conception of God in turn affects our self-image (2017, pp. 146–148). Webb then turns to the theological cases of Docetism and impassibilism as historical and contemporary examples of the way that “distorted notions of God may be related to distorted notions of the self” (2017, p. 147).

Webb begins by endorsing Joseph Ratzinger’s reflection that the Docetic rejection of Christ’s suffering was congenial to Stoic thought (Ratzinger, Behold the Pierced One, 57; cited in Webb, p. 147). She then goes on to explain that:

This ‘congeniality’ was due to Stoicism’s rejection of all passion in human experience […] Perhaps not surprisingly, then, alongside the ancient Greek world that espoused a passionless ideal of humanity, the Docetic embraced a God who was removed from the messy miseries of this world. (Webb, 2017, p. 147)

Docetism, then, resulted from a negative evaluation of the emotions. Webb makes a parallel argument with respect to impassibilism today, connecting it with negative theologies of mental illness (such as the idea that mental illness is the result of a lack of faith) that, in her view, are also the result of a suspicion of emotion and the vulnerability that accompanies it. In her words:

In the same way that Docetism may have reflected the Stoic ideal for humanity, an emotionally restrained Savior – and an entirely impassible God – may be the preferred correlates to negative […] theologies which portray Christian life as absent of all psychological distress. Such a depiction of God is more conducive to the ideal of human invulnerability found in these theologies. (Webb, 2017, p. 147)

According to Webb, then, problematic conceptions of mental illness are to some extent the result of a suspicion of emotion and the ideal of a perfect human as emotionless, invulnerable and self-sufficient, and this human ideal is reflected in impassibilist notions of God.

Webb’s argument suggests that the impassibilist is in no better position than the passibilist when it comes to anthropomorphism: while the passibilist may be projecting her ideal of relationality onto God, the impassibilist may equally be projecting her ideal of invulnerability.

I would add to Webb’s case that the impassibilist’s anthropomorphic projections might be less obvious to us because we live in a patriarchal culture and because, as some feminist philosophers and theologians have highlighted, invulnerability, independence and self-sufficiency are patriarchal or masculinist ideals (see e.g. Mackenzie & Stoljar, 2000, p. 9).

If the psychological studies that Webb cites are correct that there is a symbiotic relationship between a person’s conception of God and the image of their ideal self, then it seems likely that this will apply to the impassibilist as much as to the passibilist. And if we live in a patriarchal culture, that this is happening with respect to an impassibilist conception of God may be less obvious to us than when it happens in relation to a passibilist conception of God. (A person who has lived only in the desert doesn’t notice more desert, but only when the landscape instead becomes verdant and green.)

Regarding Objection 3, it seems that the passibilist and the impassibilist are in a similar position, even if the impassibilist is less conspicuously in that position. Here, then, the passibilist and impassibilist are in something of an impasse.

At this point, an impassibilist of a certain and distinctive kind might respond that there is in fact an asymmetry between passibilism and impassibilism that has not been taken into account in all of this. Passibility (according to this kind of impassibilist) affirms a positive attribute of God. Impassibility (on the other hand) simply denies that positive attribute; in this impassibilism is essentially an apophatic enterprise that simply rules out certain (anthropomorphic) attributes from the divine nature precisely to maintain God’s radical otherness. I will call this kind of impassibilist an “apophatic impassibilist,” because their focus is on “impassibility” as the negation of an attribute (i.e. of passibility). Perhaps the passibilist could counter that surely (according to the law of excluded middle) God will either have emotions or not have emotions; there is no third option. However, according to the apophatic impassibilist, this would be true of creatures but is a category error in the case of God. This is because, according to the apophatic impassibilist, the passibilist conception of God as a person who has emotions and is relational and vulnerable does not have its impassibilist counterpart in a conception of God as a person who is cold and emotionless and separate from others. Rather, it denies the first without affirming the second, precisely by denying that God is a person at all (see Feser, 2023). In this respect, the apophatic impassibilist is like a mathematician denying that the number three is blue: it is true that the number three is not blue, but that is not because the number three is pink or yellow instead.¹¹ Or, as apophatic impassibilist Herbert McCabe puts this point, to deny that God is a Celtic supporter is not to affirm that God supports Rangers instead (McCabe, 1987, p. 41).

Apophatic impassibilism seems like an important corrective, both to some critiques of impassibilism, which can start from the assumption that impassibilists think that God is a person and then conclude that impassibilists must think that God is a cold and emotionless person, and also to some formulations of impassibilism by impassibilists themselves. However, I think the passibilist can push back on the claim that her theology necessarily involves projection while the impassibilist’s does not. In order to do this, I want to consider the relationship between cataphatic and apophatic language of God proposed by Pseudo-Dionysius the Areopagite (henceforth “Denise”).¹² According to Denise, “there is no contradiction between the affirmations and the negations, in as much as he [God] infinitely precedes all conceptions of privation, being beyond all positive and negative distinctions” (Mystical Theology, 1.2). As I read Denise here, she is not saying that we describe God more accurately when we use apophatic rather than cataphatic language of God, but rather, that while we use both cataphatic and apophatic language of God, ultimately God transcends both.

C.S. Lewis, who knew the works of Denise, makes a similar point though he describes it in terms of anthropomorphisms and abstractions rather than cataphatic and apophatic language. In his words:

This talk of ‘meeting’ [God] is, no doubt, anthropomorphic; as if God and I could be face to face, like two fellow-creatures, when in reality He is above me and within me and below me and all about me. That is why it must be balanced by all manner of metaphysical and theological abstractions. But never, here or anywhere else, let us think that while anthropomorphic images are a concession to our weakness, the abstractions are the literal truth. Both are equally concessions; each singly misleading, and the two together mutually corrective. Unless you sit to it very lightly, continually murmuring ‘Not thus, not thus, neither is this Thou’, the abstraction is fatal. It will make the life of lives inanimate and the love of loves impersonal. (Lewis, 2020, p. 26)

In a later part of the same work, Lewis makes the link between abstraction and negative (or apophatic) language more explicit and relates this specifically to passibilism and impassibilism. Of the Bible’s representation of God, which often speaks of God having emotions, Lewis says:

We are constantly represented as exciting the Divine wrath or pity - even as ‘grieving’ God. I know this language is analogical. But when we say that, we must not smuggle in the idea that we can throw the analogy away and, as it were, get in behind it to a purely literal truth.¹³ All we can really substitute for the analogical expression is some theological abstraction. And the abstraction's value is almost entirely negative. It warns us against drawing absurd consequences from the analogical expression by prosaic extrapolations. By itself, the abstraction ‘impassible’ can get us nowhere. It might even suggest something far more misleading than the most naïf Old Testament picture of a stormily emotional Jehovah. Either something inert, or something which was ‘Pure Act’ in such a sense that it could take no account of events within the universe it had created. (Lewis, 2020, p. 69)

As I read it, Lewis follows Denise in regarding not only cataphatic language (such as the language of passibilism) but also apophatic language (such as the language of impassibilism) as inadequate.¹⁴ Both may be necessary, and both act as a corrective of the other, but God transcends both in the sense that neither succeed in pinning down the matter of the ‘quality’ of God’s life more than the other.

If there can be an apophatic impassibilist (such as I have described above), I think there can also be a Denisian passibilist (of which Lewis may be one).¹⁵ According to the Denisian passibilist, there may be some things we want to say about God (relating to relational agency) without claiming to provide a thick description of God’s emotional life. The difference between the apophatic impassibilist and the Denisian passibilist lies in this: the apophatic impassibilist still denies that God has relational agency (albeit in the same way that the mathematician denies that the number three is blue). In contrast, the Denisian passibilist regards “the number three is blue” and “God has relational agency” as disanalogous. This is because, while they regard the question “is the number three blue?” an odd one, at the end of the day the right answer is that the number three is not blue. This makes it a bad analogy for “God has relational agency,” according to the Denisian passibilist, since she wants to affirm (in an analogous and so literal sense) “God has relational agency” but append “and yet” to that clause. In saying “and yet” the Denisian passibilism points to impassibilist language as a corrective to the anthropomorphising tendencies of passibilist language but the buck does not stop with impassibilism: because the two are mutually corrective, there is an infinite dialectic between them.¹⁶ Perhaps a helpful image of this idea is of a rope made of two (equally-sized) strands, with one strand representing apophatic, impassibilist language and the other strand representing cataphatic, passibilist language. If one focuses on one part of the rope, one might see more of the apophatic or else of the cataphatic strand, but each of these is followed by the other strand, which is in turn followed by the other strand, and so on. Both parts of the rope are necessary and neither is sufficient (to be a rope) in itself.¹⁷

How does the idea of Denisian passibilism relate to the idea that language of God is analogical?¹⁸ The Denisian passibilist agrees with the passibilist who wants to say that speaking of God’s emotions and capacity for emotional responsiveness is analogical – that is, that it is literally true but that there are differences as well as similarities in the way God is passible to the way in which (some) creatures are passible (i.e. to the way they have emotions and the capacity for emotional responsiveness). In fact, Denisian passibilism can be seen as an extended reflection on analogical predication. However – and quite compatibly – the Denisian passibilist thinks that impassibility can also be analogically predicated of God: there exist things in the created world that are impassible, and God is analogically like these – both in that there are similarities and differences. In the case of both passibilism and impassibilism, for the Denisian passibilist, the dissimilarities will always be greater than the similarities. To appropriate the words of the Fourth Lateran Council, “Between Creator and creature there can be noted no similarity so great that a greater dissimilarity cannot be seen between them.” For the Denisian passibilist, the language of impassibility is there to remind the person of the differences between divine and creaturely emotion, but (unlike the impassibilist) the Denisian passibilist does not think the language of impassibilism captures more of the truth of the matter than passibilism does. In this, Denisian passibilism differs from apophatic impassibilism, since, while the Denisian passibilist wants to say that both passibilist and impassibilist language is analogical, I think that the apophatic impassibilist would want to say that passibilist language is metaphorical (it is literally false that God is compassionate, even if that points to something else about God’s nature), and that impassibilist language, and other negative predications, are univocal.¹⁹

So far in this section I have considered Objection 3, which claims that there is an asymmetry between passibilism and apophatic impassibilism which means that the former is more liable to anthropomorphism. In response to this, I have suggested another position that the passibilist can take with respect to religious language that suggests that God transcends not only cataphatic but also apophatic language that potentially evens the playing field again between passibilism and impassibilism with respect to anthropomorphism. I have called this “Denisian passibilism” and have cited the theology of C. S. Lewis as a likely example of passibilism of this kind.

I now want to be a bit more bold, and suggest that there are reasons to prefer Denisian passibilism over the other positions we have considered here: namely, over non-Denisian passibilism, and over impassibilism, including apophatic forms of impassibilism.

It is not within the scope of this paper to answer the question of how Denisian passibilism can be squared with the idea of God as the uncaused cause, which has been taken by many impassibilists to entail that God cannot be affected (emotionally, or in other ways) by other beings. However, I think two things are worth noting. First, impassibilists – or at least those impassibilists who are libertarians about free will – face an analogous issue: if God is the primary cause of all creatures that do, and a human freely chooses to do an action – say, bake a cake – then the human makes God the primary cause of the human baking the cake, and so God is still affected by creation (see Rogers, 2023).

Second, perhaps a promising response to how Denisian passibilism might be squared with the idea that God is the uncaused cause might draw on the ideas of the third-century bishop St Gregory Thaumaturgos who argued that God’s capacity to suffer (an aspect of responding emotionally to others) will be willed by God because of God’s love, and so will be an aspect of God’s omnipotence and perfection rather than something that is outside of God’s control (see Mozley, 1926, pp. 66).²⁰ Such a response will place a high value on relational agency, and view more critically traditional notions of autonomy. In keeping with the Denisian emphasis on transcendence, such as response should also, I think, appeal to mystery. This reflects the fact that, in general, unlike both non-Denisian passibilism and certainly non-apophatic impassibilism, Denisian passibilism is rather vague and we are left without a thick description of God’s mental and emotional life. I do not think this should worry us too much. Indeed it may even be a strength: as St Augustine put it, “if you understand it, then it is not God” (Sermon 52, 16; PL 38: 360).²¹