Introduction
Consider the following depiction of God:
God revealed to David: ‘O David, fear Me as you would fear a ferocious beast’ … Whoever knows God, knows that one of His attributes is that He has no concern about destroying the entire world, and that he himself is subjugated to a power such that if that power destroyed countless beings like him, it would never be moved nor distressed (al-Ghazali 1427, 2368).
A God of this kind – indifferent to the annihilation of the world, unaffected by anything outside himself, and one before whom we should stand in fear – may exist, and we may even have reasons to obey him. But is he worthy of worship? Questions about God’s existence and about obedience to God are distinct from the question of whether God is worthy of worship. A being might possess many excellent attributes and still fail to be worthy of worship, even if we have reasons to obey it. Conversely, a being might lack some attributes traditionally associated with perfection and still be worthy of worship.
In some religious traditions, worship is the central relationship between humans and God.Footnote 1 Believers worship the God of their religion, and this worship shapes their lives and communities. Belief in God is not the same as worship of God; the deep differences between religious communities often arise not merely from belief in different gods, but from the worship of different gods. Worship, rather than mere assent to God’s existence, is often what most deeply shapes religious life and distinguishes religious communities.
Worship is a central feature of many religious traditions. Through worship, believers relate to God in ways that shape their lives, practices, and moral outlook. Yet despite its religious significance, worship has not received sustained philosophical attention. This neglect was already noted by Tim Bayne and Yujin Nagasawa (Bayne and Nagasawa Reference Bayne and Nagasawa2006), whose influential paper helped draw philosophical attention to worship as a distinct topic of inquiry. Since the publication of their paper, further articles and books have appeared, and we can now speak of an emerging ‘philosophy of worship’.Footnote 2
This paper develops this emerging line of inquiry within the philosophy of worship by arguing that worship-worthiness should occupy a foundational place within philosophy of religion. My central claim is that, before asking whether God exists or what properties God has, we should ask what kind of being is worthy of worship. I refer to this as the ‘Worship-Worthiness First’ (WWF) thesis. Most discussions in philosophy of religion focus instead on God’s existence or on perfect being theism. The question of worship-worthiness rarely serves as the starting point of inquiry. Yet if worship is central to religious life, this ordering of inquiry is not philosophically neutral. Worship-worthiness is not just one topic among others. It is the question that must come first. Issues concerning God’s existence and perfection are not irrelevant, but they should be approached in light of what is worthy of worship, rather than the other way around. This paper does not aim to offer a comprehensive theory of worship or worship-worthiness. Rather, it argues that questions of worship-worthiness should play a foundational role within philosophy of religion, even if many substantive questions about worship remain open. This starting point is not confined to theistic projects. Questions of worship-worthiness can serve as the basis of critical – including atheistic – arguments within philosophy of religion, as the discussions below illustrate.
To defend this, I distinguish two broad families of theology. Worship-Worthiness Theology (WWT) takes worship-worthiness as its central concern. By contrast, Non-Worship-Worthiness Theology (NWWT) begins from other starting points. I argue that WWT enjoys axiological, prudential, and methodological priority over NWWT. If worship is central to religious life, then questions of worship-worthiness cannot remain philosophically secondary. The following sections develop and defend these claims in turn.
The paper proceeds as follows. It first divides the philosophy of worship into two branches – axiology and deontology – and clarifies that this paper focuses on the axiology of worship. It then discusses the nature of worship and offers a definition. Next, it defends the priority of WWT over NWWT, thereby establishing the WWF thesis. Finally, it presents three case studies concerning perfect being theism, arguments for God’s existence, and the nature of God, showing how the WWF thesis reshapes these debates.
Axiology and deontology of worship
Two questions must be distinguished at the outset: Is God worthy of worship? And are we obligated to worship God? Although related, the two questions come apart. A positive answer to one does not entail a positive answer to the other. A being might be worthy of worship even if we are not obligated to worship it, and we might be obligated to worship something unworthy of worship.
Suppose Jimmy tells a joke that is both funny and racist. Everyone laughs except Peter. Laughing at the joke might be fitting (because it is funny) but not correct (because it is racist). Or suppose Jimmy envies Peter, who is successful in every way. Envy might be fitting here – Peter is enviable – but not correct. In contrast, if Jimmy envies someone who is clearly less successful than him, the envy is not fitting either. Thus, there is a distinction between fittingness and correctness (for discussion, see D’Arms and Jacobson Reference D’Arms and Jacobson2000).
Recall the depiction of God at the beginning of this paper. Suppose such a being, after slaughtering many innocent people, commands you to praise him, threatening to slaughter you as well if you refuse. Many philosophers would say that you have reason to praise him, and perhaps ought to do so. The threat gives you a reason to act. Yet the threat seems irrelevant to whether he is genuinely praiseworthy. Such reasons are often called ‘non-value-related reasons’ or ‘wrong kind of reasons’. A being is worthy of worship only when there are right kind of reasons to worship it.Footnote 3
These distinctions suggest a useful division within the philosophy of worship between two importantly different kinds of questions: axiological and deontological. Although closely related, these two dimensions of worship should be distinguished.
The axiology of worship concerns questions of fittingness and worship-worthiness, including the following:
• What does it mean for a being to be worthy of worship?
• Under what conditions is God worthy of worship?
• What forms of worship are fitting?
The deontology of worship concerns normative questions about worship, including the following:
• What normative or deontic concepts – such as obligation, reasonableness, or indebtedness – are relevant to worship, and how are they related?
• What can ground or justify the worship of God?
• Are we obligated to worship God?Footnote 4
This paper focuses on the axiology of worship. Its primary concern is not whether we ought to worship God, but what it means for God to be worthy of worship. Questions concerning whether, or for whom, worship is obligatory belong to the deontology of worship and fall outside the scope of this paper.
What is worship?
At first glance, worship seems easier to recognize than to define. People bow in prayer, sing hymns, offer sacrifices, or stand in silence. Yet worship takes radically different forms across cultures and traditions. This diversity has led some philosophers to question whether worship admits of any unified definition at all.
Avital Hazony Levi, for example, argues that worship is not a single sui generis concept. Drawing on historical and anthropological evidence – from the Mishna, where people worshipped a deity named Peor through ritual excretion, to Greco-Roman practices involving orgiastic rites – she argues that the historical diversity of worship undermines attempts to identify a single essence shared across cultures and traditions. For this reason, she claims that many philosophers have been mistaken in treating worship as a unified concept with a single essence shared across times and places (Hazony Levi Reference Hazony Levi2022, 488–491).
Hazony Levi is right to emphasize the historical diversity of worship, but this does not show that worship is philosophically indefinable. The question ‘What is worship?’ is not a purely descriptive or historical question. It is what I call a metatheological question – one that depends on broader conceptions of God, the self, and the relation between them.
Like the questions ‘What is faith?’ or ‘What is hope?’, the question ‘What is worship?’ is answered differently across religious traditions, reflecting deeper conceptions of God, the self, and the relation between them. There is no fully neutral or tradition-independent answer to the question of worship. But this does not mean that no philosophical definition is possible. It means that definitions of worship are not simply discovered, but adopted in light of broader theological and philosophical commitments. Thus, philosophers of worship should proceed in much the same way that philosophers of faith do: they may draw on the diverse ways religious traditions understand worship while still offering philosophical definitions of their own.
In what follows, I offer a minimal account of worship that is nevertheless substantive enough for philosophical analysis. I do not claim that it captures the essence of worship across all cultures. Rather, I offer it as a defensible starting point for WWT.
Two families of definitions
Elliot Salinger (Reference Salinger, Segal and Lebens2024) distinguishes between two broad approaches to defining worship: attitude-based accounts (‘A-accounts’) and action-based accounts (‘X-accounts’).
Attitude-based accounts hold that worship is essentially an internal attitude – reverence, awe, love, gratitude, or some combination thereof. Action-based accounts, by contrast, understand worship primarily in terms of practices or patterns of behaviour. Neither approach, I argue, is adequate on its own.
Consider first an attitude-based definition. Suppose Declan has all the appropriate attitudes towards God: love, reverence, awe, and belief. Yet he never acts on these attitudes. He ignores God’s commands, performs no rituals, and makes no effort to align his life with his professed attitudes. It is difficult to regard Declan as a genuine worshipper. Worship seems to require some form of action – not necessarily bodily movement, but at least a disposition to act. Without any action at all, it is not clear that the alleged attitudes are genuinely present. A person who constantly tells you, ‘You are the best and fairest mechanic in the world’, but every time his car breaks down, he takes it to someone else, gives us reason to doubt that he really believes what he says. The absence of action may give us reason to doubt the sincerity or depth of the professed attitude. Thus, some degree of action is necessary for worship.
Consider now an action-based definition. Suppose someone performs all the outward rituals of worship – praying, bowing, and going through the motions – but without any accompanying beliefs or emotions. It is difficult to regard this as genuine worship. Outward performance without the appropriate attitude is hollow. Michael Sandel offers a telling example: a father promises his children a dollar for every thank-you letter they write to him. The children write the letters, but as Sandel observes, it was clear they had been written reluctantly (Sandel Reference Sandel2012, 60). The act was performed, but we would not call it genuine gratitude. The same point applies to worship. Action without the appropriate inner state falls short of genuine worship.
These considerations suggest that neither attitude nor action alone is sufficient for worship. Worship appears to require both appropriate attitudes and appropriate forms of expression. But this leaves a further question: are attitudes and actions equally fundamental, or does one have explanatory priority in an account of worship? I return to this question in the next section.
A tripartite definition of worship
The relative importance of these components becomes clearer when we consider two contrasting cases. First, suppose a person is paralysed and unable to perform any physical act of worship. Can he still be a worshipper? Intuitively, yes – provided he has the relevant beliefs and emotions, and a disposition to act if he were able. Second, suppose a person entirely lacks the relevant beliefs and emotions – for example, an infant or someone with severe cognitive disabilities. Can such a person be a worshipper? Intuitively, no. Worship requires at least some degree of cognitive awareness and appropriate affective response.Footnote 5
These two cases suggest an important conclusion: the cognitive and affective components are more fundamental to worship, while the practical component plays a primarily expressive role. Without the appropriate beliefs and emotions, no amount of outward performance is sufficient for worship. By contrast, a person with the appropriate beliefs and emotions can still be a worshipper, provided there remains a disposition to act if action were possible. These considerations suggest the following tripartite account of worship:
For A to worship B requires that: (1) A has relevant beliefs about B, (2) A has appropriate emotions toward B, and (3) A is disposed to perform actions that express or reinforce those beliefs and emotions.
These conditions are intended as necessary rather than jointly sufficient conditions for worship. At this stage, terms such as ‘relevant’ and ‘appropriate’ are intentionally left open, since specifying their content is itself one of the questions explored in the following section. This account belongs broadly within the family of attitude-based accounts, though with two important refinements. First, I break down ‘attitude’ into two distinct components: the cognitive (beliefs) and the affective (emotions). This allows us to examine separately the norms governing the cognitive and affective dimensions of worship. Second, I assign action an expressive and reinforcing role rather than an independent one. Actions are not the essence of worship; rather, they express and help sustain the inward state of worship. This view – according to which action is important but primarily expressive – is often called the ‘expressive conception of worship’ (see Salinger Reference Salinger, Segal and Lebens2024, 21). On this view, worship comprises three interrelated components: cognitive, emotional, and performative.Footnote 6
Open questions in the axiology of worship
The tripartite account I have offered is deliberately minimal. It identifies the basic structure of worship while leaving open which specific beliefs, emotions, and actions worship requires. These are normative questions that any full philosophy of worship must address before determining what kind of being, if any, could be worthy of worship. In general, the norms governing belief, emotion, and action in other domains apply to worship as well. But worship also raises distinctive normative questions of its own, questions that any account of worship-worthiness must ultimately confront. Below, I briefly indicate some of the most important among them.
Cognitive norms: what degree of belief is required for worship? Certainty? Mere hope? Tony Manela (Reference Manela2022) argues that gratitude can remain meaningful even when the benefactor’s existence is doubted; analogous questions arise in the case of worship. Must the worshipper be certain that God exists, or can worship arise from hope, trust, or some weaker epistemic state?
A further set of questions concerns not God’s existence, but the attributes a worship-worthy being must possess. What kind of being is worthy of worship? Must God be omnipotent, omniscient, and omnibenevolent? Or are lower degrees of these properties sufficient? Traditional perfect being theism holds that God must be perfect in every respect, though this view faces serious objections.Footnote 7 Some philosophers have therefore proposed weaker conceptions. For example, Gale and Pruss (Gale and Pruss Reference Gale and Pruss1999, 468, 473) suggest that a being may be worthy of worship if it possesses sufficient power to create the universe.Footnote 8 These debates remain open.Footnote 9
Affective norms: which emotions properly belong to worship? Love, fear, awe, reverence, gratitude, trust, or some combination?
Practical norms: does worship require unconditional obedience, or can conditional obedience suffice? James Rachels famously argued that worship entails unconditional submission, and since such submission is incompatible with moral autonomy, no being can be worthy of worship (Rachels Reference Rachels1971, 335). His argument has been contested (Lombardi Reference Lombardi1988; Quinn Reference Quinn1975; Ruloff Reference Ruloff2025), but the question remains open.Footnote 10
Another question concerns what kinds and degrees of disobedience exclude one from the community of worshippers. Total disobedience clearly excludes one; total obedience keeps one within worship. But what about cases in between? Does a single act of disobedience remove one from worship? I suggest not. On the attitude‑based view I have defended, actions matter because they express inner attitudes. Disobedience excludes one from worship only when it manifests the absence of the appropriate beliefs and emotions. Occasional disobedience resulting from weakness of will, for example, may not suffice for exclusion from worship. A related question is whether worship requires positive acts such as prayer and ritual, or merely the avoidance of acts incompatible with worship, such as blasphemy or idolatry. This question, too, remains open.Footnote 11
How these questions are answered will shape our understanding of what kinds of beings, if any, are worthy of worship. These are among the central questions any adequate philosophy of worship must confront. In the next section, I argue that these questions have a kind of priority over traditional theological questions about God’s existence and attributes. Before asking whether God exists or what properties God possesses, we must first ask what is worthy of worship – and that question cannot be answered without addressing the normative questions raised here.
Worship-Worthiness First
So far, I have distinguished the axiology of worship from its deontology, proposed a tripartite account of worship, and outlined the normative questions any full account must address. I now turn to the central thesis of this paper: WWF.
The concept of ‘God’ is far from univocal. Different philosophers and theologians have worked with very different conceptions of God: God as creator, perfect being, necessary being, first cause, cosmic purpose, cosmic designer, moral agent, utterly transcendent and unknowable reality, and so on. Many of these conceptions emerge from distinct philosophical projects and arguments.
Alongside these conceptions, I propose another: God as a worship-worthy being. For present purposes, I divide theological approaches into two broad families:
1. WWT: approaches that treat worship-worthiness as their central starting point and organizing concern.
2. NWWT: approaches that take some other feature – metaphysical perfection, moral agency, or something else – as primary.
I classify the conceptions listed earlier – God as creator, perfect being, and so on – under NWWT insofar as they do not treat worship-worthiness as the primary axis of inquiry. Given these two families of approaches, my claim is this:
Worship-Worthiness First (WWF): WWT has priority over NWWT.
The next question, then, is this: in what sense does WWT have priority over NWWT?
What kind of priority?
One possible interpretation of WWF is logical priority: the view that ‘God’ simply means ‘worthy of worship’. On this view, any being unworthy of worship could not properly be called ‘God’. This position has influential defenders. Rachels’s argument against the existence of God, for example, assumes that if God exists, He must be worthy of worship, and then attempts to show that no being could satisfy that condition (Rachels Reference Rachels1971, 335).Footnote 12 Ninian Smart likewise writes that ‘there is an internal connection between the concept of God and the concept of worship’ (Smart Reference Smart1972, 51).
I do not take this view. As a descriptive matter, the term ‘God’ is used in many ways that are independent of worship-worthiness. Philosophers who search for a first cause, a necessary being, or a guarantor of morality often employ the term ‘God’ without treating worship-worthiness as a central concern. Deists, for example, may speak of God without implying that such a being is worthy of worship. Moreover, the possibility of misotheism – the view that God exists but is worthy of hatred rather than love or worship – shows that one may coherently affirm God’s existence while denying that God is worthy of worship.Footnote 13
Thus, I do not claim that NWWT projects are simply mistaken. If I accepted logical priority – defining ‘God’ as ‘worthy of worship’ – I would have to conclude that these thinkers are not really talking about God and that their projects are fundamentally misguided. I do not take this view. Their projects may still have philosophical value. But they are not primary in the sense defended here. Another approach – WWT – has been comparatively neglected. Yet it has priority over NWWT in several respects. Philosophy of religion has, in my view, often prioritized questions that are less fundamental than questions concerning worship-worthiness. The priority I defend is not logical priority, but rather axiological, prudential, and methodological priority.
Axiological priority: We can distinguish two importantly different questions about God. One concerns God’s existence. The other concerns the value God would have, independently of whether he exists. Consider the following four positions:
• God does not exist, and it is good that he does not.
• God does not exist, but it would be good if he did.
• God exists, and it is bad that he does.
• God exists, and it is good that he does.
Each of these positions combines two distinct claims: one concerning existence and another concerning value. But the value in question depends crucially on what kind of God exists.Footnote 14 A worship-worthy God transforms the value of human life in a way that a purely metaphysical first cause may not.
As I have argued earlier, worshipping God is not the same as believing in God. Worship engages the believer cognitively, affectively, and practically. It reorients their life, shapes their values, and gives them an ideal to aspire to. Worship is so transformative that one could say, ‘Tell me what God you worship, and I will tell you who you are’. So, the difference between a world in which such a relationship is possible and a world in which it is not is profound. A world containing a God unworthy of worship may therefore differ less, in axiological terms, from a godless world than from a world containing a worship-worthy God. Thus, from the standpoint of the axiology of theism, the God of WWT is especially significant, since the existence of a worship-worthy God would make a profound difference to the value and orientation of human life. This is what I mean by the axiological priority of worship-worthiness.
Prudential priority: Because our intellectual and practical resources are finite, we cannot pursue every question with equal urgency. We must therefore make judgements about which questions deserve our primary attention. If, as I have argued, the God of WWT makes a greater axiological difference than the kinds of God typically emphasized within NWWT, then prudentially – given the finitude of our time – we should focus on WWT first.
The history of philosophy of religion shows that this has not always been done. Considerable philosophical attention has been devoted to the existence and attributes of God as conceived within NWWT – whether God is a first cause, a necessary being, or a perfect being – without ever asking whether such a being would be worthy of worship. Much of this work, even if philosophically successful, may have limited implications for the practical and existential dimensions of human life.
This is not to say that NWWT projects are worthless. They may have intellectual value. But prudentially – given the finitude of our time and the greater axiological significance of WWT – we should begin with WWT. Moreover, many critics of theism (for example, those who press the problem of evil) are not merely engaged in abstract intellectual puzzles. They are concerned with whether the God people actually worship can be justified. Their critiques often presuppose that questions of worship-worthiness are central to the debate. To miss this is to miss the point of the debate.
Methodological priority: If we accept both the axiological priority and the prudential priority of WWT, then a further consequence follows: we should also adopt WWT as our methodological starting point. Before asking whether God exists, whether God is perfect, or whether God is the first cause, we should first ask what is worthy of worship? Only after answering that question can we determine what kind of being we are seeking.
The traditional order of inquiry in philosophy of religion reverses this. It begins with metaphysical questions – existence, perfection, causation. But this order puts the cart before the horse. Questions of worship-worthiness provide the orientation that should guide the broader theological inquiry.
To adopt WWF methodologically is to reorient theological inquiry around questions of worship-worthiness. It is to say: before we ask what properties God has, we must ask what a being worthy of worship would have to be like. This does not mean that the answers will necessarily be different. It may turn out that the being worthy of worship is also the perfect being and the creator. But the order of inquiry matters. It shapes both the sequence of questions and the assumptions that guide the investigation.
To summarize, the priority of WWT over NWWT is not logical priority. It is axiological (the God of WWT bears most directly on the value and orientation of human life), prudential (given our limited resources, we should focus on what matters most), and methodological (the order of inquiry must begin with worship-worthiness). This is what I call ‘WWF’.Footnote 15
An abstract defence of WWF, however, is not sufficient. To see what WWF means in practice, we need to examine how it might transform specific debates in the philosophy of religion. In the next section, I turn to three case studies: perfect being theism, arguments for God’s existence, and the nature of God. In each case, I argue that beginning from worship-worthiness reshapes both the questions we ask and the answers available to us.
Applying WWF
I have argued that theological approaches can be divided into two broad families – WWT and NWWT – and that WWT has axiological, prudential, and methodological priority over NWWT. I have called this the WWF thesis.
If one accepts the axiological and prudential priority of WWT but fails to adopt it methodologically, then one is still playing the theological game on the wrong field. In this section, I examine three major topics in the philosophy of religion through the lens of the WWF thesis: perfect being theism, arguments for God’s existence, and the nature of God. In each case, I argue that beginning from questions of worship-worthiness reshapes both the central questions and the range of defensible answers. I also highlight philosophers who, despite not explicitly endorsing WWF, nevertheless treat worship-worthiness as philosophically central and thereby arrive at importantly different conclusions. These case studies are intended to show that beginning from worship-worthiness can significantly reshape philosophical debates about God.
Perfect being theism
Many influential conceptions of God emerge from philosophical arguments concerning God’s existence and nature. Among these, the ontological tradition has been especially influential, yielding the Anselmian conception of God as ‘that than which no greater can be thought’. Following Nagasawa, I prefer to speak of the perfect being thesis rather than ‘the Anselmian thesis’:Footnote 16
Perfect Being Thesis: God is the being than which no greater is metaphysically possible.
The perfect being thesis is often taken to entail the Omni-God Thesis:
Omni-God Thesis: God is (necessarily) omnipotent, omniscient, and omnibenevolent.
If God is the being than which no greater is possible, then he must possess all great-making properties to the highest degree – and necessarily so. This conception has generated numerous objections.Footnote 17
However, there is another possible way of understanding the perfect being thesis. Charles Hartshorne and Nelson Pike suggested that there are at least two interpretations of Anselm’s formula (see Pike Reference Pike1970, 149–160; Hartshorne Reference Hartshorne1965, 25–28). The first, familiar reading, takes ‘greater’ in a metaphysical sense, leading to the Omni-God thesis. The second, less commonly explored reading, takes ‘greater’ in a different sense: more worthy of worship. On this reading:
Worship-Worthiness Reading of Perfect Being Theism: God is the being than which no more worship-worthy being can be conceived.
This is not a rejection of perfect being theology, but a reorientation of its evaluative starting point. The central question is no longer simply ‘What is the greatest possible being?’ but rather ‘What kind of being would be most worthy of worship?’ The two questions may yield the same answer, but they need not.
Pike pursued this reading and drew a striking conclusion: it requires that God be powerful and knowledgeable (perhaps even omnipotent and omniscient), but it does not require that God be perfectly good or timeless (Pike Reference Pike1970, 150–157, 160–163). On this reading, goodness and timelessness do not straightforwardly follow from worship-worthiness alone.
Perfect being theism, as standardly understood, commits itself from the outset to the omni‑properties – omnipotence, omniscience, and omnibenevolence and related perfections. Critics then attack this package. But if we start from worship-worthiness, our commitments are lighter. The point is not merely to weaken the divine attributes, but to determine which attributes are genuinely required by worship-worthiness itself rather than inherited from prior metaphysical assumptions. We may not need to insist that God is omnipotent in every detail; it may be enough that he is powerful enough to create the universe (Gale and Pruss Reference Gale and Pruss1999, 468, 473). We may not need to insist on omnibenevolence; it may be enough that he is vastly superior to us in goodness.Footnote 18 We may not even need to insist on omniscience. Consider a being that is omnipotent and omnibenevolent but lacks knowledge of some trivial fact. Would such a being be unworthy of worship? Intuitively, no. It would be absurd to say that such a being becomes worthy of worship only after learning that trivial fact.Footnote 19 WWF, therefore, raises the possibility that not every metaphysical perfection is equally relevant to worship-worthiness. This suggests that worship-worthiness may not require the full Omni-God Thesis.
Consider another debate that WWF reorients: J.N. Findlay’s argument against the possibility of God:
1. The God of religious belief must be worthy of worship.
2. To be worthy of worship, a being must exist necessarily (not contingently).
3. Necessary existence is impossible.
4. Therefore, the God of religious belief is impossible (Findlay Reference Findlay1948).
Ninian Smart responded by rejecting premise (2): necessary existence is not required for worship-worthiness (Smart Reference Smart1972, 55–59). What matters for my purpose is not who is right. What matters is that both Findlay and Smart are playing the game on the right field: they are arguing about whether necessary existence follows from worship-worthiness. The crucial point is that metaphysical questions are here being approached through the lens of worship-worthiness rather than the other way around. They are not asking ‘What is the greatest possible being?’ and only then asking ‘Is such a being worthy of worship?’ They start with worship-worthiness. This is precisely what WWF recommends.
The same point applies more generally. Perfect being theism, as standardly practised, begins with metaphysical perfection. WWF reverses the order: first ask what is worthy of worship, then see what properties such a being must have. The results may not always differ. But the order of inquiry – and the light it sheds on the debate – is fundamentally different. This case illustrates the broader methodological claim of WWF: questions about divine attributes should be guided by worship-worthiness rather than settled independently of it.
Arguments for God’s existence
Arguments for God’s existence have long occupied a central place in the philosophy of religion. The ontological argument, the cosmological argument, the design argument – each has been formulated, defended, attacked, and reformulated. The appeal is obvious: if a sound argument can establish that God exists, then religious belief appears to rest on rational grounds rather than mere assertion or tradition.
From the perspective of WWF, however, the standard treatment of these arguments is methodologically backwards. Traditionally, philosophers first ask whether the argument is valid and whether its premises are true. Only afterwards do they ask whether the being whose existence is established would be worthy of worship.
Consider a hypothetical case. Suppose a sound argument proved that God exists – but the God it proves is the ‘ferocious beast’ described by al-Ghazālī at the beginning of this paper: a being who does not care about destroying the world, who is never affected, and before whom we should stand in fear. Would we worship such a being? Would we reorient our lives around it? Intuitively, no. The existence of such a being would not generate the kind of transformative religious orientation associated with worship-worthiness. The same point can be illustrated with a deliberately simplified example:
The argument from disorder
1. The world is full of disorder and systematic evil.
2. Such pervasive disorder and evil are unlikely to exist without an intelligent source or designer.
3. Therefore, the world has a designer and creator.
Suppose this argument is sound. Does it give us any reason to worship the being it proves? No. The being it proves is, if anything, a creator of evil. Its existence would not provide the kind of ideal around which religious life is typically organized. Debating such an argument for years – refining its premises, defending it against objections – would be an odd way to spend one’s intellectual energy if the central theological question concerns the existence of a being worthy of worship.
The point is not that arguments for God’s existence lack philosophical value. Rather, WWF insists that they should be evaluated in light of a prior question: does the being whose existence is established qualify as worthy of worship? If not, then the argument may retain metaphysical significance while failing to address the central religious concern. The argument may have metaphysical interest, but it does not give us the worship-worthy God.
This is not a rejection of natural theology. WWF does not eliminate traditional theological questions; rather, it reorders them around the prior question of worship-worthiness. Once we put WWF, our expectations of the arguments change. Consider one of the normative questions raised earlier: what degree of belief in the existence of the object of worship is required for worship? If mere hope is sufficient, then we do not need certain proof. It may be enough that the arguments give us reasonable hope that a worship-worthy being exists. And if an argument establishes the existence of a being without addressing whether that being is worthy of worship, then from the standpoint of WWF the argument remains religiously incomplete.
Richard Gale and Alexander Pruss provide an instructive example. They develop a new cosmological argument that does not prove that God is omnipotent. Critics objected that their argument was therefore too weak. Gale and Pruss suggest that omnipotence is not required for worship-worthiness. What matters is that God has enough power to create the universe – and that, they claimed, is enough for God to be a fitting object of worship (Gale and Pruss Reference Gale and Pruss1999, 468, 473).Footnote 20 Whether their argument succeeds is not the point. What matters is that they are playing the game on the right field: they are asking what kind of being is worthy of worship, and then tailoring their argument accordingly.
This is the methodological priority of WWF in action. Before we ask whether an argument for God’s existence is valid, we should first ask what kind of being we are looking for. Arguments that establish only a first cause, a necessary being, or a cosmic designer – without clarifying whether such a being is worthy of worship – leave the central concern of WWF unresolved. They may be interesting, but they do not give us what we seek. And that, from the standpoint of WWF, is a methodological limitation that much of contemporary philosophy of religion has yet to overcome.
The nature of God
Any discussion of the nature of God, from the standpoint of WWF, must be guided by considerations of worship-worthiness. If worship is the central relationship between God and human beings, then what God is like matters because it determines whether he is a fitting object of worship. Here, by ‘the nature of God’, I mean questions such as whether God is personal, whether God is an agent, and, if so, what kind of agent God is – moral, arbitrary, or otherwise.
Consider the question of whether God must be personal. On the assumption that worship-worthiness is our starting point, there might be a straightforward argument for divine personhood. Brian Leftow holds that worship has two essential components: (1) worship involves addressing the object of worship; (2) this addressing must take the form of praise (Leftow Reference Leftow, Buckareff and Nagasawa2016, 71). Such address presupposes a hearer capable of understanding what is addressed. Smart agrees that worship essentially involves address (Smart Reference Smart1972, 11). If worship requires address, and address requires a hearer capable of understanding, then the object of worship must be a person – a being with a mind and will. This yields a simple argument:
The impossibility of worshipping an impersonal God
1. Worship essentially involves addressing the object of worship.
2. A meaningful address requires a hearer who can understand.
3. A non-personal being (lacking mind and will) cannot understand address.
4. Therefore, a non-personal being cannot be a fitting object of worship.
5. So, if God is worthy of worship, God must be personal (Cockayne Reference Cockayne2020, 2).
This argument is not uncontroversial. Joshua Cockayne rejects the first premise, holding that worship can take forms in which address plays no role (Cockayne Reference Cockayne2020, 4–5). And Tim Mulgan, working from a careful analysis of the norms of worship, argues that personhood is not required for worship-worthiness; he defends a conception of God as a non-personal cosmic purpose (what he calls ‘Ananthropocentric Purposivism’) that is nonetheless worthy of worship (Mulgan Reference Mulgan, Kittle and Gasser2022). For my purpose, however, it does not matter who is right. What matters is that the debate is being conducted on the right field: all parties are asking what follows from worship-worthiness. Cockayne and Mulgan are conducting the debate on the right terrain.
If the argument for personhood succeeds, it has further implications. Wesley Cray takes worship-worthiness as his starting point. He argues that God cannot be worship-worthy and omniscient at the same time. To be worthy of worship, God must be personal. But a personal being cannot possess every kind of knowledge – in particular, experiential knowledge (Cray Reference Cray2011). Thus, if worship-worthiness is our starting point, we may have to give up the strongest form of omniscience. Again, worship-worthiness functions as the controlling standard of inquiry.
But even if God must be a personal agent, a further question remains: what kind of agent must God be? Many today assume that God is a moral agent. From the standpoint of WWF, however, the primary question is not whether God is moral, but whether morality is a condition of worship-worthiness. The project of defending God as a moral agent must be pursued within the framework of worship-worthiness, not the other way around. This way of framing the issue explains a striking historical fact. In earlier periods, when societies were organized hierarchically, believers saw no difficulty in worshipping a God who was not moral in the modern sense. A God who was powerful, sovereign, and irresistible was worthy of worship simply because of his power and authority. Today, many of us feel that a being who is not morally good could not be worthy of worship. This is a change in the norms of worship. We now expect that a being worthy of worship must be good (in a moral sense). But that expectation is not conceptually self-evident; it is itself a normative stance that belongs to the axiology of worship. WWF does not deny that moral goodness may be necessary for worship-worthiness; rather, it denies that this requirement should simply be assumed prior to an analysis of worship itself.
Thus, the question ‘Is God a moral agent?’ is secondary. The primary question is: ‘Does worship-worthiness require that God be moral?’ If the answer is yes, then a moral God follows. If the answer is no, then the project of defending God as a moral agent loses much of its urgency. Mulgan illustrates this point. He argues that worship-worthiness does not require benevolence, and he defends a conception of God that is not benevolent towards human beings yet remains worthy of worship (Mulgan Reference Mulgan, Kittle and Gasser2022). One may disagree with Mulgan, but one must disagree with him at the level of the norms governing worship-worthiness.
Charles Lewis writes that ‘The requirement that a God must be worthy of worship is motivated by the same perspective which takes it for granted that moral perfection is essential to divine goodness’ (Lewis Reference Lewis1983, 152). Lewis assumes that worship-worthiness is ultimately grounded in moral worthiness. But this assumption is not obviously true. The examples we have seen – from earlier societies that worshipped God primarily as sovereign power to Mulgan’s recent defence of a non-benevolent God – suggest that one can take worship-worthiness seriously without taking morality as the ultimate standard. What mattered for earlier believers was power, sovereignty, and authority; what matters for Mulgan is something else again. In neither case is moral goodness the decisive factor. If anything, the explanatory order runs in the opposite direction: whether morality is required for worship-worthiness is a question to be settled within the axiology of worship, rather than presupposed independently of it.
Conclusion
This paper has defended a central claim: before asking whether God exists or what properties God has, we should first ask what kind of being is worthy of worship. I have called this thesis WWF.
To arrive at this claim, I first distinguished two branches of the philosophy of worship: the axiology of worship (whether God is worthy of worship) and the deontology of worship (whether we ought to worship God). I then offered a tripartite account of worship involving cognitive, affective, and performative dimensions and argued that this account generates a range of normative questions concerning belief, emotion, and obedience.
I then distinguished two broad families of theology. WWT takes worship-worthiness as its organizing concern, whereas NWWT begins from other theological or metaphysical priorities. My claim is that WWT has priority over NWWT. This priority is not logical (I do not define ‘God’ as ‘worthy of worship’). Rather, it is axiological (the God of WWT makes the greatest difference to the value of human life), prudential (given our limited resources, we should focus on what matters most), and methodological (the order of inquiry must begin with worship-worthiness).
I then illustrated this priority through three case studies. In perfect being theism, starting from worship-worthiness rather than metaphysical perfection may yield a lighter set of commitments: we may not need omnipotence, omniscience, or omnibenevolence in their strongest forms. In arguments for God’s existence, worship-worthiness changes the standard of evaluation: an argument that proves only a first cause or a necessary being, without addressing whether such a being is worthy of worship, misses the point entirely. And in debates about the nature of God – including personhood and moral agency – worship-worthiness reorients the questions we ask. The fundamental issue is not simply whether God is personal or morally good, but whether personhood or moral goodness is required for worship-worthiness in the first place.
Other applications of WWF are possible. Consider negative theology, which insists on God’s radical transcendence and unknowability. If worship requires some minimal cognitive grasp of its object – and if, as I have argued, love and awe are constitutive elements of worship – then a God who is wholly unknowable may become difficult or even impossible to love. The axiology of worship therefore places significant pressure on radically apophatic conceptions of God. I leave this and other applications for future work.
It is worth emphasizing that the idea of a worship-worthiness theology is not entirely new. Murphy (Reference Murphy2017) and Kvanvig (Reference Kvanvig2021) both recognize versions of WWT, but neither argues that it has axiological, prudential, and methodological priority over rival approaches. What is new in this paper is the claim that WWT has such priority, and that this priority should significantly reorient central debates within philosophy of religion.
In the introduction, I noted that despite the centrality of worship in the lives of believers, the topic has received surprisingly little philosophical attention. The situation is even worse for WWT itself. I have tried to show not only that WWT is philosophically viable, but that it should be the starting point for any theology that takes worship seriously. The three case studies in this paper are only a beginning. If I am right, philosophers of religion need to change the field on which they play. The question is no longer simply ‘Does God exist?’ or ‘What properties does God have?’ The first question – the question that gives direction to all the others – is this: What makes a being worthy of worship?Footnote 21
Open access
