The Central Premise

Picture yourself – as vividly as possible – as living forever. Wouldn't it be wonderful to enjoy all the goods life offers, such as exotic food, love, and philosophy, in all eternity? Many people, and we count ourselves among them, do not think so. They find an eternal life undesirable. There just does not seem to be any activity, or complex of activities, that one would like to continue not just for one hundred, one thousand, or even a billion of years . . . but forever.

When we imagine ourselves to be immortal, it seems that at some point boredom would set in. And boredom apparently would not only be a contingent psychological reaction an immortal person might display or not. Rather, it seems that in the very long run, experiencing and doing things become worthless from a personal point of view. At this point, boredom seems to be the appropriate attitude towards one's never-ending existence. It would be inappropriate – a display of irrationality – never to become sick of it all.

The thought experiment of picturing oneself as living forever supports

So far we have been relying on intuition in justifying Central Premise. But this is not the only form of support. In particular, we take it that Bernard Williams has argued for something like Central Premise in his famous article on the Makropulos Case.Footnote ³ Williams (Reference Williams and Williams1973, 89) said that ‘[i]mmortality, or a state without death, would be meaningless’, and that ‘in a sense, death gives the meaning to life’. Williams (ibid., 90) holds this view on the grounds that, if an immortal person has a certain character, then every possible meaningful thing that could happen to this person will have happened after a while. The ‘character’ a person has is, on Williams's (ibid., 85–86; cf. Williams (Reference Williams and Williams1981b), 13) view, constituted by the person's ‘categorical desires’. These are desires that are not conditional on the agent's staying alive. Instead, they are what determine whether one should keep on living by giving one reasons to live, namely, in order to engage in certain projects and personal relations. In the course of an endless existence, however, it is inevitable that one's categorical desires become permanently satisfied. At this point the immortal's existence loses its meaning and boredom will set in as an appropriate reaction to his situation.

Williams's argument is best represented in the form of a dilemma. Either the envisioned immortal life consists of an infinite sequence of character shifts, of periods characterized by significant differences in personality, interests, etc. Or the life is one constituted by a stable character. The former option Williams rules out as not being something that an individual can rationally look forward to. We call this the ‘character-horn’ of the dilemma. The second option will, according to Williams, lead to repetitive monotony: the immortal will continue repeating the same set of experiences, as her character, which determines what experiences are worth for her to live through, will remain constant. In this way sooner or later she will permanently satisfy all her categorical desires, leaving her in a state of boredom and meaningless existence; in short, without any reason to continue living. We call this the ‘boredom-horn’ of the dilemma. Immortality, Williams concludes, is not a rationally desirable option, contrary, perhaps, to appearance.

This article sets out to apply Williams's argument to the case of God. However, before this happens, we would like to make clear what we set out to achieve. There are two points to emphasize here. First of all, one does not have to agree with Williams's argument in order to accept Central Premise. In light of the thought experiment entertained at the outset of this section, Central Premise seems plausible in its own right. However, this being an intuition, there is not much to argue about it: one either has it or one doesn't. What one can do, and this is also relevant for the intuitiveness of Central Premise since one can acquire (and retain) an intuition upon reflection, is to provide an argument for Central Premise. This is where Williams's argument comes into play. Of course, and this is the second point to emphasize, Williams's argument is much contested. There is significant literature on the human case that Williams discusses but none, to our knowledge, on the connection between God and boredom. This means that a comprehensive discussion of the subject would have to see the applicability of each and every attempt to respond to Williams's treatment of the human case to the case of God. This naturally goes beyond the scope of this article: although we do take up some of these arguments, we do not claim to be comprehensive in our approach. Hence, strictly speaking, our argument in this article attempts to establish a conditional conclusion: assuming that our defence of Williams's argument is not subject to objections from the part of the literature that we do not discuss and is in general sound, God cannot be omnitemporal.

The argument

We can now state our argument:

1. (1) God is the greatest possible being.

2. (2) If God were omnitemporal, then he would be bored.

3. (3) The greatest possible being is not bored.

4. (4) Therefore, God is not omnitemporal.

Our argument is based on perfect being theology. This commitment becomes apparent in premises (1) and (3). Premise (2) is based on Central Premise. We will comment on both aspects in this section and then address objections in the remainder of this article.

Perfect being theology is a discipline that deduces God's essential properties from the assumption that God is the maximally great being.Footnote ⁴ We assume that in order to determine whether a certain feature f is an essential property of God, we must ask whether the maximally great being necessarily has f. Of course, some great-making properties could be incompatible so that even the greatest possible being could not have all such properties. Another complication is that some incompatible properties (or sets thereof) could be great-making to different degrees. Also, whether and, if so, how great-making a property is could depend on several contextual features. These complications notwithstanding, God, as the maximally great being, must have a set of properties the possession of which is all things considered maximally great-making (in the sense of being at least as great-making as any alternative set of properties a being can possess). In order not to clutter the exposition, we assume in what follows that there is exactly one such set of properties and we use ‘great-making property’ to refer to any member of this set.

Perfect being theology implies, we further assume, that omniscience, omnipotence, moral perfection, and personhood are great-making properties.Footnote ⁵ For our purposes, though, the crucial question is whether the greatest possible being is bored. If it is, then premise (3) is false and our argument fails.

A necessary condition for boredom being a threat to God is that God is personal. This feature of God is also assumed to be deducible by perfect being theology. With the doctrine of the trinity in mind, we prefer to say ‘personal’ rather than ‘a person’ because we do not want to exclude the possibility that God is more than one person. We take it that a person has a mind (with beliefs, desires, emotions, and so on). God, hence, has at least one mind.

There are two reasons for thinking that premise (3) is true. First, it intuitively just seems to be true that the greatest possible being cannot be bored. Of course, we have introduced the concept of a set of properties the possession of which is all things considered maximally great-making and such a set may in principle contain properties that make someone pro tanto less great than he would be without these properties. However, it may well seem at this point that what we said above needs some qualification: there seem to be properties – and persistent existential boredom is certainly a candidate for such a property – that intuitively a greatest possible being can hardly possess. So apart from the overall balance of properties a person must or must not have in order to be maximally great, there seem to exist restrictions on the kind of properties the person may have.

The second reason to believe that (3) is true is based on comparisons. A timeless, omniscient, omnipotent, and morally perfect personal being seems to be greater than an omnitemporal, omniscient, omnipotent, and morally perfect personal being, if the latter suffers from boredom and the first, as seems plausible, does not.Footnote ⁶ This suggests that timelessness is a great-making feature and God must be timeless. However, it has often been argued that God cannot exist in a timeless mode (e.g. McCormick (Reference McCormick, Martin and Monnier2003)). If this is true, one might object, then God would have to be omnitemporal after all. This objection fails, though, because the assumption that God cannot be timeless casts doubt on perfect being theology in the first place. For it seems essential to our idea of God that he merits worship. But he hardly would, we submit, if he were significantly bored. A personal being who suffers from boredom is a proper object of pity, not of worship. The God of personal being theology, if he cannot exist outside time and could exist in time only at the cost of eternal boredom, does not exist. There could still be a maximally great being. But this being, if it were omnitemporal, would not be personal; for otherwise it would be bored.

Let us turn to premise (2). Central Premise gives us reason to believe that it is inappropriate for persons not to become bored after a sufficiently long sequence of time has passed. Since God is a perfectly rational personal being, he has all the attitudes it is inappropriate for persons not to have. And if God is omnitemporal, then a sufficiently long sequence of time (in the sense of Central Premise) has passed for him.Footnote ⁷ Therefore, if Central Premise is true, an omnitemporal God would be bored.

Premise (2) follows from Central Premise. Now, as noted in the previous section, we do not merely endorse the premise on intuitive grounds but also argue for it using Williams's ideas on the Makropulos Case. Our defence follows in subsequent sections, but before we move on to them we need to face a methodological objection that appears to be clearly looming here: how can we know what it is like for God to face eternity? We have, after all, no phenomenological access whatsoever to what things are like for God. It is no easy task to put yourself in God's shoes and picture what things must be like for God; so, how can we reason about God's boredom, which has clearly to do with God's psychology?

Two points in response. First, even though it is certainly no easy task, we are inclined to think that we have the relevant kind of phenomenological access.Footnote ⁸ We can at least try to envision fascinating activities usually ascribed to God, such as creating and sustaining worlds and making and saving souls. To us, at any rate, doing these things for all eternity does not seem to be desirable. A divine activity one would like to be engaged in forever simply does not come to mind. Second, and perhaps more importantly, most of what we do in the following sections does not concern psychological speculations about God's inner life. Central Premise puts forward a normative claim and not a psychological or phenomenological one and, accordingly, our defence mostly consists in arguing for what is appropriate and what is not appropriate for God to have, to pursue, or to be like. These are not psychological but logical, conceptual, metaphysical, or ethical claims that are therefore not subject to the methodological objection under consideration.

Having set this worry aside, in the next section we have to face in more detail an objection similar in nature. We will deal with the charge that we anthropomorphize God and that, therefore, our argument is a non-starter. The section that follows discusses the character-horn of Williams's dilemma and aims to show that God has a fixed set of categorical desires. The next five sections concern the boredom-horn of Williams's dilemma: we argue that none of God's categorical desires will prevent him from suffering from boredom if he knows the future and is omnitemporal. Finally, in the penultimate section we show that, while divine ignorance could be a way for God to escape boredom, this solution comes with high theological costs.

A category mistake?

A natural way to spell out the charge that we anthropomorphize God is to accuse us of making a category mistake. ‘Having a boring existence’ as well as ‘avoiding boredom’ are categories for humans with their psychological shortcomings but these categories do not apply to God because God is maximally great and thus has a perfect psychology.

Proponents of the objection, however, need to say more than this. We share Williams's intuition that rational persons cannot – on pain of endless character shifts – avoid boredom in an omnitemporal existence. The objection simply denies this general claim when it comes to God, pointing to his perfect psychology. But appeal to God's being the greatest possible being will not do. We accept that God is the greatest possible being and has a perfect psychology. The dialectical situation is such that we argue for this very reason that God will be timeless rather than forever bored. Recall that according to Central Premise it seems appropriate to be bored when a sufficiently long sequence of time has passed.

To see what is wrong with insisting that God, in virtue of his being maximally great, can be omnitemporal without being bored consider the following analogy. Some passages in the Old Testament seem to say that God commanded his people to commit atrocities such as genocide. Whatever the right reaction to these passages might be, it certainly is not reasonable to accept these atrocities as being justified (pace Craig Reference Craig2007). But someone who does exactly this could try to argue that, since God commanded those atrocities and God is morally perfect, the genocide was right and even obligatory. This argument, however, would be circular in a dangerous way. Rather than letting our well-reflected moral judgements determine which actions can be attributed to God because they are consistent with God's moral perfection, and which cannot, this argument leads us to give up our well-reflected moral judgements without offering any moral considerations to this effect. Such an argument bears witness to a blind, unreasonable faith.

This analogy suggests that we should take seriously our well-reflected normative intuitions when determining God's features rather than presuppose some (non-essential) features of God and adapt our intuitions or restrict their scope accordingly. Our judgement that it is appropriate for omnitemporal agents to suffer from boredom rests on well-reflected intuitions. This judgement enables us to put forward an argument for God's being timeless rather than omnitemporal. To say that God cannot be bored simply because he is the greatest possible being is as unreasonable as saying that genocide commanded by God would not be wrong (or would not be genocide) simply because it is commanded by God. We should not limit the scope of our judgement that it is appropriate for omnitemporal agents to suffer from boredom as long as we do not hear arguments that concern the very subject matter for the limitation, that is, arguments dealing with the appropriateness of attitudes of agents who face eternity.

Why God has a fixed character

With the objection that our argument is a non-starter out of the way, we can now turn to the character-horn of Williams's dilemma. In the human case one can legitimately enquire why Williams rules out an immortal whose character continuously changes in the course of her endless existence.Footnote ⁹ However, such enquiry could appear self-contradictory in God's case. For God has often been considered to be immutable in the sense that he is not amenable to change with respect to any of his non-relational properties. Consequently, the claim that God's character changes in order to avoid the first horn of Williams's dilemma seems to be ruled out because God is unchangeable.

However, this conclusion is too quick. It is often pointed out that an omnitemporal and omniscient God cannot be immutable if certain of his beliefs, namely those that contain temporal indexicals, change with time.Footnote ¹⁰ To give an example, the proposition

is only true, on a tensed view of time, at the time when it is raining, namely now, hence the only time a God who exists in time can know this proposition is now. That is, he can have neither foreknowledge nor memory of this particular proposition, nor of any other similarly indexical proposition. This does not mean that God is not omniscient. God still knows all true propositions at every point in time and believes no false ones. However, the existence of propositions with temporal indexicals does pose a threat to God's immutability, for he literally has to change his mind, i.e. his beliefs, in order to gain knowledge of these propositions. Also, this change in beliefs might well bring along changes in intention and emotion. Hence, it seems, it is far from obvious that immutability is an essential attribute of God.

There is a good way to handle this problem. It is to distinguish loose and strict immutability. Unlike the latter, the former is compatible with God's having changing beliefs; what it maintains is that God's character does not change, the idea being that strict immutability is arguably a natural attribute of a timeless God, but is unnecessarily demanding of an omnitemporal God (cf. Gale (Reference Gale1986), 332–333; (Reference Gale1991), 94–97).Footnote ¹¹ The character of a person, to repeat, is, according to Williams, constituted by her categorical desires, which are not conditional on her staying alive but which give her reasons to live in order to engage in projects and personal relationships. Loose immutability, we submit, is an essential property of an omnitemporal God. And God's having loose immutability is sufficient for ruling out character changes.

Why would an omnitemporal and omniscient God's character not be affected by his changing beliefs? First, one can argue that loose immutability is a great-making property. After all, immutability often figures on lists of God's essential properties and this seems plausible. Second, changes in God's beliefs are not fit to effect character changes in the first place. Since God is assumed to know the future, he has foreknowledge about which temporally indexed propositions he will know (cf. Neri-Castañeda (Reference Neri-Castañeda1967)). There will thus be no surprises or insights for God that would give him reason to adapt his categorical desires. To illustrate, God has known at every point in time prior to 2014 that

So God's knowledge that it is now 2014 came as no surprise to him. Of course, God can also acquire knowledge of propositions that prompt changes in his intentions and emotions, such as

or

But being omniscient, he always knew that things would go this way and therefore has no reason to change his categorical desires. So the upshot is that the changes God's beliefs undergo do not affect his character.

Desire classifications

Let us then come to the boredom-horn of Williams's dilemma. In order to avoid it, one could accept Williams's conception of categorical reasons and argue that God's categorical desires, in contrast to humans’ desires, will not become satisfied. How can one show this? Three basic ways come to mind. First, if God has a fixed set of categorical desires with infinitely many members, one can argue that even eternity will not be enough to satisfy permanently all these desires. Second, the content of a desire can render the desire forever unsatisfiable. The desire expressed by the phrase ‘I want to exist at every point of time’ provides a good example. Third, there may be desires that can be infinitely often satisfied. Suppose that Adam wants to listen to music and fulfils this desire. At some later point, he wants to listen to music again. Arguably, in this example Adam possesses only one desire (to listen to music) which becomes satisfied only temporally and cannot be fulfilled permanently but reappears in his consciousness regularly.

Whether these three ways to avoid the boredom-horn collapse into one or two depends on how one individuates desires. We will not address this issue. We also grant that, as they stand, it is at least conceivable that these scenarios disarm Williams's challenge in the human case.Footnote ¹² What we deny is that they can serve as a response in God's case. To show this, we employ a categorization of desires that is based on the contents of desires. We distinguish self-regarding, other-regarding, and impersonal desires. A desire is self-regarding if it is about the desirer. A person who wants to eat an apple has a self-regarding desire: the person wants that she eats an apple. A desire is other-regarding if it is about people other than the desirer. If Eve wants Adam to eat an apple, she has an other-regarding desire. Impersonal desires are neither about the desirer nor about other people. If Eve wants that there be an apple tree in the garden, then she has an impersonal desire.Footnote ¹³

In the next sections, we will show that none of the three types of categorical desires will help God to avoid boredom. Since our classification of desires is exhaustive, we conclude that an omnitemporal God would be bored.

Self-Regarding Desires

How could self-regarding desires help God to avoid boredom? A natural suggestion is that self-regarding desires are about something which is good for God or make something good for him. Consider the three suggested ways to avoid boredom. A first way would be for God to have infinitely many desires about himself. Second, an unsatisfiable desire, such as expressed in the sentence ‘I want to exist forever’ (if there is an infinite amount of time to come) or the aspiration for a personal ideal that cannot be achieved, could help God to avoid boredom. Third, desires which are infinitely often satisfiable – as might be, in the human case, the desire for eating lobsters or the desire to play musical instruments – would prevent boredom.Footnote ¹⁴

Self-regarding desires do not prevent divine boredom for the following reason. Notice, as a starting point, that God is not any old person, but the supreme being. Desires for trivial affairs cannot constitute his character, they cannot be essential to his personality, because this would not fit God's greatness. This is why desires for God to have pleasurable experiences, like eating lobsters, desires like the one expressed with ‘I want to exist forever’, or desires for playing musical instruments cannot function as categorical desires in God's case. Even some extraordinary desires, such as creating and sustaining worlds, do not seem appropriate to play the role of a categorical desire in God's case.

This is not to say that all self-regarding desires are trivial. Desires to strive after personal ideals, like knowledge or virtue, are self-regarding but would be non-trivial. However, such non-trivial self-regarding desires are not available to God because he already possesses all perfections. There is nothing God could aspire to in these regards.Footnote ¹⁵

Have we overlooked a candidate desire? At this point we want to address what is probably the most important objection concerning the boredom-horn: the suggestion that God delights in himself. God is the most perfect, supreme, complete being; hence, so the suggestion goes, his eternal life is spent in apprehension of this inexhaustible goodness and greatness. To put it in Williams's language, God's love of himself is a desire that is non-trivial, while being also unsatisfiable, and therefore a life based on it is immune to the threat of boredom.

Notice that this objection does not presuppose the acceptance of Williams's claim that categorical desires constitute character. One could rephrase it by simply saying that an omnitemporal God would not be bored because God has an unlimited source of meaning and joy: himself.

There are some points that are unclear about this proposal, such as that God, being what he is, really needs an infinite amount of time to apprehend even infinite values. But we let this go.Footnote ¹⁶ The real problem with this approach lies in the motivational picture that it paints of God. On this picture, what God delights in is himself, and not his deeds, not what he gets done, and so on. In other words, God has a second-order reflexive attitude towards his own image as someone who is perfect, great, complete, and so on, instead of having a second-order attitude towards his own first-order projects, of all the good deeds and so on that he has done. The challenge is to spell out how this kind of attitude is any different from being smug or self-indulgent. Williams's (Reference Williams and Lewis1976, 314) words, concerning moral self-indulgence with respect to generosity, well describe what is problematic about this picture:

Moral self-indulgence, however, is a vice, a character trait that a morally perfect being cannot possess. The same seems to be true of self-indulgence related to one's greatness. Consequently, if we cannot seize a distinction between self-indulgence and God's eternal love of himself, then the second cannot plausibly be held to help him avoid boredom.

One could object that we need to distinguish between two constellations. If an agent merely thinks himself to be great but in fact lacks this property, then delight is inappropriate and the agent may not appropriately love himself. On the other hand, so the suggestion goes, it is appropriate for agents who really are great to recognize this fact and delight in it. God, of course, belongs to the second camp.

It is true that God belongs to the second group and is, moreover, maximally great and rightly recognizes this fact. What we deny, though, is that it is appropriate for great agents to delight in their greatness. Our opinion is based on intuitions about cases of human greatness. Take a decent human person, Chris, who has many virtues and is admirable in several respects because, say, he is a great artist as well as a brave politician. Let us assume that Chris is unique among his contemporaries when it comes to goodness and greatness. Several positive attitudes towards Chris seem appropriate: we should praise him and it also seems appropriate for other people to delight in Chris. However, it would be an unfortunate aspect of Chris were he himself to take delight in his extraordinary qualities. Notice that it may be appropriate for Chris to recognize his goodness and greatness. But other things being equal, it seems to us that Chris is a better or more appealing person if he does not delight in his greatness. We conclude, therefore, that the strategy of distinguishing between appropriate delight in oneself and (inappropriate) self-indulgence does not seem to work.

A critic might suggest that the difference we are after is simply the difference between God, on the one hand, and humans, on the other. Our observations are based on the case of humans, not God, and what is appropriate delight in oneself in the case of God would be self-indulgence in the human case. After all, the critic might add, God is himself maximally great and hence he is such that it is appropriate, also for himself, to take delight in him.

However, we have seen earlier that we should take our well-considered normative judgements seriously. We therefore do not see a problem with our intuitions being based on the case of humans and think that not applying these intuitions in the case of God would be ad hoc. Of course, God is maximally great, whereas Chris has significant shortcomings (at least when compared to God). However, since it seems inappropriate for someone like Chris to be self-indulgent, despite his virtues and other great-making features, we have reason to believe that it is morally better not to delight in one's advantages and merits, even if they are extraordinary. We see no reason to make an exception to this principle when it comes to God. Again, we do not see how this strategy to distinguish between appropriate delight and self-indulgence could succeed.

The bottom line is that God's self-regarding desires will not save him from boredom because they are too trivial. God cannot have non-trivial aspirations for perfections, as he is perfect. Moreover, it is hard to see how an appetite for his own greatness would not be vicious and thus at odds with his moral perfection.Footnote ¹⁷

The trinitarian objection

If God is a person, he cannot enjoy his own greatness in eternity. However, let us consider a follow-up proposal that is based on an interpretation of the doctrine of the trinity. Trinity monotheism, as William Lane Craig and J. P. Moreland label their position, claims that:

Proponents of trinity monotheism could argue that each divine person's delight is derived from the presence of the other two persons. ‘Life so conceived’, says Craig (Reference Craig1998, 117), ‘is not only not boring, it is enthralling.’ Since none of the persons takes delight in himself, there is nothing repugnant about the proposal. God can eternally love himself without showing self-indulgence.

Notice that there are two principal ways in which the trinitarian objection can be spelled out. The first suggestion is that each divine person enjoys the interpersonal relationships with the other two divine persons. Another suggestion is that each divine person takes delight in the trinity of divine persons, as the trinity is God.

However, neither of these ideas works. Notice that, on trinity monotheism, God is not identical to any of the divine persons but to the trinity of divine persons.Footnote ¹⁸ It is for this reason that trinity monotheism avoids the charge of polytheism: there are three distinct divine persons but none of them is a god. However, this claim renders untenable the suggestion that each divine person enjoys the interpersonal relationships with the other divine persons in a way that prevents boredom: since none of the divine persons is identical to God, none of them has inexhaustible greatness, which however would be necessary in order to function as an eternal source of meaning and joy for the other divine persons. Hence this solution is a non-starter.

Let us come to the second idea. Notice that each of the divine persons is a constitutive part of the trinity alias God. On the second idea, the divine persons spend an omnitemporal existence in apprehension of the inexhaustible goodness and greatness of a tripersonal being of which they themselves are a constitutive part. The constellation seems comparable to individuals’ being members of groups with merits and qualities that go beyond the individuals’ merits and qualities. An analogue to a divine person is the member of a charity.

The problem with the second idea is that the charge of self-indulgence re-emerges. Although none of the divine persons would think something like ‘I delight in myself’, it seems that a thought along the lines of ‘I delight in us’ can properly be ascribed to each divine person. How is this form of delight different from self-indulgence? It seems that neither perfectly virtuous charity members nor perfectly virtuous divine persons would have such a thought. Charity members display humility and appear to be more virtuous if they derive pleasure only from the good results the charity achieves rather than also from the greatness of their charity. This suggests that the divine persons would have a vice, which can be best described as a kind of collective self-indulgence, if they are delighted in the greatness of the trinity. As a result, the trinitarian objection fails.

Other-regarding desires

How could other-regarding desires help God to escape boredom? We have said that we think of creating and saving souls as fascinating activities as long as they are not pursued forever. One could object that the desire to create and save souls is unsatisfiable because there is no largest number of souls God could create and save. Let us grant this. Even so, we have a serious worry concerning the viability of other-regarding desires as vehicles to avoid boredom. If the active, impertinent pursuit of good outcomes is the only way for God to avoid boredom (as we argue in the previous and the next section), this would eventually crowd out other activities, pursuits, and character traits of God, leaving him merely with the pursuit of optimal outcomes and devoid of a colourful character. In short, God would be a true moral saint, with all the problems moral saints face (cf. Wolf (Reference Wolf1982) ).

There might be a way out, though. It is to grant that these desires are permanently satisfiable, hence that creating and saving souls sooner or later becomes boring for God, but point out that he still would have moral reason to continue these activities. The general idea is that an agent can continue his life for the sake of other persons, although he might have no self-interest in doing so or might even be better off dead. This agent – God – has moral reason to embrace eternity. Call this the argument from self-sacrifice.

As further support, it can be added to this that living a moral life brings meaning to that life. Since, moreover, the meaning of a life is typically considered as a positive final value, a property that is desirable for its own sake (Metz (Reference Metz and Zalta2007)), there seems to be another strong reason for God to maintain an endless boredom-filled existence.

The objection from self-sacrifice brings to the fore further assumptions behind Williams's dilemma. For Williams would certainly deny that humans – and, by extension, God – can escape his dilemma in this way. His dilemma is premised on his claim that all reasons are internal, being based on the agent's desires (in a suitably broad sense of desire; see Williams (Reference Williams and Williams1981a)). Moral desires would be no different and hence no exception to his argumentation: they too would be permanently satisfied in an infinitely long existence. Hence, according to Williams, there would be no moral reason to choose an immortal life. There would, furthermore, be no other point to an infinitely long moral existence either, according to Williams (Reference Williams and Williams1973, 95–96). The meaning of a life has to do with the projects – categorical desires – in it. When all these projects are completed, when all the categorical desires are permanently satisfied, no meaning pertains to the life any more. Hence, if we follow Williams, the meaning of his life cannot be what gives God reason to continue his existence either – since there would be no such meaning in his life when stretched to infinity.Footnote ¹⁹

Of course, one can question these assumptions: both the theory of internal reasons and the theory of the meaning of life that we attributed to Williams can be disputed. However, engaging with these separate debates goes beyond the scope of this article. Nevertheless, for the sake of further illumination, let us set aside Williams's background assumptions for the moment. Even so, the objection from self-sacrifice is not persuasive when it comes to God for considerations that do not concern moral reasons or the meaning of life, but well-being – which for our purposes we can equate with happiness.Footnote ²⁰ In our examination of God's eternity, we are concerned with a comparison of God's greatness in an omnitemporal mode of existence with his greatness in a timeless mode of existence. Even if God's omnitemporal existence would be meaningful in virtue of his morally desirable self-sacrifice, it would be better for God to be timeless. And doing well (being happy), we submit, is a great-making property. Hence, the timeless God would be greater.

However, one might point out that happiness is not traditionally considered a perfection in perfect being theology. Which reasons are there to add happiness to the list of great-making properties? How, in general, can we determine whether a particular feature is great-making or not? Greatness is a normative property. It would go beyond the scope of this article to engage in the epistemology of normative properties in detail. However, we have already committed ourselves to a view that we take to be an application of the epistemology of normative properties to perfect being theology: we should base our views of greatness and perfection on our well-reflected normative intuitions. Let us reconsider the admirable human person Chris, whom we introduced earlier. Chris is an extraordinary artist and a brave politician. Now suppose a contemporary of Chris, Doris, shares these virtues and has accomplished similar achievements. The difference is that Doris is constantly happy and appreciates life whereas Chris is deeply frustrated and always feels miserable. It seems – to us, at any rate – that, even though Chris would merit some extra moral admiration for his achievements in virtue of the motivational difficulties he has to overcome, Doris is more of a role model for her contemporaries, more worthy to be praised as someone to aspire to. This suggests that Doris is greater than Chris, which in turn suggests that happiness is a great-making property.Footnote ²¹

There is an additional reason to think that happiness is a great-making property. In his omnitemporal mode of existence God could properly be called a tragic hero: he would be suffering debilitating boredom for the sake of morality. Owing to his heroism, he would be a proper object of moral admiration. But the tragic aspect would to some extent render him pitiful. Being worthy of pity, even if it is only to some degree, does not seem to fit God's being the proper object of worship.Footnote ²² Now, since God is thought to be a proper object of worship in virtue of his great-making features and he would be a proper object of pity in virtue of his suffering from boredom and his being a proper object of worship seems to be at odds with his being a proper object of pity, we have reason to believe that suffering reduces greatness and, in turn, happiness is a great-making property.Footnote ²³

To our explanation, in the second section, that God must have a set of properties the possession of which is all things considered maximally great-making, a critic might object that even though boredom would render God pro tanto less great, the maximally great-making set of properties could all things considered still include boredom. This is correct. However, it is not enough for the critic to point out the possibility that the maximally great-making set of properties could include boredom. Rather, the critic has to argue that it does include boredom. Since boredom makes you pro tanto less great, we have some reason for thinking that boredom is not a member of that set. We doubt that the negative effect of boredom could be outweighed by the value of great-making properties that somehow require boredom. Of course, in the case of humans, there might be miserable artists who without suffering would not be such great musicians, say. But we do not see what could compensate for God's suffering from boredom in an omnitemporal mode of existence if he could instead just be atemporal.Footnote ²⁴ At least, given that boredom makes you pro tanto less great, the burden of proof is on the critic. Finally, notice that even if the benefits of God's temporal existence in terms of greatness would compensate for his boredom, our argument could be adapted: the boredom tied to an omnitemporal existence would still give some reason for thinking that God is not omnitemporal and could be combined in a cumulative case with other considerations.

In sum, independently of whether the assumptions behind Williams's dilemma are viable, the argument from self-sacrifice does not help God to escape the dilemma. For it is not only having a meaningful life that qualifies as a great-making property; happiness is a great-making property, too. And a morally perfect omnitemporal person would lack this property. Hence other-regarding desires do not help God when it comes to boredom.

Impersonal desires

The category of impersonal desires covers all the desires which are neither self-regarding nor other-regarding. However, it is hard to find examples of impersonal desires in God's case. Desires with moral content are not impersonal because (at least) moral agents figure in them. Maybe the following works: the desire ‘that there is an eternal world’ (or ‘that there is an infinite sequence of worlds’ or ‘that there is an infinite sequence of worlds that realize natural aesthetic values’) – with neither ‘God's creating the world’ nor ‘humans inhabiting the world’ entering the content of the desire.

There are at least two problems with this proposal. First, traditionally God is believed to care about us. This does not fit with the suggestion that God's reason for existing is provided by his categorical desire ‘that there is a world’ rather than ‘that there is a world for the sake of humans’ (or ‘that there is beauty’ rather than ‘that there is beauty to be enjoyed by someone’). We do not play a sufficiently important role in the picture of God as proposed here.

Second, the triviality problem, which we encountered in the context of self-regarding desires, reappears. Why should it matter, on an existential scale, whether there is a world, even if it has aesthetic value in it, as long as it does not benefit anyone? (If the desire were about a world that benefited God or other beings, it would not be impersonal.) Why should the supreme being care? Whether there is a world seems, by itself, an issue too trivial to provide the relevant impetus for God's existence, given what God is necessarily like.

This concludes our elaborations on the boredom-horn of Williams's dilemma. Neither self-regarding nor other-regarding nor impersonal desires provide a way for God to avoid boredom. We conclude that an omnitemporal God would be bored.

Divine ignorance

In this section we will present an objection that we will not rebut for lack of space. The objection presupposes a very controversial theory. If the theory is true, our argument might fail. But since the theory is very controversial, it is not a big problem that we react with a caveat rather than full-fledged defence of our argument.

The theory we have in mind is Open Theism, according to which God does not know the future. Open Theists assume a libertarian theory of free will and hold that God is omnitemporal and cannot know the future because humans have libertarian free will.Footnote ²⁵ If Open Theism is correct, our argument might fail, for several authors have argued that cognitive limitations provide a way to avoid boredom for immortal humans.Footnote ²⁶ These proposals become relevant if God does not know the future and thus has cognitive limitations.

How can such a proposal work? With respect to self-regarding desires, we distinguished between trivial and non-trivial desires. Only non-trivial desires have a content that is intuitively apt to function in the existence-maintaining role for a being like God. The problem with non-trivial desires was that God, since he is perfect, cannot improve himself. However, if God does not know the future, then he can improve. Having knowledge is a great-making feature and the more God knows, the greater he becomes. The desire for knowledge is thus an appropriate way for God to avoid boredom.

There are several problems with Open Theism – apart from its conflict with some passages of the Holy Scriptures. For example, it is hard to draw the conclusion that God does not know the future from the assumption that humans have libertarian free will. Moreover, it is natural, at this point, to worry whether God would be omniscient if he did not know the future.Footnote ²⁷ Furthermore, if God does not know in advance what humans will do, this has some impact on his control of human history. He could not choose outcomes but only lotteries of outcomes. In contrast, in a timeless mode of existence God is not dependent on lotteries but can choose outcomes directly. In the case of a timeless God, there would be no unforeseeable events in the form of free choices and God could choose outcomes directly. One might well wonder, for these reasons, if Open Theism really provides plausible accounts of God's foreknowledge and omnipotence. His inability to secure optimal outcomes on Open Theism may also affect God's moral perfection.

The bottom line is that if Open Theism is true, our argument might fail. For cognitive limitations could possibly enable an omnitemporal God to strive for ideals and thus give meaning to his existence. However, we have also indicated why Open Theism is a controversial theory with significant problems.