Is the desire for life rational?

Abstract

The question of the meaning of life has long been thought to be closely intertwined with that of the existence of God. I offer a new theistic, anti-naturalist argument from the meaning of life. It is argued that the desire for life is irrational on naturalism, since there would be no good reason to believe that life is worthwhile on the whole if naturalism were true. As I show, the same cannot be argued of theism. Since it is clear that the desire for life is not irrational, it is concluded that we have strong reason to prefer theism over naturalism.

Introduction: God and meaning in life

So I hated life, because the work that is done under the sun was grievous to me. All of it is meaningless, a chasing after the wind. (Ecc. 2:17)

The book of Ecclesiastes is well known for its bleak assessment of the human condition, and for repeatedly equating human life with mere ‘smoke’ (Heb. hevel), usually rendered as ‘vanity’ or ‘meaningless’ by English translations, and with ‘chasing after the wind’. The latter expression indicates that for the Qoheleth (‘Teacher’), the meaninglessness of life lies primarily in its futility. This is an odd claim: activities like chasing after the wind or trying to count all the grains of sand on a beach are said to be ‘futile’ because they inevitably end in failure. Specifically, they fail to achieve their desired ends. But virtually all of our goals – having a successful career, getting married, publishing books etc. – require us to be alive in order to achieve them. If so, how can the very act of living be futile? The Qoheleth offers the following answer:

‘What do people get for all the toil and anxious striving with which they labour under the sun? All their days their work is grief and pain; even at night their minds do not rest. This too is meaningless. (Ecc. 2:22–23)

In his estimation, all the goods that life has to offer are outweighed by the pain, sorrow, disappointment, and all the other evils that life throws at us (and in any case, our enjoyment of them does not last). In short, life is simply not worth the trouble and living is not a worthwhile project. It is in this sense that life is ‘meaningless’. However, the Qoheleth hints at various points that the futility of life may perhaps be only apparent, and that we may see why our lives have meaning after all, as long as we take God into account: ‘This too, I see, is from the hand of God, for without him, who can find enjoyment?’ (Ecc. 2:24).

The question of the meaning of life has long been thought to be closely intertwined with that of the existence of God. Theistic philosophers often argue that God, and God alone, can ensure that our lives are meaningful.¹ How God does this will depend on what a ‘meaningful life’ is taken to be. On one view, human life is meaningful as long there is an objective purpose (or telos) which human beings exist in order to fulfil, independently of our particular preferences and desires. To use Jean-Paul Sartre's (1965) well-known example, ‘the purpose of a paper-cutter is to write’ is made true by the fact that paper-cutters were intentionally designed for this reason. In contrast, it would be confused to say that the purpose of mountains is to look beautiful, even if they do in fact look beautiful, unless of course one believes that they were designed. Thus, William Lane Craig (2013, 163) argues that if ‘God does not exist, then you are just a miscarriage of nature, thrust into a purposeless universe to live a purposeless life’. Similarly, Thomas Morris (1992, 56) contends that ‘[to] have meaning of any kind, a thing must be brought under the governance of some kind of purposive intention’, and infers from this that human life could only have been endowed with meaning by a transcendent purpose-giver. Combine statements of this sort with the premise that life is meaningful (in the relevant sense), and we have an argument for theism.

But it is not obvious that having an objective purpose is either sufficient or necessary to living a meaningful life. Concerning sufficiency, suppose you discovered that you and all other human beings had been created in a lab by highly intelligent aliens, for the sole purposes of harvesting our faeces, which they consider to be a delicacy.² If this scenario were true, your life would certainly have an objective purpose, in the same way that pens, computers, and aeroplanes do. But intuitively, this does not make your life meaningful – if it is meaningful at all, it is in spite of the purpose for which you were made, not because of it. Concerning necessity, many theists would vigorously deny that God's life is meaningless,³ even though it necessarily could not be the product of intentional design.⁴

We may instead follow the Qoheleth in taking ‘meaningfulness’ to refer to the value or worth of a life. Stewart Goetz (2012, 3) similarly identifies ‘what makes life meaningful’ with ‘what makes life worth living’. In that case, theists can avail themselves of the various moral and axiological arguments on offer. John Cottingham (2005) contends that objective values cannot exist unless they find their source in God, the ultimate ground of value. It follows that life (or anything else) cannot be objectively valuable if atheism is true. Here the atheist may answer that the subjective value of life is all that matters to him: he can desire and enjoy life regardless of whether God exists. It is enough for his life to be meaningful, namely worthwhile, to him, given his goals, beliefs, and values. Whether or not life is also meaningful in some mind-independent sense is irrelevant. In addition, many atheistic philosophers have advanced and defended non-theistic accounts of objective value (e.g. Wielenberg (2009)).

I will present and defend a new argument from the meaning of life. The target of my argument will be metaphysical naturalism, the popular view that reality is exhausted by the natural order. I will argue that the desire to live is irrational if naturalism is true. Since the desire to live surely is rational, it follows that we ought to reject metaphysical naturalism. As I will show, traditional theism does not face this problem. And as should become clear, this argument will not at all depend on the claim that life must be meaningful or worthwhile in an objective sense, or that it requires a purpose in order to be meaningful. I will simply take the meaningfulness of a life to be its worthwhileness to the individual who lives it.

I begin by laying out my argument and motivating its premises, before addressing objections.

The argument

Once again, I take ‘metaphysical naturalism’ (or simply ‘naturalism’) to be the thesis that the natural order is all that exists, or at least that there are no supernatural beings. The question of the exact boundaries of naturalism is of course a matter of much controversy. For instance, it is unclear whether admitting non-natural (but not supernatural, since causally inert) entities like Moorean-Platonic ‘Goodness’, as done by some atheistic philosophers like Erik Wielenberg (2009) and David Enoch (2017), allows one to remain within the naturalist fold. Fortunately, I do not believe that the success of my argument hinges on how naturalism is to be defined.

I take a ‘rational’ desire to be a desire that it is rationally permissible to hold (that is, such that it is not irrational to hold it).⁵ The desire for life is ‘reflective’ as long as we have strong reasons to believe that life is worthwhile (more on this later).

1. (1) The desire for life is rational.

2. (2) The desire for life is irrational, unless it is reflective.

3. (3) If naturalism is true, the desire for life cannot be reflective.

4. (4) Therefore, if naturalism is true, the desire for life is irrational.

5. (5) Therefore, naturalism is false.

(4) is entailed by (2) and (3), as is (5) by (1) and (4). Therefore, motivating my argument will only require me to motivate (1), (2) and (3).

Motivating (1): the desire for life is rational

Desires, like beliefs, are intentional mental states, meaning that they represent states of affairs. On one view, a desire just is a kind of belief – specifically, a belief about what is good (cf. Gregory (2021) for a recent discussion). In any case, both beliefs and desires may be justified or unjustified, rational or irrational.⁶ The rationality (or justification) of both desires and beliefs is at least partly a function of their coherence with our other beliefs and desires. My desire to write this article and to send it for publication is, I take it, a rational one: given my other desires and background beliefs about what is valuable, it is reasonable for me to desire this. On the other hand, my desire for a cigarette is irrational, given my belief in the harmful effects of smoking and my desire to remain in good health.

The great majority of us desire to live. When asked whether we do, we reply in the affirmative. And we behave in such a way as to continue to live, generally avoiding decisions and lifestyles that are likely to result in the cessation of life. We are not indifferent as to whether or not we will continue to live. Some, such as Spinoza, have gone as far as to argue that desiring to ‘persevere in its own being’ (conatus) is part of the essence of any existing thing, including human beings (Della Rocca (2008), 145). We need not go that far, of course.

Nor does it obviously follow from the fact that all or most of us desire to live, that we are rational in so desiring. As G. E. Moore (1922/1903, 56) remarked to J. S. Mill regarding the latter's ‘proof’ of utilitarianism, ‘“desirable” does not mean “able to be desired”’, but rather ‘what ought to be desired or deserves to be desired’. As he goes on to say, the Prayer Book's talk of ‘good desires’ is surely no tautology (ibid.).

Even so, I do not expect many naturalists to take issue with (1). I take this to be a strength of my argument relative the other, related ones mentioned in the introduction: while naturalists can easily deny that human life has an objective purpose or objective value, they are unlikely to disagree that it is rational for them to want to live. Indeed, many such naturalists enthusiastically affirm their gratitude for their existence in the world, as documented by Yujin Nagasawa (2018).⁷ Hence I do not intend to motivate (1) in any depth, though I will offer the following two considerations in its defence.

First, it enjoys firm support from common-sense intuition. It just seems intuitively obvious that there is nothing irrational about wanting to live. We don't typically chastise one another for expressing this want, or for acting in such a way as to satisfy it (it wouldn't even occur to us to do so). Oppositely, we find expressions of suicidal thoughts deeply disturbing. A plausible explanation of this posits an intuition to the effect that it is rational to desire to live.

Second, its denial would severely undermine eminently plausible moral beliefs, such as our belief that we should try to protect people whose lives are endangered. If Bob's desire to live is irrational, it is hard to see why I ought to save him from drowning if I am able to (compare: I don't have a moral duty to make sure that Bob successfully counts every blade of grass on his lawn, even if he passionately desires this). One might cite the physical pain of drowning, or the distress that Bob experiences at the thought of losing his life, as grounds for my duty to rescue him. But this would fail to explain why that duty would surely still exist, even if Bob's imminent demise would somehow be painless, and if he were somehow completely unaware of it.⁸ The persistence of the moral duty to protect Bob's life, even in these circumstances, seems best explained by the rational permissibility of Bob's desire to live: since it is appropriate for Bob to want to live, it is right for me to assist him in satisfying this desire, insofar as I am able to.

Note finally that (1) need not be taken to mean that it is never irrational to desire to live, without exception. Naturalists will probably point to cases of extreme pain and misery, in which (they argue) it is no longer rational for the victim to wish to remain alive,⁹ against an ‘absolutist’ reading of (1). A more moderate reading, according to which it is generally rational to desire to live, barring unusually unfortunate circumstances, will suffice for our purposes.

Much more could be said in defence of (1), but I must move on to my argument's other premises.

Motivating (2): the desire for life is irrational, unless it is reflective

Worthwhileness

I have said that a life is ‘meaningful’ as long as it is worthwhile for the individual that lives it. Plausibly, the ‘worthwhileness’ of an activity is best thought of as its aggregate value, or the sum of the values of its benefits and costs.¹⁰ To say that a life is worthwhile or ‘worth living’, then, is to say that its benefits outweigh its costs – in other words, that being alive is a good thing on the whole. I should clarify that ‘good’ here need not at all be interpreted in an objective sense. It may be that value is entirely mind-dependent: for instance, if hedonistic accounts of value are true, my life is only worth living to the extent that it produces pleasure, and that the pain that it also inflicts is outweighed by the pleasure.

Some activities are such that it is quite easy to tell whether they are worth engaging in it or not. For example, sticking my hand in a fire ant nest is evidently not a worthwhile activity, given my knowledge of fire ant stings and my desire to avoid pain. Other activities are more difficult to evaluate in this way. If I am offered a new job, accepting the offer would mean acquiring new colleagues, a new workplace, a new schedule, engaging in new kinds of activities, and various other putative costs or benefits which I must carefully examine, before reaching a conclusion as to the desirability of the job.

It should be clear that in this respect, continuing to live is much more like acquiring a new job than like sticking one's hand in an ant hill. Life involves a wide array of costs and benefits, which are to be weighed up against each other, if one is to present a positive case for the worthwhileness of living.

Reflective desires

Could one put forward a cost-benefit analysis, yielding the conclusion that life's benefits outweigh its costs, and thus is desirable on the whole? I shall return to this question later. For now, let us say that my desire to live is reflective only if I have strong reasons to believe that life is worthwhile. More generally, my desire to engage in some activity A is reflective only if there are strong reasons to believe that A is a worthwhile activity.

If a desire for some state of affairs is reflective, having been found to be worthwhile by an individual, it is thereby rational for said individual to desire said state of affairs. The converse of this implication also has some initial plausibility: intuitively, an agent A's desire for some state of affairs S needs to ‘spring’ from A's knowledge (or reasonable belief) that S is worthwhile on the whole, in order to be rational.¹¹ For example, consider your desire for some fancy gadget. Suppose that this desire is in fact entirely the product of aggressive, manipulative advertising on the internet, resulting in an intense longing for the gadget. Should you discover, after prolonged soul-searching, that this is indeed how your desire came about, you would be likely to come to the conclusion that you ought to try to withhold the desire, if at all possible.

Why would you conclude this? Because you would have become aware that your desire for the gadget is not at all the result of your having any good reasons to believe that the item is indeed worth buying, and that there are in fact no such good reasons. You thus judge the desire to be irrational. Similarly, learning that, while we do desire life, we lack good reasons to believe that living is a worthwhile activity should lead us to conclude that wanting to live is irrational.

Note, importantly, that this would be different from acquiring good reasons to believe that life is disvaluable. It may be thought that this is what the above thought experiment really illustrates: since you have a limited amount of money, arbitrarily spending it on a trinket that you know you won't need would constitute a net cost, and hence it is irrational to desire doing so. But supposing that you do not know whether you will need the item, and you have no reason at all to believe that buying it will benefit you overall (or that it will cost you overall), desiring to buy the item would still intuitively be irrational. Thus, lacking good reasons to believe that some activity is worthwhile does appear to entail that the desire to engage in said activity is irrational.¹²

But there may be exceptions to this rule. Intrinsic (or ‘basic’) desires are such that the desired state of affairs is desired for its own sake, not by virtue of some other desired good. David Hume described intrinsic desires as ‘original existences’, holding that they could never be irrational, regardless of our background beliefs (cf. Parfit (1984) for a response to Hume, and Hubin (1991) for a defence) – they are ‘just there’, so to speak. Take the case of our desire for happiness, which is one of the most promising candidates for being an intrinsic desire. If asked why we desire it, we are likely to answer ‘I just do’. Because we do not derive this desire from other desires, we cannot present reasons for desiring it, but it surely does not follow that the desire for happiness is irrational.¹³

This caveat need not worry us, because our desire for life is very unlikely to be an intrinsic desire. For if it was, we would consider being plunged into a dreamless sleep for the rest of our lives (before finally ceasing to exist) to be at least somewhat preferable to being annihilated. But we do not, which strongly suggests that we desire life in virtue of desiring other things, such as happiness or well-being.

Thus it appears that if the desire for life were not a reflective desire, it would follow that the desire for life is not a rational desire; hence my argument's second premise.

Motivating (3): if naturalism is true, the desire for life cannot be reflective

Can we show that life is worthwhile?

Once again, our desire for life is reflective only if we can put forward strong reasons to believe that life is worthwhile. As I indicated earlier, such an argument would need to take the form of a cost-benefit analysis, yielding the conclusion that the benefits of living (happiness, meaningful relationships, virtue, etc.) are worth the costs of living (suffering, to others and oneself), and therefore that living is a worthwhile project.

Naturally, not everyone would be in a position to produce the required argument. Take the book of Job, whose eponymous protagonist tragically loses his property and children, and falls dreadfully ill. Though his story does end on a positive note, Job does not know this at this stage of the story. Given his limited evidence, he has no good reason to expect his condition to improve, and is hence in no position to argue that, all things considered, it is a good thing that he is alive. If anything, the opposite seems to be true: ‘May the day I was born be wiped out’ (Job 3:3).

Interestingly, the inability to show that life's benefits are worth its costs appears to be the rule, rather than the exception. Most human beings may not be as unfortunate as Job. Nevertheless, our lives, no matter how happy, are also filled with evils of various kinds. The third chapter of David Benatar's aptly named Better Never to Have Been begins with the question, ‘How bad is coming to existence?’ and proceeds to offer an extensive survey of human suffering (Benatar (2006), 61–92). ‘Dissatisfaction does and must pervade life’, he tells us, if only because some of our most important desires, such as wanting to stay healthy, young, and not to be separated from our loved ones will inevitably be frustrated by ageing and death. For other desires, like the desire for happiness or friendship, frustration is extremely common, and satisfaction is exceedingly brief and gives rise to other desires. Pain, discomfort, stress, boredom, and frustration permeate our lives, relationships, and work. And this is to say nothing of the dreadful tragedies faced by countless individuals, from rape and murder to crippling disease and suicide. Overall, he argues, ‘the quality of even the best lives is very bad’ (ibid., 61).

Benatar's ultimate goal is to show that coming into existence is always a harm. Since none of us choose to be alive, the question whether we are rational to desire our lives should be distinct from the question whether it might have been better had we never been brought into existence. In any case, my aims are markedly more modest: I only wish to argue that, for any given life, it is impossible to show that the benefits of life are worth its costs, and thus that living is a worthwhile project (that is, unless one is a traditional theist, as I will explain shortly).

In response to Benatar, Yujin Nagasawa (2008, 676) accuses him of failing to account for the ‘balance of pleasure and pain in life’ (emphasis in original). The pleasures (and other goods) we experience in life can compensate us for the pains (and other evils): ‘Most of us believe – reasonably, I think – that coming into existence is not a serious harm because, roughly speaking, the amount and quality of pleasure one enjoys in one's life usually compensates for the amount and quality of pain one suffers.’

I am inclined to agree that most of us regard the benefits of being alive as making up for its costs. However, for living to be a worthwhile project, it is not enough for its benefits to ‘compensate’ its costs, it must also outweigh them – in other words, being alive must constitute a net gain for the living being. For if the costs and benefits of being alive cancelled each other out, there would ultimately be nothing to gain from being alive, meaning that the state of affairs of being alive would not be desirable (though its opposite, i.e. not being alive, would not be desirable either). Indeed, the appropriate attitude towards a state of affairs whose benefits are equal to its costs is one of indifference or apathy, not desire.

Now, consider the three main theories of well-being, namely hedonism, desire theory, and objective list theory (to use the standard taxonomy, cf. Parfit (1984)). On hedonism, a life goes well to the extent that it contains more pleasure than pain. For any given life, and possibly barring extreme cases like Job's, it is practically impossible to determine whether it is more pleasurable than painful, largely owing to the immense complexity of our inner lives. One would need to examine carefully one's pleasurable and painful experiences, and establish that on the whole, the former last longer, are more frequent, and/or more intense than the latter. Our lives contain simply too many such experiences for this to be feasible. The same applies to desire theory, which states that well-being is entirely dependent on the satisfaction and frustration of desires: we cannot realistically assess whether our lives involve more desire-satisfaction or frustration. Both theories face the additional worry that our memories of past pleasures, pains, and satisfied or frustrated desires are likely to be distorted, making it difficult to draw a conclusion about the overall desirability of our lives.¹⁴

Objective list theories hold that a life goes well as long as it instantiates a certain number of objective goods (e.g. meaningful relationships, knowledge, virtue) and, presumably, does not instantiate too many objective evils (e.g. having a meaningless job, failed relationships, ignorance, vice, etc.). The main worry here is the is the incommensurability of the different goods and evils: if I have a very successful and meaningful career but all my relationships have ended in dismal failure, how can I argue that the former good outweighs the latter evil? It seems that I could not, because there is no standard method of measuring the value of different goods and evils against each other. I could of course claim to prefer having a successful career over meaningful relationships, but this would bring us back to desire theory and its associated problems. Even if one could devise a system that would ascribe a numerical positive or negative value to any objective good and evil (e.g. +10 for a meaningful career, −7 for a failed relationship), the issue of complexity would arise once more, as the goods and evils to be weighed up against each other would be extremely numerous.

Finally, our highly imperfect epistemic access to the future is a concern, regardless of which theory of well-being one opts for. We cannot rule out that we will be struck by sudden devastating tragedies, like Job. More generally, it is difficult to see how one could predict that one's life will contain more good than evil.

The general worry here is not that we cannot know with absolute certainty that our lives are (or will be) worthwhile. Rather, the worry is that there are no good reasons to believe that it is: for every instance of good in any given life (happiness, desire-satisfaction, success), there is a corresponding instance of evil (sadness, desire-frustration, failure), and it is practically impossible to show (to oneself or to others) that the former outweigh the latter. Therefore, it appears that the desire for life cannot be a reflective desire, in the sense given above – though this depends on one's background philosophical commitments, as I will now argue.

Theism and naturalism contrasted

Traditional theism (of which classical Christian theism is a subset) holds that the natural order owes its existence to the activity of a transcendent, self-existent, supremely good God. If I believe that such a being exists, successfully arguing that life is worthwhile would not at all require me to produce a cost-benefit analysis of life. It is sufficient to argue that a perfect being would not have brought human beings into existence, unless their lives were worthwhile. This is much easier than weighing up all the many goods and evils of being alive against each other. Consider this simple argument:

1. (i) God would have the means to ensure that human life is worthwhile.

2. (ii) If (i), God would ensure that human life is worthwhile.

3. (iii) Therefore, God would ensure that human life is worthwhile.¹⁵

The plausibility of (i) is manifest if one bears in mind that God's resources are not exhausted by worldly goods. Indeed, on the classical Christian view, the post-mortem enjoyment of an infinite supernatural good is the very purpose for which we were made. ‘What people are for’, as G. E. M. Anscombe (1993, 124) put it, ‘is – as guided missiles – to home in on God, God who is the one truth it is infinitely worth knowing, the possession of which you could never get tired of’. The potential for this ‘beatific vision’, entailing full communion with God and perfect joy, is what makes ‘every life, right up to the last, infinitely precious’ (ibid.). It is difficult to deny that being alive in the world would be desirable, if it really did provide one with an opportunity to achieve this supremely great good (as long as the probability that one actually achieves this is not exceedingly low). Whatever miseries we may suffer while physically alive would pale in comparison to eternal happiness and blessing. Not only this, the biological process of life, however painful at times, would be the process by which we are sanctified, that is, by which our character is formed and prepared for union with God.

This point is often made in discussions of what Stephen Maitzen (2009, 122) called the ‘Heaven Swamps Everything’ theodicy. But my present aims are more modest than those of a theodicy: I am not arguing that a heavenly afterlife would justify God in causing or allowing suffering, but rather that God, if he exists, could ensure that our lives are worthwhile, even in the midst of great suffering.¹⁶ Neither am I arguing that there are no other ways in which a perfect being could ensure that our lives are worthwhile (that is, other than by providing an opportunity for supreme happiness after death). Alternative ways may perhaps exist, but motivating (i) only requires me to show that there is at least one way in which he could accomplish such a feat.

As to (ii), a world in which human lives are worthwhile is intuitively greater than one in which they are not (all else being equal). Thus, a perfectly good being would prefer to actualize the latter possibility, and should be expected to act accordingly. Since (i) and (ii) jointly entail (iii), it follows that the theist has a valid and sound argument for the conclusion that life is worthwhile, meaning that the theist's desire for life can be reflective.

As God does not figure in his ontology, the naturalist obviously lacks access to this argument. But could he deploy an analogous argument showing that the naturalistic, evolutionary forces that shaped our cognitive faculties would only have given us an innate, distinctive desire for life if life was in fact worthwhile?

An evolutionary explanation of the innate desire is easily articulated: being innately disposed to desire life from an early age is conducive to survival and reproductive success. Organisms disposed to desire life (and thus, to fear death) are more likely to survive long enough to reproduce than those that do not, everything else being equal. So, the in-built disposition to desire life was selectively retained in our ancestral populations, and passed on to us through the standard mechanisms of genetic inheritance. Thus, evolutionary psychologist Geoffrey Miller (2007) speaks of the ‘the hardwired fears and reactions that motivate humans to avoid death’: ‘Suffocate me, and I'll struggle. Shoot me, and I'll scream. The brain stem and amygdala will always do their job of struggling to preserve one's life at any cost.’

Significantly, however, the nature of the evolutionary process is such that the desire for life would have been selected for, regardless of whether life is worthwhile or not. This is so on both subjectivist and objectivist accounts of value. For instance, take the simple hedonistic view that some activity is worthwhile just as long as it is pleasurable on the whole. The sad truth is that wanting to live would still be conducive to greater reproductive success, even if life invariably involved far more pain than pleasure. For even in this scenario, human beings who desired to live would fare better in the evolutionary race. The same holds if we take more objectivist positions, such as the view that ‘worthwhileness’ is an irreducible, mind-independent property: whether life instantiates this property or not is irrelevant to natural selection.¹⁷

The upshot of this is that holding irrational desires – specifically, desires whose objects are undesirable – may be an evolutionarily beneficial trait. There is hence a crucial asymmetry between theism and naturalism, insofar as theism gives us strong reasons to believe that life is worthwhile, whereas naturalism does not.

Conclusion: (3) is true

I have advanced a serious challenge to the claim that the desire for life can be reflective, and presented a solution to this challenge, which is compatible with traditional theism but incompatible with naturalism. The existence of God would give us strong reasons to believe that life is worthwhile. If instead naturalism were true, we would not be entitled to appeal to such reasons, hence (3).

Having motivated all key premises, I now wish to compare briefly my argument to another, related one.

Nagasawa's ‘Problem of evil for atheists’: a comparison

The reader may have noticed similarities between the above argument and Yujin Nagasawa's recent ‘Problem of Evil for Atheists’ (2018). Nagasawa argues that certain kinds of evil raise a difficulty for naturalists¹⁸ who hold to ‘existential optimism’ – that is, ‘that the world is overall a good place and that we should be grateful for our existence in it’ (ibid., 157). The cruelty and violence of natural selection threatens to make such optimism irrational. If my existence in the world is the end-product of a fundamentally evil system, which required the suffering and death of countless living beings in order to produce me, how could I possibly be thankful for it? And how could I claim that the world is ‘good overall’, if the natural laws that govern it are inherently oppressive?

Theists are comparably well equipped to deal with this threat, because theism implies that there is much more to reality than the natural world in its present state. If we think of the actual world as an enormous ‘spacetime worm’, the unsavoury aspects of reality (like natural selection and other ‘systemic evils’) need only constitute a tiny spatiotemporal slither of said worm. In particular, temporary evils, however horrible, may be vastly overshadowed by the existence of a supremely benevolent being and the possibility of a blessed afterlife in perfect communion with him, such that it would be reasonable to be grateful for one's existence in the world. Naturalists obviously cannot vindicate their optimism in this way, since their ontology includes only the natural world.

Existential optimism and the desire for life, while not identical, are closely related: one can hardly rationally desire to live while rejecting the proposition that it is good to be alive. Both my argument and Nagasawa's aim to show that given the extent of suffering in the world, a positive propositional attitude (i.e. a desire or a belief) towards life is irrational on naturalism.

However, since Nagasawa does not put forward any considerations in favour of existential optimism, his argument is not strictly against naturalism, but against the conjunction of naturalism and existential optimism. In contrast, my argument includes the crucial premise that the desire for life is rational (motivated in (3)) which, combined with the premise that the desire for life would not be rational given naturalism, yields the conclusion that naturalism is false. One could say that while both arguments try to show that naturalism commits one to a certain kind of pessimism about life in the world, only my argument draws the further conclusion that naturalism must therefore be false.

Moreover, it should be noted that the grounds for saddling naturalists with pessimism about life are different in both arguments. In mine, the worry arises from the practical impossibility of determining that life is worthwhile given naturalism, not from the systemic nature of natural evil.

My argument and Nagasawa's thus constitute distinct contributions to the field, despite their many similarities.

Objections

An objection to (2): elitism

My argument's second premise states that the desire for life cannot be rational unless it is reflective – that is, unless one has good reasons to believe that life is worthwhile. One could object that this is too restrictive, and that it implausibly rules out many patently rational desires. For example, very young children desire food, sleep, affection, play, etc. without giving any thought at all to the value or desirability of such things. Having a reflective desire requires the ability to represent reflectively their own desires to themselves, an ability that such children almost certainly lack. But it seems absurd to call these desires irrational. ‘Rational’ and ‘irrational’ are normative terms, and to call a desire ‘irrational’ is to say that there is a sense in which we ought not to have it. And there is surely no sense in which very young children ought not to desire the goods listed above.

In short, it could be argued that the reflectiveness requirement is unduly elitist, since it denounces the natural desires of very young children as irrational.

Since it is indeed implausible to claim that all (non-intrinsic¹⁹) desires must be reflective in order to be rational, the objection compels us to specify the conditions under which non-reflectiveness is problematic. Recall the example given above: you learn that your desire for the fancy gadget is the result of manipulative marketing, and this causes you to consider this desire as irrational. Note that in the example, your awareness of the true origins of your desire, and of the absence of reasons to regard it as a worthy object of desire, appears to play an important role. Intuitively, one cannot rationally continue to desire something after learning that there are no good reasons to regard said thing as valuable.²⁰ This seems to motivate the following principle:

Loss of Innocence (LI): If you know that you have no good reasons to believe that some state of affairs S is desirable, your (non-intrinsic) desire for S is irrational.

LI would explain why the non-intrinsic desires of very young children can still be rational, even if they are not reflective: since they give no thought to the desirability (or lack thereof) of the objects of their desires, they are unaware of any lack of reasons to believe that said objects are desirable. Should they begin to reflect on the value of these objects, they may come to realize that some of them are unworthy of being desired, in which case some of their desires would thereby become irrational.

However, LI is insufficient in its present form. Returning to our example, your desire for the gadget did not become irrational upon your discovery of its true origins. Rather, it was irrational all along; you simply became aware of its irrationality. This suggests the following modification to LI:

LI-2: If you know or are able to know that you have no good reasons to believe that some state of affairs S is desirable, your (non-intrinsic) desire for S is irrational.

LI by itself sanctions my desire for a cigarette, just as long as I am ignorant of the fact that I have no good reasons to value smoking, even though I have easy epistemic access to this fact. As such, LI is an overly permissive principle, hence the need for LI-2.

LI-2 is also capable of explaining why very young children can have rational, non-reflective (non-intrinsic) desires: they not only lack the knowledge that some of the objects of their desires may not be worthwhile, they are unable to acquire this knowledge, since they lack the relevant metacognitive capacities. In contrast, we adults are able to reflect on our desires and the value of their objects, and are fully aware that some desired states of affairs are not desirable.

Let us finally turn to our desire for life, and suppose that we have no reason to believe that living is a worthwhile project, that is, it is an unreflective desire. We are either aware of this, or not. If we are, then by both LI and LI-2, our desire for life is irrational. If we are not, we are still capable of becoming aware of this, assuming that we are not infants, since we are fully capable of understanding the considerations raised against the claim that we can show life to be a worthwhile project (see above). Thus, by LI-2, our desire for life would still be irrational. Whether we would know it or not, then, our desire for life would be irrational if unreflective, and premise (2) stands unscathed.

Another objection to (2): the benefits of desiring to live

A second way of responding to (2) appeals to pragmatic considerations. It is sometimes argued that, in cases where it is unclear whether a certain belief B is true, it may be rational to (as it were) elect to hold to hold B, as in Pascal's wager. Perhaps a similar principle applies to desire, and the naturalist's desire for life in particular: since it is unclear, given the evidence, whether or not life is worthwhile on naturalism, it may be rational for the naturalist to elect to desire to live.

To show that this choice is in fact rational, the naturalist would need to show that the likely benefits of desiring life would outweigh its likely costs, just as Pascal argued for belief in God. He could, for instance, argue that those who desire to live are thereby more likely to be happy, since they are alive and thus have what they wish. The obvious reply here is that the desire to live carries with it the fear of losing one's life in a dangerous world, and the sadness that comes with the knowledge of one's inevitable demise. Conversely, not desiring to live may deprive one of the happiness and gratitude for being alive, but also of the associated anxiety and sorrow. In each case, it seems impossible to show that the benefits outweigh the costs, for the same reasons that it was impossible to show that the benefits of being alive outweigh its costs (see above), chiefly, the sheer complexity of our emotional lives. But in that case, the naturalist cannot defend the desire for life on pragmatic grounds.

An objection to (3): perspectivism

My argument's third premise states that if naturalism is true, our desire for life cannot be reflective, since there are on naturalism no good reasons to believe that life is worthwhile. ‘Perspectivists’ could object that the truth or falsehood of naturalism is irrelevant, because reasons depend on perspectives, not mind-independent facts. To borrow an example from Bernard Williams (1991, 101), suppose Bill believes – and has every reason to believe – that the contents of his glass is gin and tonic, when it is in fact petrol and ice. Intuitively, he still has a good reason to drink his glass (assuming he enjoys drinking gin and tonic) despite the fact that doing so would make him seriously ill. Similarly, one could argue that the human desire for life can still be rational, despite the truth of naturalism, as long as human beings have compelling reason to believe that theism is true. For then, they could appeal to the argument above concluding with the proposition that God would ensure that our lives are worthwhile if he existed. It would hence be reasonable for us to believe that living is a worthwhile project. But this would be a case of us forming a reflective desire for life even while naturalism is true, against (3).

Needless to say (and notwithstanding the intuitive plausibility of perspectivism), this objection should be deeply unappealing to naturalists, for two reasons. First, it would require the concession that there are compelling reasons to believe that theism is true, and that naturalism is therefore false. Second, and even less attractively, they would need to accept that they themselves cannot have a reflective desire for life, and thus (given (2)) that their desire for life is not rational. This would amount to saying that (1), which claims that the ‘desire for life is rational’, applies to theists, but not to naturalists. Insofar as most naturalists are probably inclined to agree with (1), I cannot imagine that very many of them would consent to being excluded from its scope.

We may thus leave this objection to the side, and turn to the next one.

Another objection to (3): Kahane's problem

Guy Kahane (2018) has argued that the existence of God or any otherworldly good or realm is unnecessary to making our lives worthwhile, since sufferings undergone in this life could in principle be more than compensated for by an eternal, purely physical afterlife involving only worldly goods and pleasures (meaningful friendships, intellectually stimulating activities, exciting trips, etc.), whether or not God exists. In that case, it may be thought that the naturalist is no worse off than the theist, since both world-views are compatible with a blissful post-mortem existence, and thus with life being worthwhile.

However, recall that the motivation for (3), which states that the desire for life cannot be reflective on naturalism, is that if naturalism were true, we would have no reason to believe that life's goods would outweigh its evils. This is very different from saying that given naturalism, it is impossible that this would be the case. The latter is far more difficult to substantiate; fortunately only the former is required to motivate (3) (since for my desire for life to be ‘reflective’ just is for me to have strong reasons to believe that living is a worthwhile activity).

Now, there are indeed conceivable naturalistic scenarios in which human beings enjoy blissful afterlives – say, if immortal aliens somehow upload the consciousness of every human being into an unending ‘Virtual Reality’ world involving all sorts of pleasure.²¹ But there would be no grounds for believing that this sort of scenario will obtain, given naturalism – only grounds for believing that they could obtain.²² By contrast, I argued that there are strong reasons to expect the God of traditional theism, should he exist, to give us an opportunity for supreme happiness after we die. So, while both theism and naturalism would make possible the worthwhileness of lives, only on the former position would we be justified in trusting that they are actually worthwhile. I conclude that Kahane's problem does not apply to my argument.

Another objection to (3): life just seems worthwhile

Once again, a desire to engage in activity A is ‘reflective’ just as long as there are good reasons to believe that A is a worthwhile activity. I have argued that there are, on naturalism, no good reasons to believe that life is a worthwhile activity, and thus that the desire for life cannot be reflective on naturalism.

The naturalist could object that I have not in fact shown this. Instead, I have shown that there are (on naturalism) no good arguments whose conclusion is that life is worthwhile. Perhaps, life simply seems worthwhile to us, and this could, in itself, be a good reason for thinking that life is worthwhile, even in the absence of a cost-benefit analysis showing that life is worthwhile. After all, it is widely held that at least some beliefs can be justified simply in virtue of seemings, without the need to be inferred from other beliefs. Phenomenal conservatism states that one can justifiably believe that P just as long as it appears to one that P, unless there are defeaters for the belief that P (for instance, if it seems to me that there is a cat on the mat, my belief that there is a cat on the mat cat can be justified, provided that I have no reasons to doubt this belief). If so, then perhaps it is sufficient that life seems worthwhile to us. The ‘seeming’ in question would have to be an intuitive one (rather than, say, a visual seeming). My defence of premise (1) suggested that the rationality of the desire for life seems intuitively obvious to us, and that this is reason enough to accept the premise. If so, then why shouldn't the same be said for the worthwhileness of life?

Of course, the objection stands only if our intuitive conviction that life is worthwhile, namely, that its benefits outweigh its costs, is faced with no defeaters. It may perhaps be thought that the considerations raised in defence of (3) entail that there can be no such defeaters: due to the extreme complexity of our lives and incommensurability of many goods and evils, we cannot tell whether said evils outweigh the goods, and thus whether the initial optimistic intuition is false. But a defeater D for a belief that P need not show that P is false, in order to be effective. All that is needed is for D to constitute significant evidential support for not-P. For example, suppose I believe that it would be worthwhile to take on a new job that I have just been offered. Suppose a former employee assures me that the manager I would be working under is quite an unpleasant and incompetent person. This new information would not show that my initial belief was false – for all I know, the generous work benefits and the stimulating nature of the job more than make up for having a sub-par manager. Even so, it could still count as a defeater for my belief that taking the job is worthwhile, insofar as it significantly lowers the justification of said belief, which it may do without compelling me to hold the opposite belief (i.e. ‘taking on the job would not be worthwhile’). This defeater may or may not itself be defeated, depending on whether I can show that the costs of the job are outweighed by its benefits.

Now, take the fact that to be alive is to risk experiencing horrible tragedies, of the kind only partially enumerated above. This does not entail that life is not worthwhile, and that its evils are not outweighed by its good. But it is certainly evidence in favour of the latter proposition. The horrors that so many suffer, and the very real risk that everyone faces of suffering them, would be expected given the hypothesis that life is not worthwhile. This evidence seems sufficiently strong to count as a defeater, that is, to give us a reason to doubt our initial optimistic intuition about the worthwhileness of life, and to suspend judgment on the matter until we acquire reasons to believe that the goods of life are worth its evils. This, I have argued, is precisely what naturalists, unlike theists, cannot access.

Contrary to the objection, then, there is a defeater for the widespread belief that life's evils are outweighed by its costs. In fact, part of the point of my argument is to draw attention to this defeater, before showing that it cannot itself be defeated on naturalism. An intuitive seeming to the effect that life is worthwhile therefore does not remove the need for an argument showing that it is worthwhile. Without this, the intuition that life is worthwhile will remain faced with an undefeated defeater, in which case there is no good reason to hold this belief.

One last objection to (3)

Granted that there needs to be an argument showing that life is worthwhile, the naturalist may propose the following one:

• (1′) Our desire for life is rational.

• (2′) If our desire for life is rational, then life is worthwhile.

• (3′) Therefore, life is worthwhile.

(1′) is obviously identical to (1), so I cannot reasonably take issue with it. (2′) seems plausible enough, and entails (3′) when conjoined with (1′). Has the naturalist found an argument? If so, it would follow that the desire for life can be reflective on naturalism, against (3).

Unfortunately, (2′) is incorrect. For it could be that life is not worthwhile, even though we have strong reason to believe that it is, and are thereby rational in wanting to live. Since it is possible that the antecedent is true while the consequent is false, the implication in (2′) is false. To be sound, the argument should be reformulated as follows:

• (1*) Our desire to live is rational.

• (2*) If our desire to live is rational, we have strong reason to believe that life is worthwhile.

• (3*) Therefore, we have strong reasons to believe that life is worthwhile.

(2*) is far more plausible than (2′), but the change from (3′) to (3*) is not inconsequential. The reformulated argument simply tells us that there are reasons to believe that life is worthwhile, without telling us what those reasons are. If, as I have been arguing, naturalism entails that there are no such reasons, (3*) entails that naturalism is false. Thus, the naturalist cannot deploy the above argument without presupposing that my argument is unsuccessful (since if it is successful, the above argument would disprove naturalism). To object to my argument in this way would therefore be question-begging.

A tu quoque objection: heaven and hell

I have argued that given naturalism, there are no good reasons to expect the goods of life to outweigh its evils, and thus that our desire for life cannot be reflective if naturalism is true. By contrast, traditional theism opens up the potential for a state of perfect happiness after death, which would vastly outweigh whatever suffering we may have endured during our physical lives. One world-view is thus in a significantly better position to defend the worthwhileness of life than the other.

But the naturalist could charge that the asymmetry is illusory, for two reasons. First, consider the candid words of the apocryphal 2 Esdras:

Let the human race lament, but let the beasts of the field be glad; let all who have been born lament, but let the four-footed beasts and the flocks rejoice!

For it is much better with them than with us; for they do not look for a judgment, nor do they know of any torment or salvation promised to them after death.

For what does it profit us that we shall be preserved alive but cruelly tormented?

For all who have been born are involved in iniquities, and are full of sins and burdened with transgressions.

And if we were not to come into judgment after death, perhaps it would have been better for us. (7:65–69)

The author laments that the lives of animals are preferable to those of humans, precisely because our souls live on after physical death, and are thus potentially subject to a miserable afterlife. The naturalist need not go so far. But he can argue that, just as it is unclear whether worldly goods outweigh worldly evils, it is unclear whether otherworldly goods outweigh otherworldly evils. For many traditional theists, being alive gives one an opportunity to progress in virtue and holiness towards the supremely great good of eternal salvation. Conversely, it involves the risk of damning one's soul, through poor life choices and acquired habits. One the face of it, damnation seems at least as bad as salvation is good, in which case the cost and the benefit cancel each other out. If so, the traditional theist appears to be in no better position than the naturalist to show that life is worthwhile, since expanding one's ontology beyond the natural fails to increase the net value of being alive.

In response, I argued earlier that God, a supremely good being, would ensure that our lives our worthwhile if he existed. Now, if the risk of damnation entailed that life is not in fact clearly worthwhile (as per the above objection), it would not follow that God would not after all ensure that our lives are worthwhile. Rather, it would follow that contrary to traditional Christian teaching, theism is inconsistent with the potential for damnation. Thus, the objection suggests the following argument:

1. (iv) If God exists, life is clearly worthwhile.

2. (v) If there is a real risk of ending up in hell, life is not clearly worthwhile.

3. (vi) Therefore, it cannot be the case that God exists and that there is a real risk of ending up in hell.

Various theists have defended (vi). Universalists like Thomas Talbott (2014) and Daniel Howard-Snyder (2003) argue that all human beings will ultimately receive salvation, even if some will need to undergo a lengthy purgatorial experience first. Less radically, annihilationists (e.g. Swinburne (2003), 182)²³ contend that damnation amounts to annihilation after a certain period of post-mortem suffering, in which case the goodness of salvation would vastly exceed the evil of damnation, since eternal happiness is intuitively far better than the simple cessation of existence is bad. At most, then, the objection shows only that there are compelling reasons to deny that God would allow anyone to suffer everlasting torment in hell. It does not show that theism can give us no positive assurance that our lives are worthwhile.

In truth, I am unconvinced by (v) of the above argument. For it seems to me that life could be worthwhile even with the possibility of eternal suffering. Suppose, for instance, that damnation were significantly less likely than salvation, and/or significantly less unpleasant than salvation is pleasant. In that case, the potential rewards of being alive would be well worth its risks, as high as these would be. And it would not be difficult for the God of traditional theism to actualize this state of affairs.

[here]One possible rejoinder here is that a chance to gain infinite happiness, however great, could never outweigh the risk of infinite pain, however slight, owing to the nature of infinity. But this is simply implausible: suppose I could participate in a contest which I had a reasonable chance of winning, and such that the reward if I won would be an eternal, supremely joyful state, while the penalty if I lost would be a barely noticeable but unending itch on my toe. It seems absurd to say that the reward wouldn't be worth the risk of participating here.

One could also object that, just as we are incapable of knowing through introspection whether our earthly lives are more pleasurable than painful (see above), we are likewise in the dark about the moral state of our minds, and thus about whether we will receive salvation or damnation. Kantian worries about our inability to know the motives of our actions, and Freudian ones about the opacity of our mental states more generally, spring to mind. But in that case, I am in no epistemic position to argue that God will grant me eternal happiness, and hence that my life is still worthwhile despite its pains.

But surely I can know that a given project is worthwhile, even if I cannot know what the outcome of engaging in it will be, as when I apply for jobs or write and submit requested revisions to an article for a philosophy journal. If the desired outcome does not obtain, and an undesirable outcome obtains instead – for example, my application is rejected, and I experience sadness and a loss of self-confidence as a result – this would not entail that the project in question was not worthwhile, and that I was wrong to think that it was.²⁴ It is sufficient that, upon deciding whether or not embark on the project, I know that my chances of succeeding are reasonably good. These need not necessarily be higher than my chances of failure, as it is surely sometimes rational to embark on projects with a low chance of success (e.g. applying for a job with a hundred other applicants), but they need to be high enough to warrant the associated risk.

But, the objector insists, how can we know that our chances of receiving salvation are ‘reasonably good’ in this sense? I propose the following alternative to the above argument:

• (iv′) If God exists, life is clearly worthwhile.

• (v′) If there is a real risk of ending up in hell, and no reasonable chance of receiving salvation, life is not clearly worthwhile.

• (vi′) Therefore, it cannot be the case that God exists and that there is a real risk of ending up in hell, and no reasonable chance of receiving salvation.

• (vi′) entails that if theism is true, we can know that we have a reasonable chance of acquiring salvation, and thus that life is worthwhile, even if hell is a real possibility.

Thus in my view, the possibility of hell would not entail that life is not clearly worthwhile. But even if it did, this would not show that theism would fail to give us good reasons to believe that life is worthwhile – once again, it would simply show that theism and hell are incompatible. I conclude that the asymmetry between theism and naturalism with respect to the rationality of the desire for life still stands.

Can the naturalist turn the tables?

In order to show that the desire for life does not pose the same problem to theism, I defended the claim that ‘God would ensure that human life is worthwhile’ (see above). This contention may be thought to backfire severely against theism. Consider the following argument, suggested (though not defended at length) by Jason Megill and Daniel Linford (2016):

1. (I) If God exists, then all lives have meaning.

2. (II) There is or has been at least one human life that lacked meaning.

3. (III) Therefore, God does not exist.

Suppose that for a life to ‘have meaning’ here is for it to be worthwhile. If God's existence would guarantee that all human lives are meaningful in this sense, good evidence of a single meaningless human life would be sufficient to disprove theism. Hence, my argument appears to make theism vulnerable to an easy refutation.

However, recall that part of my motivation for the claim that God would ensure the worthwhileness of our lives was the claim that God could do so. As I argued, God's capacity to provide us with the opportunity to achieve eternal happiness in union with him entails that he is able, if he so wishes, to make it the case that even the worst human lives are worthwhile. Thus, being alive in the world would be desirable just as long as the probability that one actually achieves this is not too low, even if the quality of one's earthly existence is truly wretched. I then argued that if God could accomplish this, he would.

If so, and unless the naturalist is able to argue that this is not in fact the case, and that the opportunity for eternal joy would fail to make a miserable life worthwhile, the claim that there is at least one meaningless life (i.e. to claim (II)) would thus simply beg the question against theism.

Furthermore, the naturalist presumably accepts that some lives which contain a significant amount of misery would nevertheless still be worthwhile given a realistic opportunity for supreme, eternal happiness. In that case, it would seem strange and arbitrary for the naturalist to argue that there is a certain level of earthly misery, beyond which even the chance of attaining a blessed eternity would fail to make life worthwhile. In any event, the onus is on the naturalist to show that such a line exists, and that it is actually crossed by some human lives in the world.

Conclusion

I have presented and defended an argument against naturalism, from the rationality of the desire for life. The crux of the argument is that there are, on naturalism, no good reasons to believe that life is worthwhile, in which case it would not be rational to desire life. But since it surely is rational to desire life, it follows that naturalism must be false.

While the argument laid out in this article does not directly support theism (since it concludes with the falsehood of naturalism), it should be clear that it gives us strong reason to prefer theism over naturalism, insofar as no parallel argument from the meaning of life can be levelled against theism, as I have argued. I conclude that the meaningfulness of human life constitutes a significant consideration in favour of theism.