Sceptical theism, divine commands, and love

Abstract

Sceptical theists respond to the problem of evil by arguing that we should be sceptical of our abilities to understand God's plan and the justifying reasons for his actions. A major difficulty faced by sceptical theism is the problem of moral paralysis. Some sceptical theists have proposed a divine command response: theists can appeal to God's commands in acting, and this circumvents the need to exercise value judgement in moral deliberations. This article provides an objection to the divine command response by arguing that it renders love impossible and practically undermines the possibility of the theistic way of life. As a result, this article demonstrates a constraint on any potential solution to the problem of moral paralysis in sceptical theism: the access to values of loving relationship and human well-being, as well as their role to play in agents’ deliberative process, should be safeguarded.

Introduction

Articles on sceptical theism often open with a case that consists of some horrendous evil, and this article is no exception. Consider William Rowe's fawn case:

Picture a fawn in a distant forest. A lightning strikes. A dead tree falls and catches on fire. The fawn is trapped, and it burns with the tree. It lies in terrible pain for a couple of days before it dies. No-one witnesses the tragedy.¹

Apparently, the fawn's death doesn't bring any greater goods or prevent anything worse from happening. Although the exact case might not have actually happened before, it is very probable that some similar event has once happened. If God exists, why would God allow such pointless suffering? Sceptical theists argue that though we perceive such suffering as pointless, there are justifications under God's consideration that are beyond our cognitive ken. As finite beings created by God, we should not expect ourselves to be able to understand God's plan and God's considerations.

Sceptical theists hold the view that the goods and evils we know of are not representative of all the possible goods and evils that God takes into consideration.² Michael Bergmann develops his sceptical theist position relying on three claims (Bergmann 2001, 279):

We have no good reason to think that the goods and evils that we can know of, as well as the entailment relations between them that are accessible to us, are representative of the goods, evils, and entailment relations between them that are accessible to God. Due to our wanting cognitive ability, we cannot infer from our not seeing the fawn's death bringing about any greater goods or preventing any greater evils that there are no sufficient moral reasons for God to allow it to happen.³ In general, we are never justified in claiming that there are no God-justifying reasons for God to permit horrendous evil in the world.

But scepticism about our cognitive faculties tends to generalize. Many worry that sceptical theism implies a form of moral scepticism. If the goods and evils and the entailment relations between them that we can know of are not representative, there seems to be no good reason for us to be confident in our moral deliberation. Some sceptical theists, in turn, have responded by appealing to divine commands. With God's guidance, we can retain our faith in our moral practice, despite being sceptical of our cognitive ability. Rather than being troubled by ST1–ST3, sceptical theists may think ‘God's commands provide all-things-considered reasons to act’ (Bergmann and Rea 2005, 244). In what follows, I provide a practical objection against the divine command response. I contend that it conflicts with the theistic teaching of love and threatens to render love impossible.

We begin by briefly discussing the moral scepticism objection and the divine command response. Then, I argue that not all kinds of divine commands would do the work that the divine command response needs. Next, I present my argument against the divine command response. I then turn to some potential objections. Lastly, I indicate some general lessons for any potential solutions to the problem of moral paralysis in sceptical theism.

Moral paralysis and divine commands

To start with, it would be useful to discuss briefly the moral scepticism objection and the divine command reply. The moral scepticism response argues that the value scepticism embodied by sceptical theism forces agents into a state of moral paralysis – a state of moral puzzlement that disables agent from acting morally. Consider the following case:

John, a sceptical theist, enters a room and sees his wife, Carol, painfully struggling on the floor. John realizes that she is having an asthma attack and is struggling to breathe. Her rescue inhaler is in the drawer and is immediately accessible to John.

What should John do? Any sensible person would say that the right thing for John to do is to immediately reach for the inhaler and try to save her. However, one important part of the story is that John is a firm believer in sceptical theism. He worries that behind his wife's suffering there are goods and evils that he is unaware – or unable to be aware – of. Just as in the fawn case John cannot claim to know a greater good would be served if the fawn were saved, he cannot claim to know that a greater good would be served if Carol were saved. His allegiance to sceptical theism prevents him from claiming knowledge of what course of action serves goodness the best, if he is consistent.

The value scepticism embodied by sceptical theism infects our everyday moral deliberation. When faced with a situation demanding moral judgement on the consequences of an action, a sceptical theist, if deeply committed to sceptical theism, may find their moral deliberation interrupted by the acknowledgement of their cognitive limitations in the value realm. In discussing whether a sceptical theist should intervene and stop a crime, Almeida and Oppy, two prominent proponents of the moral scepticism response, write:

But, if we could easily intervene to stop the heinous crime, then it would be appalling for us to allow this consideration to stop us from intervening. Yet, if we take the thought seriously, how can we also maintain that we are morally required to intervene? After all, as a result of our acceptance of ST1–ST3, we are allegedly committed to the claim that it is not unlikely that it would be for the best, all things considered, if we did not do so. (Almeida and Oppy 2003, 506)⁴

Sceptical theism not only leads to a state of moral paralysis in which agents cannot decide between intervention or inaction, it also poses a tricky problem to theists who follow the practical teachings of theism. Theists are taught to do good for others, and prevent or remove others’ suffering, but sceptical theism implies that, due to our poor grasp of value, we are unable to do so. It removes our rational warrant for following these central moral teachings in theistic religion. This amounts to an important objection to sceptical theism because it directly undermines the practical possibility of moral action – a consequence that is extremely hard for any theist to accept.

In a direct reply to Almeida and Oppy's criticism, Bergmann and Rea appeal to divine commands: sceptical theists, as theists, ‘very typically believe that God has commanded his creatures to behave in certain ways; and they also very typically believe that God's commands provide all-things-considered reasons to act’ (Bergmann and Rea 2005, 244). Granting that ST1–ST3 implies our limited access to the realm of value, Bergmann and Rea think theists can appeal to divine revelation and seek refuge in God's commands despite our cognitive limitation. After all, sceptical theism does not undermine our ability to understand and follow God's commands.

Ira M. Schnall, drawing on the medieval Jewish commentaries on the Bible and the Talmud, provides a similar response (Schnall 2007, 60–65). According to Schnall, the medieval people also ponder on the question of whether one should intervene and remove the suffering of a dying, injured, or ill person. Given that God is supremely wise and merciful, the suffering of the person is presumably consistent with, or even sourced from, the will of God – intervening may risk preventing some greater goods that are beyond our ken from obtaining. Schnall, following the wisdom of the medieval Jewish thinkers, believes that

once we know that God has given us permission to heal the sick . . . and that He has at least encouraged us to alleviate suffering, we are no longer plagued by the sceptical doubts as to whether saving lives and alleviating suffering is the right thing for us to do. (Schnall (2007), 61)

By consulting scriptures, we can acquire knowledge about what God permits and forbids us to do. Although sceptical theism questions our ability to make judgement in choosing the correct course of action, we can use the scriptures as a moral instruction manual to help make the judgement.

Let's call Bergmann, Rea, and Schnall's responses ‘the divine command response.’ In sum, the divine command response suggests that divine commands provide theists with direct moral guidance, including those who pledge allegiance to sceptical theism. Despite our limited access to the realm of value, we can seek moral instruction from scriptures or revelation. Sceptical theists, then, are no longer forced into a state of moral puzzlement and paralysis. In what follows, I provide an objection to the divine command response. I contend that it leads to the impossibility of love, a result that theists can hardly accept.

Divine commands and valuation

Before turning to our objection, I first argue that not all kinds of divine commands will do the work that the divine command response needs. On one level, the divine command response has its strength because it is largely consistent with widely held theistic moral beliefs and it circumvents the need of making value judgements by moral agents. However, precisely because the response only works if it requires no value judgement from the agents, the range of divine commands that the sceptical theist can appeal to is greatly restricted. To see this, let's try to apply the divine command solution to John's case.

John sees his asthmatic wife struggling painfully on the floor. He is deciding whether he should reach for the inhaler. How does the divine command response help? Suppose John is extremely familiar with the Christian scriptures, and he tries to conjure up the biblical verses that may help him to make the decision. Galatians 6:10 flits through his mind: ‘Therefore, as we have opportunity, let us do good to all people.’ John, with this verse in mind, would still face a problem of valuation because it requires John to evaluate what is good to all people. The divine command, if it is to avoid the sceptical theists’ moral puzzlement, must be concrete enough to bypass the need of value evaluation from the agents’ perspective and provide direct action guidance to agents.

How to identify useful divine commands from scriptures is therefore a difficult problem. Commands that require value evaluation are not at the sceptical theists’ disposal, so we have to find commands that are more concrete. However, as Stephen Maitzen points out, biblical commands are often too unspecific to be useful for sceptical theists (Maitzen 2007, 240–241).⁵ Maitzen's example is that the verses do not contain any specific commands on whether one should protect a child against abuse. Similarly, we can find no scriptures specifically prescribing what one should do when one sees one's partner struggling on the floor. The circumstances in which we find ourselves often require specific action guidance that are not present in the scriptures.

According to Maitzen, we can neither extrapolate God's other commands to find out God's thought about child protection because this presumes a substantial understanding of God's purposes from our perspective, an understanding that sceptical theism thinks unobtainable. Sceptical theism, as Erik Wielenberg powerfully points out, entails scepticism about divine assertions. We lack justification for believing that it is unlikely that God might make assertions untruthfully for some beyond-our-ken purposes (Wielenberg 2010). As a result, we are not justified to extrapolate other parts of the scriptures to find out God's commands on what one should do in these specific moral circumstances.

But consider Schnall's advice: God has permitted us to heal the sick and alleviate the sufferings of people. It seems that this does not require value evaluation and is general enough for us to apply in a wide range of circumstances. The problem with Schnall's response is that God's alleged permission for us to heal the sick and alleviate others’ sufferings is an extrapolation from discreet commands rather than a direct command that can be found in the scriptures. Schnall's argument heavily relies on the medieval commentaries, which contain interpretations of biblical verses. For example, Schnall argues that, drawing on Talmudic gloss, ‘Do not stand still over your neighbour's blood’ (Leviticus 19:16) is a divine command ‘telling us that we are not only permitted to save lives; we are actually obligated, or commanded, to save lives whenever we are in a position to do so’ (Schnall (2007), 61). However, the content of this interpretation is simply not found in the verse; to arrive at this not very literal understanding of the very literal verse requires the sort of insight into God's purposes that sceptical theists find problematic.

One way to get out of this problem is to appeal to other forms of divine revelation. Perhaps, the medieval commentators received direct revelation from God and infused God's commands into their interpretation of the ancient texts. Perhaps, sceptical theists are able to feel God's presence and his commands directly and act on them. This view is not implausible, for traditional theists often claim that they ‘hear’ God's voices in their prayers, and the authors of the scriptures are often claimed to be carried along by God's spirits as they wrote. For the divine command response to work, we need to appeal to some form of non-scriptural divine revelation in circumstances where the scriptures’ commands do not straightforwardly apply. Moreover, such revelation must provide specific instruction on what one should do, for it must bypass the need of value evaluation. Potentially problematic cases aside,⁶ this seems to be the best account the divine command response may take. Let's now turn to our objection.

Divine commands and love

Let us start by revisiting John's case. Suppose John feels God's presence and hears God's answers to his inner prayers: ‘save Carol.’ What would John's psychology be like in deciding to save his wife? As a theist who obeys God's commands, he decides to save Carol. But he does so only out of his respect for God's command and not for any other reason. Recall ST1–ST3. As a sceptical theist, he is not justified in claiming full knowledge of the values of the consequences of his action. John should believe that he has no good reason to think that the perceived goods he brings to his wife by saving her in the due action are a justifying reason for him to save her. Although God commands him to save her, God might do so out of some justifying reasons that are beyond John's ken. Then, the perceived goods that John is aware of do not play a justificatory role in his moral deliberation. Being commanded by God is his only reason to do so.

A question worth asking is: Is a loving relationship still possible with such psychology? Suppose John saves Carol according to God's instruction. She catches her breath and is sent to the hospital. When a friend remarks on how caring and loving John is, John responds very sincerely, with Carol's presence in the room: ‘I saved her because that's what God told me to. I'm not sure what good it will bring to her, but that's part of God's divine plan.’ We can reasonably expect Carol's disappointment. The disappointment arises from John's not saving her for her own sake or for the sake of their loving relationship. As the beloved one in a loving relationship, we expect ourselves to have an unmediated relevance in the lover's concern – the fact that I am in this relation with them should play a direct and special role in their deliberation. However, in John's deliberation, Carol plays no role – let alone a primary role. John saves her only out of a theological concern to accord with God's divine command. On the assumption that love requires taking one's beloved as a final goal in one's deliberation and action (an assumption that I will defend shortly), John does not save his wife out of love.

To make things worse, John can even add, ‘If God told me to do otherwise, I would leave Carol there and let her die.’ This kind of response would very likely break the loving relationship between John and Carol. John is completely prepared to stop committing to their relationship if God commands otherwise. Although God's commands might align with John's concerns for Carol, John, as a sceptical theist, should not act on his concerns for Carol, and should be prepared to act in a way that does not align with his concerns for Carol. This counterfactual capriciousness suffices to break Carol's trust in his love.

Even if John is unable to transform his psychology completely in order to accord with ST1–ST3 and the divine command response, the problem remains. Suppose John is motivated to act by his concerns for Carol. He would undergo a confusing moral phenomenology: He cares about his wife and wants to save her, but he realizes that ST1–ST3 suggests that he has no good reason to think that the well-being of his wife is a justifying reason for him to act on; he appeals to God's commands and understands that he should save Carol – not for Carol's sake but for God's sake. A form of moral disharmony thus arises. He acts with a divided psychology: what motivates him cannot be the reason for his action. As a sceptical theist, he must admit that the goods that he is aiming at for his wife cannot be a genuine reason for his action. However, love does not work this way. The lover's concern for the beloved should be part of the final goal of his reason to act. In a sense, when John is acting solely out of respect for God's commands, he no longer acts out of his love for his wife.

One might insist that John's love for Carol can still constitute a reason for him to save Carol.⁷ It is true that John, as we have just discussed in the last paragraph, may still be motivated by his love for Carol. He might even be motivationally prepared to save Carol even if it were true that God commanded otherwise. However, if John is serious enough about his allegiance to sceptical theism, this should not constitute a reason for John to do so. To act out of love is to act on one's direct concern about the beloved's interests, desires, and well-being, and as we have shown, a sceptical theist should regard these as inadequate reasons for his moral actions. The divine command response displaces all other reasons in John's deliberation process. Therefore, given John's allegiance to ST1–ST3, though John's love for Carol may remain as his motivation to save Carol, it could not be a reason for John to act on.

If John is a firm believer in sceptical theism, this problem infects many interactions between him and his wife. Suppose John is deciding whether they should have a child, or whether he should donate his left kidney to his seriously ill wife, or whether he should help clean the wound on his wife's knee after she tripped over the stairs. In all these cases, as a sceptical theist who consistently bears ST1–ST3 in mind, John can only decide, according to the divine command response, based on God's commands. Moreover, this also extends to non-romantic sorts of personal love, including parental love, filial love, and friendship. Then, the above problem is not a one-off issue, but a recurring problem throughout John's life, within his relationships with his partner, children, parents, and friends. This renders different kinds of loving relationships in John's life highly improbable.

In sum, it seems that sceptical theism together with the divine command response renders love psychologically impossible and the establishment and sustainment of loving relation highly improbable. This is due to two reasons. First, one is not allowed to take the welfare or well-being of the beloved, or the loving relationship itself, as one's final goal of one's action and concern. Second, it requires a kind of prepared capriciousness to act against one's concerns for one's beloved. Regardless of whether it is psychologically possible for the sceptical theist to transform their way of thinking and set of motivations to align with ST1–ST3 and the divine command response, love is rendered extremely difficult to maintain.

This is a devastating consequence for sceptical theism because love is one of the most important elements in theistic ethical teachings. Christianity regards love as the highest vocation of human beings: ‘Above all, love each other deeply, because love covers over a multitude of sins’ (1 Peter 4:8). The Hebrew Bible says, ‘you shall love your neighbor as yourself’ (Leviticus 19:18). In Hadith one finds, ‘None of you will have faith until he loves for his brothers what he loves for himself’ (Hadith 13). To love others is an embodiment of the theists’ religious beliefs and commitments: ‘No one has ever seen God; but if we love one another, God lives in us and his love is made complete in us’ (1 John 4:12); ‘You will not enter Jannah until you believe, and you shall not believe until you love one another’ (Hadith 13). It is therefore reasonable to say that love is a non-negotiable element of theism. Sceptical theism, combined with the divine command response, leads to an internal tension: on the one hand, we ought to love other people; on the other hand, the divine command response makes love psychologically impossible. If love is impossible, then it is impossible for us to fulfil the theistic duty to love others. This is a strong reason to resist the divine command response – it only solves the problem of moral paralysis at a much more expensive cost, namely the impossibility of love.⁸

Our objection substantially depends on two assumptions about the nature of love. First, personal love requires the lover to take the beloved as a final goal in their moral deliberation. Second, the love in theistic teachings indeed coincides with this kind of personal love. I will defend these two assumptions in the remainder of this section.

First, consider the nature of personal love. ‘What is love’ is an ancient philosophical question, and there have been many attempted answers since Plato. In contemporary discussion, the most common theories of personal love, according to Bennett Helm, can be categorized into three views: the Love as Union view, the Love as Robust Concern view, and the Love as Valuing view (Helm 2021). These theories disagree on what love consists in. There does not seem to be a consensus among philosophers on which is the correct theory, and I would not attempt to settle the debate here. Rather, I would show that all major theories of personal love support our first assumption.

According to the Love as Union view, love consists in the formation of a significant union, a ‘we’, where the distinction between the lovers’ interests dissolves, and they share cares, concerns, and emotional responses.⁹ This view goes beyond mere emotional attachment between the lover and the beloved, emphasizing the actual fusion of the lovers’ concerns and actions. The view sees love not as an attitude, but as a relation that the lovers stand in to each other. It requires that when I act out of any concern, I act not for my sake or my beloved's sake, but for our sake. The shared ‘we’ becomes the subject of the lovers’ shared concerns, and to retain the loving relation, the lover has to take the beloved's well-being and concerns – in a way that is fused together with their own well-being and concerns – as a part of the central focus of their deliberation.

Taking the interest and desires of the beloved as one's final goal is the most salient feature of the Love as Robust Concern view, for this is, according to this view, what love consists in.¹⁰ This view is well encapsulated by Gabriele Taylor:

If x loves y then x wants to benefit and be with y etc., and he has these wants (or at least some of them) because he believes y has some determinate characteristic ψ in virtue of which he thinks it worth while to benefit and be with y. He regards satisfaction of these wants as an end and not as a means towards some other end. (Taylor 1976, 157, my emphasis)

According to this view, for one to love another, one must regard the satisfaction of one's desires to bring about the beloved's benefits and to get together with the beloved as a final goal of one's deliberation. The beloved has to have an unmediated relevance in his concern that ‘guide[s] and limit[s] his conduct. What a person loves helps to determine the choices that he makes and the actions that he is eager or unwilling to perform’ (Frankfurt 1999, 129).

Lastly, consider the Love as Valuing view. According to this view, love is constituted by the way we value a person. It can be further divided into two views: the view that to love is to appraise the value of the beloved and the view that to love is to bestow value on the beloved. David Velleman, as the major proponent of the appraisal view, suggests that:

Appreciation for someone's value as a person is not distinct from loving him: it is the evaluative core of love. I do not mean that love is a value judgment to the effect that the beloved has final value as an end in himself. Love is rather an appreciative response to the perception of that value. (Velleman 2008, 199)

According to Velleman, to love is to perceive the final value of the beloved as an end in himself, and to value him ‘singularly’ in a way that shuns ‘any thought of alternatives’ (Velleman 2008, 200). This requires one to see the beloved as an end in himself that is infinitely valuable and cannot be compared with other goods. For such value is normative, the lover should have ‘an arresting awareness of [the beloved's] value’ and is ‘open to caring about [the beloved] in all sorts of ways’ (Velleman 2008, 201). So, it is clear that the appraisal view aligns with our understanding of personal love.

The bestowal view, on the other hand, suggests that to love is to bestow value on the beloved.¹¹ To love is not to respond to the value in the beloved, but to create value in the beloved, to have a commitment to one's beloved to treat them as an end in themselves and respond to their interests and concerns for their own sake. As Helm suggests, this view resembles the Robust Concern view in emphasizing one's response to the beloved's interests and concerns, but while the Robust Concern view thinks that love is constituted by this response, the bestowal view thinks of it as an effect of love (Helm 2021).

There are advantages and difficulties with each of these views, but they are beyond the scope of this article. What is important to us is that, although they disagree on what constitutes love, they agree that if one is in a loving relation with another, it involves, in one way or another, taking the beloved's interests and concerns to figure as a final goal in one's deliberation.

Second, consider the nature of theistic love. Our objection fails if theistic love does not coincide with ordinary personal love in its requirement to take the beloved as the final goal in one's deliberation. To begin with, many prominent interpretations of the scriptures on ‘love’ seem to support our understanding. C. S. Lewis, in The Four Loves, calls the highest form of love, the form of love that God commands, ‘charity.’¹² Charity does not ‘belittle the natural loves but indicate[s] where their real glory lies’ (Lewis 1960, 163); rather than eradicating the ordinary nature of love, it extends the scope of natural love to people that are ‘not naturally lovable: lepers, criminals, enemies’ and so forth (Lewis 1960, 177). The scriptures teach us to love unconditionally and not to discriminate. The difference between theistic love and natural love, according to Lewis, lies in the scope and conditions of love. Unlike natural love, theistic love ‘is wholly disinterested’, but like natural love, theistic love ‘desires what is simply the best for the beloved’ (Lewis 1960, 177).

To give another example, according to Eleonore Stump, Aquinas thinks theistic love is ‘primarily the love of persons’, and ‘the overarching genus of personal love . . . is friendship’ (Stump 2006, 27). Aquinas believes God is the perfect and highest goodness, and his divine goodness is reflected in every creature. By desiring the goodness in the creatures and in ourselves, we affirm God's goodness and love God in return. Aquinas suggests that ‘to love is to wish good to someone’, and ‘that which is loved with the love of friendship is loved simply and for itself’ (ST I-II, 26, 4, my emphasis). On this view, theistic love is coincidental with ordinary love. It requires one to act in a way that treats the goods of the beloved as the final goal, or else the result is that it ‘loses the character to true friendship’ (ST I-II, 26, 4). Given that these major interpretations of theistic love understand love in a way that aligns with our analysis, we have a strong prima facie case that theistic love indeed coincides with ordinary love in this respect.

The sceptical theists who disagree with my assumption may object that this argument from authority holds no weight against them. Lewis's, Stump's, and Aquinas's views on love ultimately stem from interpretations of scriptures. But the scriptures do not explicitly tell us whether theistic love coincides with our conception of ordinary love. Their views can only be arrived at by extrapolating other contents in the scriptures, for example, by analysing stories in the scriptures. However, this falls back on Maitzen's challenge: extrapolating other parts of scriptures to discover God's view on the nature of love requires an insight into God's ultimate purpose. If sceptical theism is true, theists cannot presuppose any insight into God's purposes behind the production of those parts of the scriptures. However, this does not leave our opponent in a more advantageous position. If one gives a non-ordinary meaning to an ordinary term used in the text, they bear the burden of proof. If one thinks ‘love’ in the scriptures has a non-ordinary meaning, one has to find scriptural evidence to support that view. Likewise, since merely by consulting the literal contents of scriptures would not settle this issue, they can only find evidence by extrapolating other contents in the scriptures – which, again, is a blocked route. So, our opponents encounter a difficulty in providing evidence for this response. Given the lack of scriptural evidence that suggest we should understand ‘love’ otherwise, the best way to interpret theistic love is to adopt the ordinary understanding of the word.

Some may raise the possibility that theistic love and ordinary love, though they largely coincide with each other, may come apart in extraordinary scenarios. For example, some life-saving treatment such as receiving a blood-transfusion may be at odds with the religious convictions of some people (e.g. Jehovah's witnesses). So, in some special cases, refusing to perform such life-saving treatments on one's beloved in face of their death may be considered, by the beloved, as an honourable and laudable thing to do. Even though he acts against his concerns for his beloved,¹³ the beloved may feel a sense of love and adoration for the lover because he steadfastly adheres to the principles that both of them hold sacred. This indicates that theistic love and ordinary love can diverge from each other in extraordinary circumstances.

Consider John and Carol's case again. Assume that both John and Carol firmly believe in sceptical theism. Then it seems she would understand John's choice and deliberation process. Perhaps, she even feels a sense of love and adoration for John because he steadfastly adheres to principles that both of them hold sacred, namely, sceptical theism (combined with the divine command response). And perhaps, this sacred kind of love is what the objection above points to. The problem with this view is that the moral paralysis problem not only happens in extraordinary occasions, but in a wide variety of ordinary moral circumstances. John deliberates in this problematic way when he is deciding whether he should help his mother up after she fell in the middle of the road, or whether he should lend a small sum of money to the close friend who is having a hard time, or whether he should speak against the injustice that his colleague encounters. In these everyday moral scenarios, the sceptical theist has to appeal to divine command and make his decision. If theistic love requires the divergence from ordinary love in all these scenarios, it simply implies that theistic love is largely separated from ordinary love – which is, as we have shown, an unlikely conclusion.

To summarize, I have argued that given that the sceptical theists can only appeal to concrete commands that bypass value evaluation, the divine command response prevents theists from taking the beloved as a final goal in their deliberation and action, thus undermining the possibility of love. I have also argued that theistic love coincides with ordinary love in requiring the lover to take the beloved as the final goal in their deliberation process. Given that love is a central teaching in theism, this provides a strong reason for us to reject the divine command response. We now turn to some potential objections.

Objections and replies

I will first provide a clarification, then I will consider and reply to two potential objections.

An important clarification about our objection is that it is not a general critique of divine command theory. In general, divine command theory does not restrict our access to the value realm. While the divine command response, within the context of sceptical theism, leads to the consequence that God's commands are the only legitimate moral reason in our deliberation, divine command theory can admit our personal concerns for values, along with complying with God's commands, as legitimate moral reasons. One's concern for the beloved can be part of the final goal of one's action. Indeed, God's teachings in the scriptures often only provide abstract guidelines for theists to follow and requires value evaluation and sincere value commitments from the theists. For example, Paul writes in Romans 12:9, ‘Love must be sincere. Hate what is evil, cling to what is good.’ Therefore, in addition to the problems introduced above, the divine command response also faces difficult interpretative issues in reconciling with the religious teachings because the sacred scriptures at many places seem to require theists to exercise their value judgements.

The first objection is about the divine commandment ‘Love your neighbour as yourself.’¹⁴ Some may think that this command is a permission from God for us to stick with our ordinary moral deliberation, in other words, God commands us to act in a way that takes our beloveds as final goals in our deliberation. If our assumption about the nature of theistic love is correct, this suggestion seems not to fall prey to Maitzen's worry about the identification of commands. ‘Love’ here just means ordinary love, and the whole commandment is directly read off from the scriptures and understood literally.

But this does not help to resolve the problem of moral paralysis. The problem of moral paralysis arises because if sceptical theism is true, then we have to admit that we have no good reason to trust our moral intuitions and reasoning. In moral deliberation, one typically aims to act in a way that best serves the good, or at least, in a way that serves the good sufficiently well. This is especially true when we are deciding what we should do with respect to our loved ones. However, merely by being commanded to love our nearest and dearest with our ordinary deliberation, we do not thereby arrive at the knowledge that goodness is best served in this course of action. Constrained by sceptical theism, we cannot obtain knowledge about God's purposes behind this commandment. Perhaps, by saving his wife out of his sincere love, John is committing a local evil that eventually contributes to the ultimate good that is beyond our ken. John cannot even be certain that this course of action is best for Carol, since there may be sorts of goods and evils for his wife that are beyond his cognitive ability. The sceptical theists do not know what good they are creating, and in what way they are contributing to best serve the divine good – not even whether their action would best benefit their beloveds. This doubt, again, disrupts our moral deliberation and leads us back to where we began.

The second objection. The problem of moral disharmony, the separation of motivation and reason, is not a new problem in ethics. Bernard Williams famously criticized utilitarianism for its impartial moral demand, as he believes it leads to an attack on integrity and alienation from self (Williams 1973).¹⁵ Peter Railton has attempted to respond by distinguishing the demands of consequentialism as a criterion of right action on the one hand, and as a decision procedure on the other (Railton 1984, 148–156).¹⁶ He argues that one does not have to adopt the consequentialist mode of practical deliberation in order to actually maximize consequence and act rightly. In other words, consequentialism can remove itself from the scene while still working as the criterion of right action in the background. Can a sceptical theist provide a reply that is parallel to Railton's response? God's commands are only the criterion of right action, but not the practical guide to action. John can deliberate according to his concerns for Carol, while still fulfilling God's commands.

My response is twofold, corresponding to two understandings of the nature of divine commands. If theists understand the very state of affairs of obeying God to be constitutive of the fulfilment of a divine command, divine commands simply cannot be removed from our practical deliberation process. Obeying consists of a mental component in which one has the attitude to be subject to God's commands, and hence one cannot obey God's commands indirectly, without subjecting one's deliberation process to God's commands. If theists think God's commands only require one to comply with them behaviourally, a Railtonian sceptical theist can still only act in a state of bad faith. First, there are no good reasons for the sceptical theist to believe that his moral deliberation would align with God's commands, if he does not constantly consult them. Second, this falls back on the moral puzzlement problem, for the sceptical theists must acknowledge the unreliability of their moral deliberation. If the sceptical theist continues to act on the goods that he conceives of as reasons, this only implies that he is not serious enough in acknowledging the truth of ST1–ST3. Either way, the Railtonian response does not work for sceptical theism combined with the divine command response. So, the second objection also fails.

Conclusion

This article has rejected the divine command response that attempts to save sceptical theism from the charge of moral paralysis. We have shown that for the divine command response to work, sceptical theists can only appeal to concrete commands that do not require value evaluation from their side. However, the concrete divine commands, as a moral guide, displace other reasons in the agent's deliberative process. As a result, this undermines the psychological possibility of love, and thus conflicts with the central theistic teaching that we ought to love other people.

The foregoing discussion indicates a more general lesson for any potential solutions to the problem of moral paralysis in sceptical theism. As argued in this article, the two elements in sceptical theism puts a double constraint on any potential solutions: (a) regarding the sceptical element, the solution should avoid requiring agents to exercise value judgements; and (b) regarding the theistic element, the access to values of loving relationship and human well-being, as well as the role they play in agents’ deliberative processes, should be safeguarded by the solution. However, this dual constraint has an inherent tension because (b) seems to require the exercising of certain kind of value judgement that (a) seeks to avoid. While this article remains open on whether the problem of moral paralysis can be tackled by other responses, any promising attempt to solve the problem should work under this double constraint. The inherent tension, nonetheless, shows that this task is extremely hard to accomplish.