My aim in ‘Christianity and the ethics of belief’Footnote ¹ was to show that the following claims are compatible: that it is always morally wrong to believe anything without sufficient evidence; that there is insufficient evidence for some religious propositions; and that it is sometimes morally permissible to take a religious stance that includes such unsupported propositions. I argued that they would be compatible if faith was not belief but commitment to a set of assumptions in the hope of salvation. I also argued that Christianity was implicitly committed to the first of the claims. In ‘The ethics of belief and Christian faith as commitment to assumptions’,Footnote ² Rik Peels denies the first claim, denies the Christian commitment to it, and denies that faith involves assumptions. He fails in his attempts to justify his denials.

The first claim is the central tenet of W. K. Clifford's ethics of belief.Footnote ³ Here is a very brief argument for it. The total losses from believing without sufficient evidence, or overbelieving, outweigh the total gains. Some overbeliefs are falsehoods that result in the overbeliever doing harm to others. Other overbeliefs may not result in actual harm but can still put others at risk. If there are benefits, they often accrue to the overbeliever himself while the harms and risks are borne by others. It follows that we ought to do something to reduce the actual and potential harm from overbelieving. The only way to reduce it is to reduce the number of pernicious overbeliefs. And the only way to reduce the number of pernicious overbeliefs is to try to avoid overbelieving altogether.Footnote ⁴

Obviously, doing nothing at all will not reduce the number. Moreover, every conceivable alternative to avoiding overbelief altogether is pointless because no-one can believe at will (where a person believes at will when he acquires a belief deliberately despite knowing that he has insufficient evidence for its truth). Since people cannot believe at will, no-one can recognize that he lacks sufficient evidence for p, determine correctly that believing p would nonetheless be harmless or beneficial, and thereafter acquire the belief that p. In other words, trying to distinguish good and bad overbeliefs will prevent us from overbelieving anyway, so there is no point in trying. Therefore, we must try to avoid overbelieving altogether. It is certainly possible to try to avoid it: people can improve their standards of evidence (and evidence acquisition),Footnote ⁵ thereby reducing the probability of their acquiring new overbeliefs in the future and eliminating past errors as they become aware of them.

Peels tries to eviscerate the ethics of belief with excuses (105). I did suggest that excuses are possible and I did suggest some possible excuses, but I did not commit myself fully to them. Peels distorts my opinion by making it seem more definite than it actually is.Footnote ⁶ His distortion is most extreme when he brings in the notion of a sensus divinitatis as an excuse for religious belief. There is not an iota of evidence that anyone has an innate belief in God. Moreover, Cliffordians find notions like the sensus divinitatis particularly objectionable because they regard them as illicit attempts to shift the burden of proof and because the attempts are often supplemented with ad hominem attacks: anyone who posits the existence of a sensus divinitatis is liable to infer that anyone who disagrees with him has an ulterior motive for denying the existence of God.

Clifford's insight that we should always try to avoid overbelieving is a philosophical discovery. There may be no justification for condemning people who lived before the discovery or who have not heard of it (despite their doing bad things out of ignorance) – just as there is none for condemning people for spreading bacteria if they lived before the discovery of antisepsis or if they have not heard of it (despite their doing bad things out of ignorance). I hesitate to say outright that they are excused because it did not take Clifford to tell us that there are cases in which people ought to know better. Be that as it may, explanation and education are probably the most appropriate initial reactions. Once it is known that overbelieving is risky or harmful, however, it is hard to excuse overbelievers – as hard as it is to excuse those who ignore hygiene and thereby put the health and lives of others at risk. It is possible, I still suspect, but nowhere near as easy as Peels makes it out to be.

Other moves Peels makes are also questionable. Contrary to Peels, there is no ambiguity in my position (98–99): it is always wrong to overbelieve in the light of objective standards. In fact, subjective overbelief is impossible unless you are unaware of it or you cannot immediately rid yourself of it when you become aware of it, so it is pointless to argue that it should be avoided.

Peels contends that I need it to be the case that we can weaken our own standards of evidence and that my view that someone who believes a proposition typically also has what he takes to be sufficient evidence for it precludes any weakening of standards (101). But, at most, he shows that someone could not weaken his standards of evidence through an act of conscious will. He does not show that a person's desires, say, could not corrode his standards without his consciously willing it. Even if his argument were better, it would be moot: in this paper, I have presented a streamlined argument for the ethics of belief that involves neither contention.

The fact that there are many religions that distinguish themselves from others doctrinally shows that, as a matter of logic, there must be a considerable number of religious propositions that are not supported by sufficient evidence. This is also true within Christianity with its many branches.Footnote ⁷ In order to minimize the impact of the ethics of belief on Christianity, I argued that we should construe faith not as belief but as commitment to a set of assumptions in the hope of salvation. The definition has a number of virtues. Among others, the terms are used in their ordinary senses and the concepts are familiar, yet the definition clearly distinguishes faith from belief. Also, it is easy to show that it is compatible with Clifford's views because Clifford allowed that we may act on assumptions.

Peels denies that faith could be a matter of assumptions (100–101) but his arguments are questionable. Contrary to Peels, making and acting on assumptions over a long period of time is not likely to lead to belief. Pascal's potential Christian wants to believe, works hard at it, and will probably fail anyway – I suspect that many lapsed Christians have tried the Pascal method without success. In contrast, a Cliffordian Christian would want to avoid becoming a believer. Now, people who have a policy of making and acting on the assumption that the simplest hypothesis is most likely to be true – because making and acting on the assumption enables them to explore the world most systematically – are never surprised when the simplest hypothesis has to be replaced. Yet they would be surprised at some point if assumptions were liable to turn into belief. There is no reason why a Cliffordian Christian could not be just as successful at avoiding belief.

Contrary to Peels, faith as assumptions does not involve circularity. It is logically possible that someone might hear the Christian story, decide that the salvation mentioned was worth seeking, and commit himself to the whole package, including the assumption about the purpose for making the assumptions. Peel's alternative description, which entails that this could not happen all at once, is question-begging. Peels also thinks that accepting such a package would be irrational because the existence of many other possibilities makes it improbable that it is true. However, it is my understanding that this is a point at which the grace of God is supposed to come into play, so there should be no problem from the Christian perspective.

It is true that I argued a number of years ago that religion would not be psychologically comforting if people did not believe and that its purpose is to provide psychological comfort.Footnote ⁸ But, for his charge of inconsistency to be a serious problem for me, Peels would also need to show that I could not modify the argument and say that religion would not be comforting if people either did not believe or did not make assumptions. Moreover, although the issue of psychological comfort arises in connection with an argument to the effect that people invent religion, Christians can counter that Christianity presupposes the view, on the ground that only people who tend to invent religion are liable to be receptive to revelation.

Finally, I argued that if Christians are commanded to love God and their neighbour, then Christians are committed to the ethics of belief, because overbelieving about metaphysical matters would increase the probability of believing things that might harm others. Peels thinks there is no problem because the beliefs are about different things, but our standards of evidence do not enable us to distinguish beliefs on the basis of their content. He also thinks that being a Christian might be preferable to avoiding overbelief when it comes to benefiting neighbours (102). That may be true but, to adapt one of Peels' analogies to my purposes, a dentist who did not use anaesthesia would be a sadist, however much he reduced the amount of pain experienced by his patients. Similarly, a Christian who put his neighbour at risk through overbelieving could not be said to truly love him, however much he might benefit him otherwise.

Contrary to Peels, you do not need to believe in God in order to love Him. If the love is the kind that can be commanded, there is no reason why the believer would be at an advantage over the maker of assumptions. Both can do what they take to be God's will. And if it is the kind that cannot be commanded, you have to have an experience that you take to be an experience of God. Otherwise, you can only love the idea of God. There is no reason why the believer should be better placed to have what he takes to be an experience of God than is someone who makes assumptions.

My aim in ‘Christianity and the ethics of belief’ was to show that it is possible to reconcile religion and ethically motivated rationality without denaturing either. Peels does not show that there are any substantive and irremediable problems with my arguments. He has not shown that there is no need for religion to conform to the ethics of belief or that it is impossible for it to do so.