In this ambitious and thought-provoking book, Fiona Ellis argues for the stunning claim that there is a way to expand philosophical naturalism in such a way that it becomes compatible with theism, and that this way is, in fact, already implicit in the concept of (a particular version of) naturalism. While naturalism and theism are usually understood as opposing, mutually exclusive views, Ellis shows that we can (and ought to) have it all: ‘we should be naturalists’ (2) but nevertheless reject the view that ‘believers in God are intellectually challenged’ (3).

Ellis begins her argument by discussing what is probably the most well-known form of naturalism, i.e. ‘scientific naturalism’. According to this view, reality can be fully described using the language of the natural sciences (e.g. physics, chemistry, biology) only. Aspects that cannot be described in the language of the natural sciences are considered ‘supernatural’ in the sense of non-real, illusory, and intellectually indefensible. Ellis demonstrates that naturalism so understood is unreasonably strong, given that it discredits a whole range of important aspects of human life, most notably values, which influence and even guide human behaviour: ‘we are motivated by kindness, justice, and benevolence, appalled by cruelty and greed, and we shudder at the idea that someone could treat another person as a mere opportunity for satisfying their desires. It is in this sense that values make their demands on us and provide us with the relevant reasons for action, and it is in this sense that we come to find value in things, when, say, we are struck by the cruelty or kindness of a person or an act’ (66f). Given that values cannot be accounted for in scientific terms, especially not in cases where it is impossible to explain a particular value in terms of a satisfaction of interests, the scientific naturalist is forced to dismiss their relevance for an overall project of describing reality. This, Ellis argues, seems like an implausibly strong way of understanding naturalism.

However, many naturalists are aware of this problem and would not define their position in such extreme terms. Rather, ‘expansive scientific naturalists’ (Expansive Naturalism I) aim to formulate their view in such a way that it allows for the accommodation of objective human values as described by the human sciences (anthropology, sociology, psychology, etc.). Peter Railton, for example, whose works Ellis discusses extensively, holds that the moral dimension of reality can be explained in terms of the (non-moral) desires and interests we hold qua human beings.

Despite the fact that, in order to describe reality, the expansive scientific naturalist has a larger vocabulary at her disposal than the scientific naturalist, Ellis argues that also the expansive version of scientific naturalism does not have the means to capture the realm of human value adequately. This is because values can only be understood appropriately from an engaged, i.e. non-detached, point of view, whereas scientific inquiry implies thinking in abstract, disengaged, and reductive terms. Consequently, Ellis argues, ‘treating values as objects of detached observation … has the effect of removing them from the picture’ (44). This becomes clear, for example, when we think of the concept of moral obligation: a scientific (i.e. reductive) account of moral value may be able to explain value to some degree in terms of human well-being, but it will never be able to explain why we should indeed feel obliged to act accordingly (47).

At this point it becomes clear that (both natural and human) scientific investigation alone is insufficient to make appropriate sense of the normative force of human values, particularly of moral values. The discussion thus proceeds to yet another form of naturalism, ‘expansive naturalism’ (Expansive Naturalism II), a version of naturalism according to which values are part of the furniture of the world, even if they cannot be investigated by scientific means. Drawing on works by David Wiggins and John McDowell, Ellis argues that there is a way of thinking about moral properties as irreducible to scientific facts that doesn't make them seem ‘intolerably odd’ (51). Crucially, we have to understand moral properties as inextricably linked to the sentient beings conceiving of them, i.e. beings whose ‘moral sensibility’ can account for the normative force of moral properties. Moral properties are objective, but finding out about their existence requires the acquisition of an appropriate body of recognitional and conceptual capacities, for example through education, habituation, or training (62). In understanding moral properties in this way, ‘morality is humanized [and hence, naturalized] in the sense that the domain it involves is essentially within the reach of human beings. Our capacity to respond to this domain is a rational capacity which stems from our human nature rather than from any supernatural addition’ (63).

So it is possible to be a naturalist while at the same time conceding that the world is, at least to a certain degree, ‘enchanted’ (73) by values that cannot be reduced to scientifically measurable objects or event. The central move of Ellis’ argument is now to use this concession to argue that expansive naturalism can be ‘generalized in a theistic direction’, i.e. that ‘there is an intellectually respectable version of the claim that the natural world is divinely enchanted, that our responsiveness to such a world is continuous with our natural being, and that this responsiveness precisely does have a bearing upon our humanity’ (87).

Drawing on McDowell's notion of ‘second’, i.e. acquired, human nature, Ellis argues that we must reject the idea that God is entirely ‘other-worldly’ and embrace the idea that the supernatural is ‘a quality or dimension which enriches or perfects the natural world’ (91). This does not mean that we should think of God as one particular being amongst many other beings in this world, distinguished only by His immortality and superior powers (95), nor does it mean that we should conflate God with nature (pantheism) or with some remote and isolated causal principle. Rather, in agreement with Karl Rahner, Ellis argues that God must be understood in a way that does justice both to His radical otherness and to His intimate connection with the world (and us human beings).

This can be achieved once we understand the idea, due to Levinas, that we relate to God via human values, i.e. through our moral relations with other human beings. Our moral responsibilities towards other (human) beings open up the realm of moral ideals to us, inducing a desire for God in the form of a desire for (moral) goodness: ‘to know God is to know what is to de done’ (135). This seems to lead to the following dilemma: the idea that we relate to God by relating morally to other human beings can be seen as achieving the goal of naturalizing theism, but at the same time, it explains theism in such narrow terms that we end up with a picture that looks dangerously similar to a form of non-theistic expansive naturalism.

Ellis’ key move is to turn this objection against itself: instead of worrying that Levinas’ view ‘squeezes God out of the picture’, ‘we could conclude equally that the expansive naturalist has introduced Him − minus the name’ (174). The Christian God can then be understood as playing a morally motivating role (174), with Jesus being ‘the supreme instantiation of God's omnipresence to the world’ (164), an idea not to be understood historically (196) but rather, in the way we understand myths: as telling us something substantial about the world without being historically substantial events. Ellis thus argues for a view according to which ‘the world is irreducibly God-involving, but God is not reducible to the world’ (198), a way of understanding God that, as she sums up, combines ‘the wisdom of Plato and the sanity of Aristotle’ (204).

Ellis shows, in an impressively undogmatic way, that every naturalist who recognizes the relevance of values for human life, and who understands that values can only be fully understood in conjunction with the beings responsive to them, ought to be at least open to the idea that the world is not only morally but also divinely enchanted. A staunch theist may consider this result too weak: nothing in Ellis’ argument forces us to accept the divine as part of the world. A different way of looking at this, however, is to say that it is precisely this undogmatic stance that makes Ellis’ argument so convincing. Arguing for the intellectual respectability, rather than necessity, of seeing the world as divinely enchanted enables both the theist and the atheist to stay on board of Ellis’ argument until the end − one of the most remarkable strengths of Ellis’ book.

In light of this argumentative inclusiveness, on the other hand, one might wonder why the important question of how we are to make sense of religions other than Christianity does not receive more attention: are all religions, including all versions of Christianity, equally appropriate expressions (or ‘myths’) of the divine enchantment of nature? If so, why choose one religion over another? If not, where should we draw the line?

But perhaps Ellis does not need to answer these questions. Her explicit aim is to demonstrate that there is a philosophically respectable way to combine theism with naturalism: a very interesting result both for those who are ‘sensitive to morality, but not to God’ and for those who are ‘sensitive to God as well as to morality’ (6) but who think that precisely these sensitivities preclude them from taking up a naturalistic point of view. For Ellis, Christianity offers the best framework for expanding naturalism in a theistic direction, but nothing in her main argument hinges on the acceptance of this claim.

God, Value, and Nature thus achieves a remarkable goal: it offers reconciliation between theists and naturalists, yet without forcing either side to give up any of their dearest beliefs. Rather, she shows that their opponent's beliefs are already, to a certain extent, implicit in theism and naturalism respectively − with reconciliation being only one cooperative step away. As such, Ellis’ book is not only a fine example of first-rate philosophy, but also of a first-rate philosophical attitude.