Dombrowski's ontological argument

Anselm's argument may be characterized as the claim that God exists necessarily, since God would not otherwise be ‘that, than which nothing greater can be conceived’.Footnote ³ Gaunilo of Marmoutier objected that the argument fails because we cannot conceive of God. This is because we do not understand God's nature and, since God is unique, we cannot infer this from the nature of some other reality (Gaunilo (Reference Deane1962), 306–307). A similar objection is found in Aquinas (2008, I, 1:2), in response to whom Dombrowski remarks that the ontological argument requires only ‘a comprehension of the abstraction that God is that than which no greater can be conceived: no comprehension of the essence of God beyond this level of understanding is required’ (Dombrowksi (Reference Dombrowski2006), 19). Dombrowski pursues a related objection, however, rejecting the ontological arguments of Thomas Morris, Katherin Rogers, and Alvin Plantinga on the grounds that the ‘classical’ conception of God with which they begin is incoherent. Dombrowski argues that the religious tradition has privileged one list of divine attributes without good reason; for example, a God whose omniscience entails detailed knowledge of the future and whose omnipotence entails that creatures have no power is difficult to reconcile with human freedom.Footnote ⁴ On his own ‘neoclassical’ or process view, there is no conflict between divine and human knowledge and power since the divine and human can work together: ‘God is seen as possessing maximal power that is compatible with the other divine attributes, especially omnibenevolence, but not as possessing omnipotence if this means that other existents ultimately have no power of their own’ (ibid., 141).

Gaunilo is, however, better known for highlighting the second difficulty, i.e. that Anselm's argument implies that there is no distinction between having an object in the understanding, and understanding that it is instantiated. This is illustrated by the famous reductio ad absurdum: just as we cannot prove the existence of an island which is ‘more excellent than all other countries’ by arguing that it is more excellent to exist in reality than in the understanding alone, so we cannot, by similar means, prove the existence of a being than which nothing greater can be conceived (Gaunilo (Reference Deane1962), 304–309).

As Dombrowski (Reference Dombrowski2006, 122) has pointed out, Gaunilo's ‘perfect island’ objection is not unrelated to Kant's famous objection to Descartes's version of the argument that ‘being’ is not a real predicate, and that there is therefore no difference between the concept of 100 existing thalers and the concept of 100 merely possible thalers (Kant (Reference Kant1929), 504–506). Dombrowski suggests, however, that the existence of both islands and thalers is contingent whereas, for Anselm and Descartes, the existence of God is necessary. Contingent existence might – or might not – be a predicate, but necessary existence necessarily is (Dombrowski (Reference Dombrowski2006), 273). Much subsequent discussion has centred on the question of whether the nature of God's existence is such that, for God at least, non-existence is inconceivable. Any modern version of the ontological argument must therefore endeavour to show how this can be so.

Dombrowski attempts to formulate a version of the ontological argument which avoids the difficulties outlined above. The fundamental premise of his argument is similar to that of Norman Malcolm's, i.e. the assertion that, since the existence of a perfect being cannot be contingent, it is either impossible or necessary (Malcolm (Reference Malcolm, Taliaferro and Griffiths2003), 273–275). Malcolm claims that God's existence is not impossible on the grounds that the concept of God is neither self-contradictory nor logically absurd; God's existence is therefore necessary – the only remaining possibility (ibid., 275). While it is not impossible for us to think of God as existing, it could, however, be argued that it is neither contradictory nor absurd for us to think of God as not existing – as Gaunilo (Reference Deane1962, 304, 310) pointed out.Footnote ⁵ Plantinga offers a more complex version of Malcolm's argument. For Plantinga, it is possible that there is a being with maximal greatness, i.e. one which has maximal excellence in every possible world. Since our world is a possible world, it must contain a being with maximal greatness (Plantinga (Reference Plantinga, Peterson, Hasker, Reichenbach and Basinger2007), 181–193).Footnote ⁶ Plantinga nevertheless acknowledges that only someone who accepts his main premise – that it is possible that there is a being with maximal greatness – would accept his argument (ibid., 193). This seems to require an independent proof of God's existence.

Drawing heavily on the work of Charles Hartshorne, Dombrowski argues that God's existence is either impossible or necessary and that, since other arguments for God's existence, especially the argument from religious experience, show that God's existence is not impossible, God's existence must be necessary (Dombrowski (Reference Dombrowski2006), 96–97). For Dombrowski, it is not the case that God's existence is necessary only if God exists (ibid., 126). The necessity in question is unconditional de re necessity; if we have a coherent concept of God, the nature of that concept entails its instantiation in any possible world. Dombrowski claims that, unlike other versions of the argument, his own version does not attempt to move from an abstract concept to a concrete reality. Instead, it moves from an abstract concept to the abstract conclusion that God exists necessarily in some concrete state (ibid., 86). In other words, Dombrowski acknowledges that we cannot affirm the existence of a clearly defined object; we can affirm only that some kind of God exists necessarily. We can know that God exists, but we cannot know exactly how God exists.Footnote ⁷

Yet again, however, the argument seems to fall into the trap encountered by Malcolm and Plantinga. Dombrowski attempts to provide a basis for the claim that God's existence is not impossible – i.e. an argument from religious experience – but, given the well-known objections to this argument (concerning the reliability of testimony, the conflicting claims of believers from different traditions, the cogency of alternative explanations for alleged experiences of the divine, and so on), it appears to offer little support for such a claim. Further, even if the argument from religious experience shows that the existence of God is not impossible, the desired conclusion is reached only if one allows that divine necessity is the only alternative. Dombrowski attempts to support such a view by suggesting that, if the existence of God is not impossible, then the idea of God is necessarily instantiated, and ‘an idea is an experience of a certain kind, with the idea of perfection being a very particular sort of experience that has to be taken seriously’ (ibid., 127).Footnote ⁸ This seems to imply that there is nothing beyond the experience of the concept with which the concept corresponds, however, an implication which, elsewhere (e.g. ibid., 56), he clearly rejects.

A second difficulty for Dombrowski's position lies behind his claim that the nature of the concept of God entails the instantiation of that concept in any possible world. If the concept of God is such that we are unable to think of it as not instantiated, the supporting argument from religious experience is rendered superfluous. If the concept of God, by its very nature, must be instantiated, we need no longer be concerned with the question of whether its existence can be shown to be possible. Nevertheless, questions remain about the nature of that concept and whether it is necessarily instantiated, and these are, perhaps, addressed more effectively by Murdoch's version of the ontological argument.

Murdoch's ontological argument

Murdoch, like Dombrowski, has reservations about the concept of God as it has often been understood, although her reasons for rejecting it are different. She argues that belief in a personal God is untenable, partly because she thinks that scientific developments have made it more difficult for us to accept the supernatural elements of religious belief (Murdoch (Reference Murdoch1992), 159, 459), and partly because she thinks that belief in a personal God offers us false consolation in the face of life's difficulties (ibid., 106, 119–20, 442);Footnote ⁹ the truly moral person must confront the terrible reality of suffering and act well without hope of reward – she must be good ‘for nothing’ (Murdoch (Reference Murdoch1985), 71).

Given the egocentric nature of the human psyche (ibid., 51), this is very difficult to achieve. Following Plato, Murdoch therefore recommends contemplation of the Good, a transcendent reality which can help us to see other people and their situations as they really are, unclouded by our own illusions, and to act appropriately; the Good is ‘an active principle of truthful cognition and moral understanding in the soul … a “reality principle” whereby we find our way about the world’ (Murdoch (Reference Murdoch1992), 474).Footnote ¹⁰

The Good shares many of the attributes traditionally ascribed to God, defined by Murdoch as a ‘single perfect transcendent non-representable and necessarily real object of attention’ (Murdoch (Reference Murdoch1985), 55). She rejects the view that we should retain the word ‘God’ to refer to a non-personal ‘symbol of an absolute demand … a traditional image of commitment to religious values’ (Murdoch (Reference Murdoch1992), 327, 143–144), however. For Murdoch, ‘“God” is the name of a supernatural person’ (ibid., 419) and the use of this word to refer to any other kind of spiritual reality leads to confusion. Nevertheless, while the Good is not to be understood as ‘God in disguise’, it is ‘what the old God symbolised’ (ibid., 428). It is therefore perhaps not surprising that Murdoch appeals to what she regards as a version of the ontological argument to support her notion of the Good.

Like Malcolm, Hartshorne, and Dombrowski, she identifies two forms of the argument in Anselm's Proslogion. The first is the argument from perfection of chapter 2 in which Anselm maintains that, since God is ‘that than which nothing greater can be conceived’ and it is greater to exist in reality than in the mind alone, God must therefore exist. The second is the argument from necessary existence from chapters 3 and 4 of the Proslogion and the reply to Gaunilo in which, according to Murdoch, Anselm claims that he is ‘not speaking of something which is, or happens to be, greater than all other beings, but of something than which a greater cannot be conceived, and whose non-existence is impossible’ (1992, 395). Murdoch sometimes refers to these arguments together as Anselm's ‘first’ or ‘logical’ argument (e.g. ibid., 404).

Murdoch notes Malcolm's claim that Kant was mistaken in understanding the proposition ‘God is a necessary being’ to be equivalent to the conditional proposition ‘If God exists then He necessarily exists’, which implies that it is possible that God does not exist (Malcolm (Reference Malcolm, Taliaferro and Griffiths2003), 278, quoted in Murdoch (Reference Murdoch1992), 409). She also notes Hartshorne's claim that ‘We should, instead, say “If the phrase ‘necessary being’ has meaning, then what it means exists necessarily and if it exists necessarily then a fortiori it exists”’ (Hartshorne (Reference Hartshorne1941), 312–313, quoted in Murdoch (Reference Murdoch1992), 409). For Malcolm and Hartshorne, the nature of God entails that God is not like Kant's triangle, the possible existence of which can be denied without contradiction. The idea of necessary existence is uniquely God's property, and it is not relevant to point out that ‘in other cases (triangles or islands) existence is not a predicate’ (Murdoch (Reference Murdoch1992), 409).

Thus, according to Murdoch, a reformulation of the hypothetical interpretation of the argument from necessary existence yields the argument that, ‘if the concept of God is meaningful (not self-contradictory) God must necessarily exist’ (ibid., 410). But, she asks, what counts as meaning? Is the concept merely in intellectu? She notes that ‘[o]f course the word “God” means something, sentences containing it are usually not nonsense, the concept has a history open to believers and unbelievers’ (ibid.). Murdoch rejects Malcolm's view that the concept of God is necessary only within the language-game of religious belief, however. This, she believes, is ‘a wrong turning’, ‘a perilous excursion’ (ibid., 413); Malcolm rescues the ontological argument from the criticisms of Kant, ‘but is not the argument which is thus rescued still an empty one with merely grammatical merits?’ (ibid., 412).

Instead, Murdoch holds the view that the concept of God is connected with absolute moral values, values which cannot be ‘thought away’ from human life (ibid.), and this is supported by a supplementary argument in Anselm, ‘a metaphysical argument which is also an appeal to experience’ (ibid., 395). Murdoch acknowledges that both Malcolm and Hartshorne offer, although without significant discussion, ancillary arguments from experience (ibid., 408). Murdoch rejects Malcolm's argument that it is the experience of overwhelming guilt, giving rise to the need for forgiveness, which has led human beings to form the concept of God, however, on the grounds that the extreme state of mind which Malcolm describes is not experienced by all human beings, and that there is no guarantee that forgiveness is available. By contrast, her own argument appeals to experience which is available everywhere, to everyone. The origins of her argument may be found in chapter 8 of Anselm's reply to Gaunilo, which, she says, supports the ‘logical’ argument that that which is perfect, the ‘object’ of our best thoughts, must, in some sense, exist,Footnote ¹¹ by claiming that:

Murdoch notes that this argument from moral experience, sometimes referred to as the ‘degrees-of-goodness argument’, may also be stated as ‘a more homely ad hominem appeal to our sense of God (Good) as discovered everywhere in the world’ (Murdoch (Reference Murdoch1992), 404–405). In the experiences of everyday life, ‘we are continually shown the reality of what is better and the illusory nature of what is worse. We learn of perfection and imperfection through our ability to understand what we see as an image or shadow of something better which we cannot yet see’ (ibid., 405). This is sometimes referred to as the ‘ubiquity of goodness argument’. For Murdoch, our consciousness of failure can also be a source of knowledge: ‘We are constantly in the process of recognising the falseness of our ‘goods’, and the unimportance of what we deem important … We find out in the most minute details of our lives that the good is the real’ (ibid., 430).Footnote ¹²

Thus, Murdoch argues that the ontological argument might enable us to claim a ‘uniquely necessary status for moral value as something (uniquely) impossible to be thought away from human experience, and as in a special sense, if conceived of, known as real’. This, Murdoch thinks, ‘might be associated with concepts of religion which reject a personal God and other supernatural beliefs’ (ibid., 396). Although she does not wish to offer a new interpretation of the word ‘God’, she does want to preserve ‘religion’, defined as ‘a mode of belief in the unique sovereign place of goodness or virtue in human life’, as she thinks that there can be ‘a religious outlook, religious preoccupations, a religious psychology which is detachable from dogma’ (ibid., 426):

For Murdoch, Christ is ‘a personal Ontological Proof’ who is ‘partly a creation of art’ in that he is ‘a compact of everything we know about goodness’; he points towards ‘the mystery of good’ (ibid., 429), although he is not, himself, perfect.Footnote ¹³ She suggests that Christ might be seen as occupying a place within Christianity which is analogous to that of the Buddha, thus enabling Christianity to preserve and renew itself by changing into something which can be believed, as it has always done (ibid., 419).

In summary, then, Murdoch has argued that Anselm's ‘logical’ argument of Proslogion, chapters 2–4 shows that the Good exists necessarily, and that we can derive from chapter 8 of his reply to Gaunilo two arguments which support this: first, the ‘degrees-of-goodness argument’ shows that we can deduce from our experience of imperfect goodness the existence of something better, of which we are as yet able to see only an image or a shadow. Second, the ‘ubiquity of goodness argument’ suggests that experience shows us ‘the uniquely unavoidable nature of God (Good …), its omnipresence, its purity and separateness from our fallen world, in which its magnetic force is everywhere perceptible. God either exists necessarily or is impossible. All our experience shows that he exists’ (ibid., 405).

Objections to Murdoch's ontological argument: Heather Widdows

It was suggested earlier that Dombrowski's argument is no more successful than earlier versions of the ontological argument because he relies on support from an argument from religious experience which that argument is unable to provide. At first sight, Murdoch's argument for the reality of the Good seems less vulnerable to a parallel objection; whereas experiences of God may be mistaken, we clearly do live in a world containing many kinds of goodness. Generally speaking, however, those who doubt the inference from an experience to a divine cause doubt not the reality of the experience but the validity of the inference which is made on the basis of that experience. Similarly, although we undoubtedly experience goodness in all its varieties, it may be the nature of that which is inferred from this experience which gives cause for concern.

Heather Widdows is particularly unsympathetic to Murdoch's ontological argument, which, she thinks, fails to support belief in either God or the Good; it may help to clarify the nature of the Good, but it is not effective as an argument for its existence. Widdows claims that using the argument and necessary existence to assert that goodness is ‘a constant certainty of the human condition and the enduring factor of the human quest … takes the reader directly into the old debate about predicates and attributes which has surrounded the argument from its inception’ (Widdows (Reference Widdows2005), 85). Some suggest that talk of the possession of attributes implies the existence of an object or being to which they belong, and this, she thinks, leads the reader to think of the Good as an object. Widdows therefore wonders whether Murdoch ‘might have achieved her aim more easily by avoiding Anselm's argument altogether, especially given the contempt in which many philosophers and theologians hold it, regarding it as little more than a word game’ (ibid., 85). Murdoch's fundamental point is that goodness is an essential part of experience, but, Widdows suggests,

Widdows's main misgiving about Murdoch's ontological argument, then, is that, for Murdoch, existence is an attribute, the possession of which may be taken – by some, at least – to imply that the Good is an object. But the key to interpreting Murdoch's argument lies in her understanding of necessary existence as different in kind from contingent existence. Neither God nor the Good exist in the same way as islands or triangles, but this does not entail their non-existence. Widdows's objection perhaps arises because she sees the argument from perfectionFootnote ¹⁴ and the Platonic degrees-of-goodness argument as separate from Murdoch's ontological argument when they are, as I suggested earlier, integral parts of that argument. For Murdoch, the ontological argument is not simply an attempt to move from a concept to the mind-independent reality of that concept by adding ‘existence’ to a list of attributes. It is, rather, an invitation to examine the concept of a perfect God/Good and to observe that a case for its existence may be supported by our moral experience.

Admittedly, there is some ambiguity in Murdoch's writings. While she refers to Plato's Good as ‘an impersonal object of love’ (Murdoch (Reference Murdoch1992), 344), however, she also explicitly denies that the Good is ‘a particular thing among other things’ (ibid., 38; see also 405), and notes that ‘in an important sense goodness must be an idea’ (ibid., 478; quoted in Widdows (Reference Widdows2005), 82). Widdows acknowledges this, suggesting that, although Murdoch sometimes refers to the Good as ‘an object of attention’, ‘the fact that the good is not an object forms part of her reason for claiming that the proof works for good but not for God’ (ibid., 82). Widdows objects that Murdoch has misconstrued the concept of God since, as Fergus Kerr (Reference Kerr1997, 75) has pointed out, there is a strong tradition in Christianity which teaches that God is not an object either. She claims that, if Murdoch were to accept this, it would weaken her distinction between God and Good. Murdoch does, indeed, refer to God as an object (e.g. Murdoch (Reference Murdoch1985), 55 and (Reference Murdoch1992), 455), but her claim that God's existence is not like that of islands or triangles suggests that God, for Murdoch, is no more ‘an “object” in the material sense’ (Widdows (Reference Widdows2005), 82) than the Good. Thus, Murdoch's distinction between God and the Good may not be, as Widdows suggests, a distinction between an object and a non-object. Rather, the distinction may be, as Widdows implies that it must be, between a personal God and an impersonal Good. Given the difficulties which she perceives to be associated with belief in the former, for Murdoch it is the latter of these which is more plausibly underpinned by a version of the ontological argument.

Objections to Murdoch's ontological argument: Stephen Mulhall

Stephen Mulhall's assessment of Murdoch's argument is initially more positive than that of Widdows. He accepts that Murdoch shows, against Kant, that God's existence is not like that of physical objects. And, just as Anselm argues that God's existence is necessary, so Murdoch shows that our sense of value is indispensable and points towards a perfection lying beyond our experience. Thus, Murdoch ‘detects and aims to recover far more of metaphysical and moral value from … [the ontological argument] than most philosophers’ (Mulhall (Reference Mulhall and Rowe2007), 30). But Mulhall is concerned about Murdoch's rejection of a personal God and, in particular, her rejection of belief in God as Father and God as Son. Mulhall acknowledges the difficulties associated with thinking of God as Father, and in particular the danger of seeing God as a projection of a consoling human fantasy. Murdoch thinks that the only way to avoid this threat is to dispense with this way of talking about God altogether, and yet she also holds that ‘God is not in any way to be grasped in terms appropriate to entities within the world of our experience’ (ibid., 31). Mulhall notes that any Christian who accepts this will agree that it is idolatrous to regard God's fatherhood as an ‘infinitely magnified version of human parenthood’ (ibid.), and concludes, first, that Murdoch's critique of the traditional conception of God the Father does not require her to step outside of the Christian tradition; any believer would share her misgivings about the picture of God she describes. Second, there may be another way of understanding this use of language about God which avoids the difficulty. Mulhall suggests that God's fatherhood expresses the idea that we are God's children and that our existence is a gift, an expression of love understood as a kind of kenosis, self-emptying, in which God comes to see the point of His own existence as lying outside of himself. On such a view, Christianity asks us to view our nature as ‘calling us to express such loving self-abnegation in our own mode of life’; it might therefore ‘actually support the mode of moral perfectionism so central to Murdoch's own thinking’ (ibid.). Mulhall suggests that talk of God as Son can also be framed in language which sidesteps Murdoch's misgivings. For Mulhall, the incarnation articulates ‘the conviction that not only human reality but reality as such is not essentially distant from that of God’ (ibid., 32); God's creation ‘is essentially consonant with, even expressive of, His own nature’ (ibid.). In Christianity, perfection has become fully real (ibid., 33).

Mulhall suggests, however, that the most significant problem with Murdoch's view is that ‘her sense of the ubiquity of goodness is insufficiently earned’ (ibid.). Murdoch does seem to concede this in the penultimate chapter of the Metaphysics, ‘Void’, where she raises the possibility that there may be some experiences – ‘black misery, bereavement, remorse, frustrated talent, loneliness, humiliation, depression, secret woe’ (Murdoch (Reference Murdoch1992), 498–499) – which militate against her view that the Good is ‘always and everywhere magnetic, that loving attention to the world will attract us to a clearer image of reality, that our transformative energies are never entirely in abeyance’ (Mulhall (Reference Mulhall and Rowe2007), 33). According to Mulhall, Murdoch claims that such episodes pass, and that, as Simone Weil suggested, they sometimes give spiritual succour by teaching us that we are nothing. But Mulhall points out that Weil was working within a Christian framework; Christ's despair on the cross expressed affliction, but also ‘a moment in the internal dialogue of God with God’ (ibid.). Since God can suffer God-forsakenness, this ‘recuperates even this most extreme crucifixion of the human self for the work of spiritual pilgrimage’ (ibid., 34). Thus, Mulhall argues, our ability to maintain ‘a perfectionist conception of the human self’ (ibid.) depends upon the teachings of Christianity, and ultimately upon the person of Christ.

Mulhall's first objection aims to show that Murdoch's position is closer to that of theism, specifically Christian theism as traditionally understood, than she would care to admit. He agrees with Murdoch that divine fatherhood is not to be understood as a magnified version of human parenthood, and commends to us a metaphorical interpretation which conveys the thought that our existence is a gift of God's self to us which we must emulate by giving ourselves to others. Similarly, the idea of divine sonship shows us that God's creation expresses God's nature; the person of Christ demonstrates the reality of perfection. But Mulhall's metaphorical interpretation of divine fatherhood seems to imply an element of divine agency which, given her rejection of interpretations of the divine as personal, Murdoch would be unlikely to accept – unless the notion of divine agency is also understood in a metaphorical sense. Mulhall's interpretation of divine sonship is, however, reminiscent both of Murdoch's ubiquity of goodness argument, and of her understanding of Christ as ‘a personal ontological proof’ – although, for Murdoch, Christ points towards perfection without, himself, being perfect.

Mulhall's claim that Murdoch's ‘sense of the ubiquity of goodness’ is ‘insufficiently earned’ is more problematic. If, as Mulhall suggests, God's creation expresses God's nature, the evils which Murdoch lists in the penultimate chapter of the Metaphysics are no less a challenge for his own view.Footnote ¹⁵ And Murdoch perhaps concedes too much when she acknowledges that some experiences militate against the ubiquity of goodness; although many experience terrible suffering, the world also contains numerous examples of the goodness on which Murdoch bases her argument. In addition, Mulhall focuses on the ubiquity of goodness version of her argument rather than the Platonic degrees-of-goodness argument, which may be less vulnerable to his objection. The degrees-of-goodness argument requires only examples of goodness which we can place in order of increasing goodness, and from which we might deduce the existence of a perfect standard. The existence of evil does not necessarily prevent us from making this deduction – and the Good thus deduced is the means by which evil, and the suffering to which it gives rise, may be resisted or alleviated.Footnote ¹⁶

Mulhall also objects that, while Murdoch draws on Weil for the view that ‘the void can give spiritual succour, insofar as it teaches us that we are absolutely nothing’ (Mulhall (Reference Mulhall and Rowe2007), 33),Footnote ¹⁷ this makes sense only within Weil's Christian framework, in which even God, in the form of Christ on the cross, experiences affliction which becomes an opportunity for spiritual pilgrimage. By the time of the Metaphysics, however, Murdoch concludes that ‘it is possible, but very often just too difficult, to “learn” from deep despair’ (Murdoch (Reference Murdoch1992), 501–502), and recommends instead that we should ‘hold on to what has really happened and not cover it with imagining how we are to unhappen it … What is needed here, and is so difficult to achieve, is a new orientation of our desires, a re-education of our instinctive feelings’ (ibid., 503). We should seek a reorientation and relation to goodness: ‘a deep, or a real, or a proper, recovery demands … some sort of moral activity, a making a spiritual use of one's desolation’ (ibid.).

Objections to Murdoch's ontological argument: Peter Byrne

While the existence of the Good may not be like that of physical objects, it remains legitimate to ask what kind of existence it does have. Peter Byrne (Reference Byrne1998) argues that Murdoch's Good is no more than a name for human aspirations. Since it does not causally account for anything, there is no reason to believe in the mind-independence which the ontological argument allegedly supports, and it is thereby distanced from traditional interpretations of theism.

Byrne points out that, for Murdoch, human life has no external point or telos, and human beings are subject to necessity and chance (Byrne (Reference Byrne1998), 111, referring to Murdoch (Reference Murdoch1985), 78–79). Without an external point or telos, goodness is only good if it is good for nothing (Murdoch (Reference Murdoch1985), 71), and, for Byrne, this ‘implies that we positively require good acts to produce no visible good for anyone in the long run, especially not ourselves, in order to see them as good’ (Byrne (Reference Byrne1998), 112). Further, if we accept that everything that happens is subject to necessity and chance, we have no adequate response to the secular problem of evil which questions whether the human good can be attained by means of good acts (ibid.).

Byrne argues that the central problem which arises from Murdoch's rejection of teleology is that ‘the concept of the Good seems to refer … to something which is inert … There should be something the Good does’ (ibid., 112–113). If that which is real is that which has causal power, the Good is not real because there is nothing for which it causally accounts; if the Good existed independently of our perception of it, human beings would not be subject to necessity and chance. Thus, Byrne thinks, for Murdoch, the transcendent makes no contribution to the structure of reality, and this raises the question of ‘how far she can pretend to have preserved anything like the spirit of the ontological proof’ (ibid., 114). For Byrne, Murdoch's proof does not point towards divine things because perfection and value do not have the kind of existence which would put them at the heart and ground of everything that exists (ibid., 115). On Byrne's view, a religion must provide an answer to the secular problem of evil by showing that the structure of reality is such that our moral actions are appropriately rewarded; in doing so, it preserves a minimal metaphysical realism. Since Murdoch's view makes no link between moral action and the nature of reality, it cannot be regarded as a form of religious belief.

It is not the case that that there is nothing which the Good does, however. I mentioned above the possibility that, on Murdoch's view, the notion of divine agency might be understood in a metaphorical sense. In response to Byrne, I would suggest that there are at least three ways in which, on Murdoch's view, a metaphorical interpretation of divine agency might be understood.

First, the existence of the Good explains our ability to recognize degrees of goodness. An interpretation of Anselm's argument similar to that of Murdoch is offered by Gene Fendt (Reference Fendt2005). According to Fendt, Anselm's argument, as it is found in both the Monologion and the Proslogion, aims to discover ‘the evident presence of God in any judgement of value’ (ibid., 150). Developing an argument first put forward by Plato, Fendt asks us to consider five men, each of whom exhibits a different level of justice. He suggests that Anselm's point is that it is Justice itself which enables us to place these on a scale of increasing justice (ibid., 152). Justice itself is not on the scale of justice, however; that through which we judge the variety of natures ‘must exist in a way that is different from the way the five exemplary men exist … because it is through Justice that we understand them and place them on the scale’ (ibid., 156). Fendt suggests that all human beings have the ability to recognize degrees of justice; if someone cannot or will not acknowledge that some things are more just than others, he recommends that we should ‘put out his left eye – and see if he complains about something more than pain’ (ibid., 161). The same argument applies to all goods; we can ‘start this meditation from anywhere on earth and get to God’ (ibid., 157). To deny the existence of that through which any scale of greatness exists and can be known is to render oneself unable to make any judgements at all (ibid., 161). Thus, Gaunilo's ‘lost island is … entirely lost: it touches the world at no point; it is but a figment of the fool's imagination’, but ‘“that than which nothing greater can be conceived” does touch the world at one point: your conscience, which testifies that it recognizes an order of justice, good, etc.’ (ibid., 162).Footnote ¹⁸

The reality of the Good, then, is a condition for our ability to place the varieties of goodness which we experience on a scale of increasing goodness. Although we sometimes disagree about the rankings of different types of goodness, even our disagreements may be taken to imply the existence of a standard of perfection the nature of which our opponents have failed to understand.

A second function of the Good is to help us to overcome our ‘natural’ egoism, a feature of Murdoch's account which has been influential in subsequent developments in virtue ethics (e.g. Taylor (Reference Taylor1989), 3). Simon Blackburn objects that Murdoch's moral vision is repellent because it ‘exalt[s] the mystery of the necessary and unconditional Good by debasing the everyday’ (Blackburn (Reference Blackburn1992), 3), but this overlooks her claim that attention to the Good enables ‘the purification and reorientation of an energy which is naturally selfish, in such a way that when moments of choice arrive we shall be sure of acting rightly’ (Murdoch (Reference Murdoch1985), 54). Contemplation of ‘the everyday’ – perhaps by means of something analogous to prayer or meditation (e.g. Murdoch (Reference Murdoch1985), 54–56, 83, and (Reference Murdoch1992), 73, 245, 448), or with the aid of great works of art (e.g. Murdoch (Reference Murdoch1985), 85–87), which she describes at one point as ‘a sort of Ontological Proof’ (Murdoch (Reference Murdoch1992), 429) – can point us towards true goodness and this, once we have understood it to the best of our ability, can enable us to behave more ethically in the realm of the everyday.Footnote ¹⁹

So the Good, first, accounts for our ability to recognize degrees of goodness, and, second, helps us to overcome our ‘natural’ selfishness. Third, as I have already argued in response to Mulhall, contemplation of the Good and the action to which it gives rise is a means by which evil may be resisted and suffering may be alleviated. Byrne is, of course, correct to note that Murdoch's Good does not ensure that the human good is ‘attainable through doing good acts’ (Byrne (Reference Byrne1998), 112), but this does not mean that she fails to address the ‘problem’ of evil. Byrne suggests that, for PlatoFootnote ²⁰ and Kant, in order to solve the secular problem of evil there must be some guarantee that human beings will achieve happiness in proportion to their moral strivingsFootnote ²¹ and that, in this respect, there is a difference between their views and those of Murdoch.Footnote ²² But Murdoch does not understand the secular problem of evil in the same way. She acknowledges that, without a personal God, there is no problem of evil in the traditional sense, but there remains ‘the almost insuperable difficulty of looking properly at evil and suffering’ (Murdoch (Reference Murdoch1985), 73), i.e. the difficulty of representing evil and suffering accurately to ourselves, without introducing any element of consolation. For Murdoch, attention to the Good helps us to understand the true nature of evil and suffering, and it is the means by which we attempt to overcome the challenges which they present. Clearly, there must be some expectation of success in at least some cases for this to be a worthwhile pursuit, but there is no need for success on every occasion. Further, any success that might be achieved must be understood as an increase in virtue and not as a reward of some kind for oneself. Of course we may experience personal benefits as a consequence of virtuous action. As Byrne (Reference Byrne1998, 41) notes, most of our actions are likely to affect people from whose flourishing we may derive some satisfaction. Further, as we noted above, moral activity can enable us to focus away from our own concerns and thereby to endure or recover from terrible suffering (Murdoch (Reference Murdoch1992), 503). Any such benefits are necessarily secondary, however; it is not merely the case that there may be no reward, but that there should be no expectation of reward. Indeed, while Byrne sees death and chance as fatal to Murdoch's position, Murdoch herself thinks that ‘there is a special link between the concept of Good and the ideas of Death and Chance … A genuine sense of mortality enables us to see virtue as the only thing of worth’ (Murdoch (Reference Murdoch1985), 99). For Murdoch, the anticipation of a reward for virtuous action is incompatible with true virtue; contemplation of the Good helps us to be good, not to achieve a reward for our goodness.

Thus, Byrne thinks that the Good has no referent because there is nothing that guarantees the success of moral endeavour, but I have argued that there is no need to see the Good as something which provides a guarantee of this kind in order to see it as an objective existent. On Murdoch's view, there is no external end or telos, in that there is no guarantee of moral progress or ultimate destination at which we will be appropriately rewarded for our actions, but this does not mean that our actions have no goal; as Murdoch herself suggests, ‘there are properly many patterns and purposes within life’ (Murdoch (Reference Murdoch1985), 79). Murdoch has not abandoned teleology, therefore; she has only abandoned teleology of what she regards as an objectionable kind.

The nature of the Good

I have argued for the reality of the Good on the grounds that there are at least three things which the Good does: it explains our ability to make moral judgements, it helps us to overcome human selfishness, and it addresses the secular problem of evil. The nature of that reality is, however, still unclear. I have already noted Dombrowski's claim that his argument moves from an abstract concept to the abstract conclusion that God exists necessarily in some concrete state (Dombrowski (Reference Dombrowski2006), 86) – that we can know that God exists, but not how God exists. Murdoch, particularly in her use of the degrees-of-goodness argument, is perhaps more successful than Dombrowski in showing that God/Good exists. It might be to her advantage if, like Dombrowski, she cannot say exactly how it exists, as this might enable her to avoid the objection, common to many other versions of the ontological argument, that her argument attempts to move from an abstract concept to a concrete reality.

Murdoch herself notes the ‘indefinable and non-representable character’ of the Good (1985, 69, 97–101). She suggests that we know some things about the Good; the ordinary person believes that some things are objectively better than others, and that we can make moral mistakes (ibid., 97). We also recognize the existence of evil. But the concept of Good ‘remains obscure and mysterious. We see the world in the light of the Good, but what is the Good itself?’ (ibid., 98). Good cannot be defined in terms of other virtues, such as Truth or Courage, since these must be defined in terms of the goodness which we are trying to elucidate. A refined Love is, perhaps, ‘practically identical with goodness’ (ibid., 102), but Love cannot be identified with Good because, even when Love is purified, Good and Love have different functions; Good is that which purifies Love. Murdoch argues that the Good is mysterious for three reasons. First, the world is ‘aimless, chancy and huge’ (ibid., 100), which makes it impossible for us to comprehend all the ways in which virtue might be required of us. Second, human selfishness makes it difficult for us to recognize the Good: ‘If there were angels they might be able to define good but we would not understand the definition’ (ibid., 99). And, third, it is difficult to look at the Good because it is not like looking at other things. We have the idea that there is a Good, but we see only instantiations which point towards it; we do not and cannot know what the Good is like in itself (ibid., 100).

Nevertheless, Murdoch's writings do contain a number of positive references to the nature of the Good. In The Sovereignty of Good she suggests that the Good is a metaphor – ‘a very important metaphor and one which is not just a property of philosophy and not just a model’ (Murdoch (Reference Murdoch1985), 93; Plato's Good is described as a metaphor at ibid., 71). As we saw earlier, Dombrowski seems to suggest that an idea of God might constitute an experience of God which, in turn, supports belief in God. In a similar manner, Murdoch suggests that the Good is ‘real as an Idea’. For Murdoch, however, this Idea is ‘also incarnate in knowledge and work and love … experience of the reality of good … is a discovery of something independent of us’ (Murdoch (Reference Murdoch1992), 508). At the end of the Metaphysics, she seems to contradict this by suggesting that the first version of the ontological argument, the ‘proof by perfection’, supports the view that ‘[t]he good artist, the true lover, the dedicated thinker, the unselfish moral agent solving his problem … can create the object of love’ (ibid., 506). The contradiction may be only apparent, however, since the good artist creates, at best, an instantiation of the Good which points towards the metaphor of the Good. If the Good is a metaphor, it is, indeed, as Alan Jacobs suggests, ‘an imaginative creation’ (Jacobs (Reference Jacobs1995), 6), and, perhaps, something which we can make incarnate, but, for Murdoch, we do not create that to which the metaphor points.

The Good, then, represents our best attempt to describe something which must exist, but which most of us cannot fully understand or explain. Murdoch notes Wittgenstein's claim that his propositions, the metaphysics of the Tractatus, are merely steps which, once we have understood them, we can climb beyond; we should throw away the ladder after we have climbed it (Wittgenstein (Reference Wittgenstein1961), 6:54; quoted in Murdoch (Reference Murdoch1992), 422). She invites us to compare Plato's myths, which are not to be taken literally (Murdoch (Reference Murdoch1992), 423).Footnote ²³ The implication would appear to be that the religious teachings which have been handed down to us, and even Murdoch's own ideas about the Good, are also ladders to be thrown away after use.Footnote ²⁴ But, as Murdoch acknowledges, few – except perhaps the saints and mystics to whom she frequently refers – will be able to be without ladders at all. Aquinas rejected the ontological argument because we do not know God's essence and therefore cannot know that it includes existence (Aquinas (2008), I, 1:2), but it is not necessary for us to be able to conceive of God/Good as it is in itself; we can deduce its existence from our limited experience and then struggle continuously towards a better understanding of its nature and what it requires of us.

Thus, for Murdoch, God is a symbol of the Good (Murdoch (Reference Murdoch1992), 428), which is itself a symbol for something we are unable fully to understand; although the Good is not God, both God and the Good are attempts to describe the same ultimate reality. But, while Murdoch rejects personal conceptions of the divine, she thinks that morality without religion is too abstract: ‘Religion symbolises high moral ideas which then travel with us and are more intimately and accessibly effective than the unadorned promptings of reason’ (ibid., 484). She recommends the use of religious imagery, which may be seen as images of a higher reality (ibid., 451), or sources of spiritual energy (ibid., 249), and thinks that religious rituals can serve as ‘vehicles of enlightenment, as exercises likely to strengthen good desires’ (ibid., 433).

But is belief in the Good a type of religious belief? I have argued against Byrne that, if religion entails realism, Murdoch's view cannot be regarded as irreligious on the grounds that it depends on a non-realist outlook. If religion requires belief in a personal God, however, on her own terms, Murdoch's view is not religious – although, even here, it might be argued that Murdoch's Good shares, at least in a metaphorical sense, some of the characteristics of a personal deity.Footnote ²⁵ But if, as Murdoch suggests, religion is defined as ‘the passionate love of the Good’ (Murdoch (Reference Murdoch1992), 326),Footnote ²⁶ this clearly encompasses her own position. She describes her view as a ‘sort of mysticism … a non-dogmatic essentially unformulated faith in the reality of the Good’ (Murdoch (Reference Murdoch1985), 74), and the ontological argument as ‘not exactly a proof but rather a clear assertion of faith … which could only confidently be made on the basis of a certain amount of experience’ (ibid., 63). If a religion explains how we are able to recognize goodness and helps us to do so by suggesting individual and/or communal practices which enable us to become more virtuous, then perhaps it may reasonably be argued that contemplation of the Good and the conduct it requires of us can be regarded as a form of religious belief. With the assistance of Murdoch's ontological argument, it is a form of religious belief that might be more plausible than many of the alternatives.