CC BY
9Философия религии: аналитические исследования
Philosophy of Religion: Analytic Researches 2022, vol. 6, no. 2, pp. 56-66 DOI: 10.21146/2587-683X-2022-6-2-56-66
2022. Т. 6. № 2. С. 56-66 УДК 21
M. Wynn
Some Reflections on Richard's Swinburne's 'God's Moral Goodness'
Mark Wynn - DPhil, Oriel College, University of Oxford, Oxford OX1 4EW, UK; e-mail: mark.wynn@oriel.ox.ac.uk
In this paper, I consider Richard Swinburne's treatment of the idea of divine goodness in his new article "God's Moral Goodness" with reference to his understanding of these three themes: the prior probability of theism; the predictive power of theism relative to naturalism with respect to various generic features of the world; and the question of why God should have made this world in particular. In each case, my intention is not fundamentally to challenge Swinburne's account, but to seek to draw out its implications, and to consider at what points it might invite elaboration, if is to be located within, for example, a Christian conception of the divine reasons for creating.
Keywords: moral goodness, Richard Swinburne, Anselm of Canterbury, God, Ethics Citation: Wynn M. "Some Reflections on Richard's Swinburne's 'God's Moral Goodness'", Philosophy of Religion: Analytic Researches, 2022, Vol. 6, No. 2, pp. 56-66.
Марк Уинн
Ориел Колледж, Оксфордский университет, Оксфорд OX1 4EW, Великобритания; e-mail: mark.wynn@oriel.ox.ac.uk
В данной статье я рассматриваю то, как Ричард Суинберн в своей новой статье «Моральная благость Бога» трактует идею божественной благости в связи с его пониманием следующих трех тем: априорная вероятность теизма; предсказательная сила теизма в сравнении с натурализмом при объяснении базовых особенностей мира; и, в частности, вопрос почему Бог должен был сотворить этот мир. В каждом случае мое намерение
Некоторые размышления о «Моральной благости Бога» Ричарда Суинберна
© Wynn M., 2022
заключалось не в том, чтобы фундаментальным образом оспорить подход Суинберна, а в стремлении выяснить следствия, которые порождает этот подход и рассмотреть, в каких пунктах он мог бы потребовать доработки, если бы он, к примеру, был бы совмещен с христианским пониманием причин, в силу которых Бог осуществляет творение. Ключевые слова: моральная благость, Ричард Суинберн, Ансельм Кентерберийский, Бог, этика
Ссылка для цитирования: Wynn M. Some reflections on Richard's Swinburne's 'God's Moral Goodness' // Философия религии: аналит. исслед. / Philosophy of Religion: Analytic Researches. 2022. Т. 6. № 2. С. 56-66.
Some decades ago now, Professor Swinburne very kindly supervised the later stages of my doctorate - and my own perspective on the issues he discusses in his contribution to this volume has been, and continues to be, very much informed by his. Since the reader will not want me simply to record the many respects in which I find myself in agreement with Swinburne, I shall try to pick out a number of themes in his essay where I am if not in disagreement with him, then at least interested to think about how his treatment of a given issue might be expanded, to accommodate an objection, or to throw some facet of his position into new relief by placing it in a wider intellectual context. Broadly speaking, my comments on these matters will follow the order in which they are presented in Swinburne's text. I shall examine in turn Swinburne's account of the prior probability and explanatory power of theism, and his understanding of the role of reasons in orienting the divine will.
[1] The goodness of God and the prior probability of theism
At the beginning of his paper, Swinburne draws a distinction between Anselmian and Thomistic approaches to natural theology, noting that while the first proceeds a priori, the second tends to think of God, from a philosophical vantage point, as whatever would best serve to explain various fundamental features of the world. On this general methodological question, Swinburne aligns himself, of course, with the Thomistic perspective (P. 7). Given this way of representing his position, it may be of some interest to consider briefly the extent to which Swinburne himself gives a weighty role to a priori judgements when developing his probabilistic case for the existence of God.
In his paper, Swinburne distinguishes between two kinds of hypothesis: those that appeal to the beliefs and intentions of persons, and those that cite the powers and liabilities of inanimate agents (P. 7), taking these two kinds of hypothesis to be logically exhaustive. Hypotheses of the first type can be distinguished from one another according to the number and character of the personal agents they postulate. And Swinburne argues, plausibly I think, that if we take the whole array of such hypotheses that are in principle available to us, the simplest will postulate a single agent, 'to whose power and length of life there are 0 limits' (P. 8)1. Swinburne also
It is worth noting one complication: from his account of the doctrine of the Trinity, it is clear that for Swinburne, a single such agent is likely to give rise two, and only two, additional such agents. His position here is consistent with his treatment of simplicity as a measure of a priori probability,
thinks that the simplest such agent will have '0 limits' on its knowledge, and be subject to no non-rational influences, from which it follows that this agent will also be perfectly good - and granted this conception of its properties, Swinburne is surely right to identify this agent with God as traditionally conceived.
Of course, a further key claim in Swinburne's text is that if two hypotheses are of equal merit in all other respects, then we should take the simpler hypothesis to be the more probable. And it follows that if we are considering the relative probability of two hypotheses purely a priori, then we should treat the simpler hypothesis as the more probable - since in this case, it will not be possible to distinguish these hypotheses with respect to, for instance, their predictive power (since there is no evidence to be predicted) or their fit with background knowledge (since we are proceeding on the assumption that we have no substantive background knowledge). Accordingly, for Swinburne, the key determinant of the a priori probability of a hypothesis is its simplicity2.
It is worth being clear that on this account, it is not just that theism will be a priori more probable than any one of the alternative personal hypotheses that are in principle available to us - for instance, the hypothesis that postulates a single agent that is like God in all respects save in having a range of power that falls somewhat short of the divine power. The thought is, rather, that theism will be a priori more probable than the disjunction of such hypotheses. This position reflects, Swinburne thinks, the practice of scientists when, for example, they postulate a single agent to explain a given body of evidence, even if there should be an array of alternative hypotheses citing two, or three, or however many, agents of the same general type that would predict the evidence just as well [see: Swinburne 2004: 146]. Let's consider now the implications of this position for the feasibility of an a priori case for the existence of God.
As we have seen, we are to suppose that the theistic hypothesis is more probable a priori than the disjunction of those hypotheses that postulate one or more personal agents, but not God. Let us suppose next that our background knowledge of the world comprises only the truth that there exists at least one person. In that case, it seems we should conclude that, probably, there is a God. Why? Well, we know that the theistic hypothesis is more probable a priori than the disjunction of all those hypotheses that postulate one or more persons other than God. So if we know simply that either theism or one of these alternative hypotheses is true (this is our background knowledge), then we should infer that most likely theism is true, since it is a priori more probable than the disjunction of its alternatives, and there is, it seems, nothing in our background knowledge that should lead us to revise this
since on this view, it is considerations of simplicity that should lead us to postulate one such agent, and then considerations of what would make for perfection in that agent - and specifically, perfection in love - that should lead us to postulate two further such agents. See [Swinburne 1994: Ch. 8].
Of course, if one hypothesis asserts more than another, then it will be less probable, and the suggestion that the simpler hypothesis is more probable a priori needs to be read accordingly, that is, the simpler hypothesis where the two hypotheses are of broadly comparable scope.
a priori assessment of its relative probability. Here, then, is a case for the existence of God which proceeds on the basis of very minimal background knowledge.
Of course, this argument differs from the Anselmian approach to proving the existence of God that Swinburne mentions at the beginning of his paper - both because it is not purely a priori, and because its conclusion is taken to hold simply as a matter of probability. Nonetheless, the claim that there exists at least one person is a very minimal claim. It is indeed a more modest claim than the claim that forms the key premise of contemporary modal versions of the Ontological Argument, namely, the so-called 'possibility premise', according to which it is possible that it is necessary that there should be a God - reading this premise so that it entails that God exists in all possible worlds. (I say 'more modest' because the possibility premise, so understood, entails that there in fact exists at least one person, but is not entailed by it.)
It would be interesting to consider, I think, whether a similar case could be made if we take our background knowledge to comprise a still more modest truth -say, the truth that something or other exists, without further specification. However, that is not a project I shall pursue here. My suggestion is simply that while Swinburne is, of course, right to suppose that he is not committed to the kind of a priori epistemology that is typical of the Anselmian tradition, it is also true that his case depends upon a weighty a priori claim concerning the probability of theism relative to other hypotheses - a claim so weighty that, if true, it would, it seems, give us good reason to suppose that there is a God granted very minimal background knowledge.
Let's turn next to the wider, broadly Thomistic, natural theological argument that Swinburne presents in his paper. So now our focus will move from theism's prior probability to its explanatory power. Here, we will see how on Swinburne's account, the divine goodness is not only entailed by the divine simplicity, and therefore integral to any assessment of the prior probability of theism, but also crucial to securing the predictive power of theism.
[2] The goodness of God and the explanatory power of theism
In his paper, Swinburne alludes to various features of the world that in his view significantly raise the probability of theism relative to what it would otherwise have been. I'll briefly discuss two of these features.
Swinburne notes that our universe comprises 'an uncountable number of fundamental particles in which every particle has the same power of gravitational attraction (and of the other natural forces) as every other particle' (P. 11). And elsewhere, he has commented that 'The universe might so naturally have been chaotic, but it is not - it is very orderly', and that 'there is a strong presumption of randomness' [see: Swinburne 1991: 136, 299]. Putting together these various observations, we might conclude that we ought to explain the world's regularity or order (for instance, the fact that 'every particle has the same power of gravitational attraction as every other particle') because this state of affairs is so little to be expected, given that 'the universe might so naturally have been chaotic'. (Compare the familiar situation
in everyday life where I seek to explain some phenomenon because it was not to be expected.) But given Swinburne's commitment to the idea that the simplest of the available hypotheses will be not only more probable a priori than each of the (relatively complex) alternatives considered individually, but also more probable than the disjunction of these alternatives, it seems that it may be best to ground the idea that the world's regularity calls for explanation in some other consideration. Let me expand on this point briefly.
In his paper for this volume, Swinburne notes that for a hypothesis to be simple, 'it must postulate the existence and operation of few substances, few kinds of substance, with few easily describable properties correlated in few mathematically simple kinds of way' (P. 8). This stance would seem to suggest that a chaotic universe is to that extent relatively complex, or lacking in simplicity, and in turn therefore relatively improbable a priori, when compared with our own universe -because in a chaotic universe, there will not be a 'few kinds of substance', or substances with 'few easily describable properties', whereas the order of our own universe consists in part, it seems, in the fact that there are relatively few kinds of fundamental particle, and in the fact that the behaviour of those particles can be represented by means of laws that have at least a somewhat elegant mathematical form. So if we allow, once again, that the simplest hypothesis is to be preferred to the disjunction of alternative, relatively complex, hypotheses, then it seems we should conclude that the natural theological case for the idea that the world's order is to be explained is best expressed not by supposing that there is a strong presumption of randomness, or that chaos is somehow the 'natural' state, so that order is, therefore, a priori improbable, but by maintaining that our picture of reality will be simpler if we take the order of the world to derive from (a supremely simple) God than if we take that order to derive from one or more inanimate causes or to exist unexplained3. And in his paper, Swinburne develops a case of just this form.
When presenting his a posteriori argument for theism, Swinburne appeals not only to the world's order, but also its goodness. And this further strand of his case is in its way still more fundamental, because the superiority of theistic explanation over its naturalistic rivals rests, for Swinburne, on this thought: as perfectly responsive to reasons, God is perfectly good, and accordingly, God will favour good outcomes over bad, notwithstanding the fact that there are no limits on the divine power, that is, no limits (other than logical limits) on what God could in principle do, supposing that the scope of his power is conceived independently of reference to his moral perfection. Accordingly, the theistic hypothesis is able to combine high prior probability - because there are '0 limits' on, for example, the divine power -with high predictive power - because while God could in principle bring about any of a very wide range of outcomes, granted the scope of the divine power, God can even so be relied upon to bring about those outcomes that involve a significant surplus of good over bad, since God is perfectly responsive to reasons. So in this way,
For a fuller account of Swinburne's view on these questions, see the first appendix of the 1991 edition of the Existence of God.
the superiority of theism as an explanation of the order or regularity of the world flows from these truths about the divine goodness: the theistic hypothesis combines a lack of specificity with respect to the divine power (so has a relatively high prior probability) with a high degree of specificity with respect to the divine effects, so far as those effects involve a significant surplus of good over bad (hence the hypothesis also has a relatively high predictive power).
Of course, we could explain any surplus of good over bad in the world by reference to one or more inanimate causes, by defining the relevant hypothesis so that it has a high predictive power with respect to this surplus: for instance, we could take these inanimate agents to have precisely those powers, and precisely the tendency to exercise those powers, that would lead us to expect this surplus of good over bad. But any such hypothesis will be highly contrived: it will need to be 'gerrymandered' in this sort of way, in order to raise its predictive power, and accordingly it will be of relatively low prior probability. I think Swinburne is right to say that theism enjoys an advantage in this respect, since only a personal kind of explanation can account in a principled way for any propensity there may be in the world for good to predominate over bad - by appealing to the capacity of a personal agent to be sensitive to moral considerations.
The key challenge for this kind of approach is then, I take it, to show that the universe does indeed have a tendency to generate a surplus of good over bad of the kind that theism would lead us to expect. Swinburne notes that theism is committed to a certain reading of this excess of good over bad: given the truth of theism, it is not enough that there should be simply overall more good than bad; it is also necessary that each individual bad (each individual case of suffering, or frustration, or affliction of whatever kind) should be 'defeated', that is, should be the logically necessary condition of a good that is at least as great as (or perhaps it would be better to say: that is greater than) the bad, or else the logically necessary condition of warding off an evil that is at least as bad as the bad that in fact obtains (Pp. 12, 13). This account of the nature of the surplus of good over bad that we would expect God to bring about suggests, potentially, a difficulty for Swinburne's claim that theism is the best explanation of 'a universe with the kind of balance of good and evil which our universe has' (P. 11). The difficulty is that with respect to the portion of the universe's history that is open to our view, many bads seem not to be defeated: for instance, from this vantage point, there seem to be cases of human pain that do not serve any greater good.
Perhaps Swinburne will say in reply that in such cases, the pain (and similarly for other forms of affliction) derives from some system of pain responsiveness that works for good in most cases - so that individual episodes of pain can, to this extent, be subsumed within some larger framework that generates a surplus of good over bad. Supposing this response to be right, we might think that there remains a further condition that any surplus of good over bad needs to satisfy within a theis-tic universe: not only should bads be defeated by goods when those bads are located within the relevant system, but in addition the bads that affect a particular individual should be defeated within the life of that individual - for if we were to suppose otherwise, then we would be representing God as a kind of consequentialist agent, who is willing to allow some to suffer simply for the sake of benefits that are to be
enjoyed by others4. If we introduce this additional condition, then our difficulty seems to return, because, with respect to the portion of the universe that is open to our view, there seem to be many cases of, say, human suffering that are not de -feated within the life of the individual who suffers.
Of course, Christians, including Swinburne - and theists more generally - have a ready response to this challenge, since Christians are anyway committed (that is, committed independently of reference to the problem of evil) to the idea that there is an afterlife, and accordingly they are free to say that the defeat of bads by goods, within the life of an individual, will be achieved, if not in the course of this life, then eschatologically. However, once this move has been made, then it may seem that theism will no longer generate any very precise expectation about the kind, or extent, of the surplus of good over bad that we will find within the universe: once we associate theism with the idea of a post-mortem existence, then from a theistic vantage point, it may seem that there is no great cause for surprise if certain bads are not evidently defeated by goods within the span of this universe's history, or within the ante-mortem lives of the individuals who suffer those bads, because after all, God, if there is a God, can be counted on to defeat those bads in a further life, or perhaps at the point at which the universe reaches its eschatological culmi -nation. In brief, once we have introduced the afterlife as a context within which bads may be defeated, many more universes will be consistent with theism than would otherwise have been the case: in particular, many universes within which bads are not defeated by goods in the history of that universe, or within which bads are not defeated by goods in the lives of those who suffer those bads, will now be in principle consistent with theism. This is helpful for the project of theodicy, of course, but also makes it harder to confirm theism by showing that it generates a relatively precise expectation about the world's character, and the nature of any surplus of good over bad, that is in fact fulfilled.
Moreover, a naturalist, or in Swinburne's terms, one who takes an inanimate explanation of the world's character to be more fundamental than any personal explanation, might urge that the failure of bads to be defeated by goods, in many cases, within the context of the history of this universe, or within the context of the lives of individual sufferers, is exactly what we should expect if naturalism is true, since on the naturalist view the fundamental principles that account for the character of the world will make no reference to an agent that is sensitive to moral values.
It is perhaps significant that when illustrating the difficulty that naturalism has in accounting for the surplus of good over bad that we find in the world, Swinburne writes: 'an inanimate entity would have to have built into it a very specific power to produce just that balance [of good over evil], rather than a universe with less bad or more bad in it (for example, some rational beings suffering for ever, not through their own choice...)' (P. 11). Here, showing the explanatory inadequacy of naturalism relative to theism seems to depend, if not on an appeal to an afterlife, then at least on introducing the idea of rational beings with an infinitely
This kind of perspective, which locates the defeat of bad by good within the life of the individual, is evident in the work of, for example, [Adams 1999].
extended future life. This move means that we can reach a view about whether there is a surplus of good over bad in the course of such a person's life considered in its full extent. And I think Swinburne is right here: theism does generate a different expectation from naturalism regarding the character of a person's life in its full extent, where that life may in principle stretch over an infinite future, since theism leads us to expect the life of an individual to exhibit a surplus of good over bad over the full extent of their life, and indeed for the bads of individual sufferers to be 'defeated' in the course of their life, whereas naturalism as such surely generates no such expectation.
But once again, if the distinctiveness of the outcomes that we are to associate with theism is understood in these terms, then the problem remains that on the theis-tic view of the matter, even a comprehensive knowledge of the history of this universe would not be enough to disclose in full the character of any surplus of good over bad in the lives of human beings and potentially other creatures. So in brief, if bads are not defeated by goods within the ante-mortem lives of creatures, or within the history of the universe, that truth perhaps confirms naturalism to the extent that it is, I think, what naturalism would lead us to expect, but it does not follow that theism is thereby disconfirmed, since theism's claim about the surplus of good over bad is one that holds only across the lives of individuals considered in their entirety, and for theism, from our ante-mortem perspective, we do not have access to the lives of individuals considered in their entirety. So in brief, while Swinburne is right, I think, to say that theism and naturalism generate different expectations in these matters - concerning the surplus of good over bad in the lives of human beings and other creatures considered in their full extent - this difference of expectation need not extend to the relationship of bads and goods within our ante-mortem lives, so posing a problem for the idea that theism's predictive power is in this regard superior to that of naturalism.
I don't think this consideration simply overturns Swinburne's argument for design from the surplus of good over bad that we find in this universe. For instance, the fine-tuning version of the argument from design (an argument that Swinburne has discussed at length) suggests that to be consistent with the development of life, a universe, supposing it to be broadly of the same kind as our own universe, will need to satisfy a very restrictive set of conditions, concerning its initial expansion rate and the values of its fundamental forces. Life itself is, I take it, a fundamental good, and indeed the precondition of the other goods that are of interest to Swinburne in his discussion of the balance of goods and bads in the lives of human and other creatures, and the existence of a life-friendly universe seems, therefore, unsurprising on theism. But if the drift of the fine-tuning argument is correct, life is surely not to be expected on standard forms of naturalism - that is, on those forms of naturalism that do not entertain the existence of multiple universes. I mention these matters only to suggest one way in which the idea that the universe exhibits a surplus of good over bad, with respect now to the very existence of life, might be cited as a consideration that favours theism over naturalism, notwithstanding the complications for arguments of this general type that we have been discussing.
[3] The goodness of God and God's sensitivity to reasons
Let me conclude by introducing, very briefly, one further set of questions. Swinburne proposes that God is a 'best-acting God', where this entails that God 'always does the best action where there is a best action' (P. 9). It is a consequence of this view, I take it, that were there to be a uniquely best pos sible universe, then God would make that universe. Mark Murphy's position, which Swinburne discusses in the course of his paper, seems to suggest, on the contrary, that there is no expectation at all that God would produce the best possible universe, supposing there to be one, because God is 'holy' - and therefore has powerful, 'requiring' reasons not to be drawn into any kind of relationship with creatures, whose ontological status is necessarily far beneath his own5. Each of these ways of representing the goodness of God - goodness understood as perfect responsiveness to the balance of reasons, or perfect responsiveness to prospective good, and goodness understood as holiness - is potentially problematic, it seems to me, if our purpose is to represent the idea of divine goodness not simply from a philosophical vantage point, but in ways that accord with the Christian conception of that goodness.
On his own account of the matter, Murphy's view seems to imply that God's becoming incarnate, or making atonement (or indeed, creating) is very much contrary to what we should expect. And this assessment of the divine motivations does not seem to comport very well with the Christian narrative, according to which God is, presumably, essentially loving, and therefore at least somewhat inclined to seek intimacy with creatures, even if not perhaps intimacy of these very radical kinds. As it happens, Murphy's view also seems hereby to place an additional apologetic burden on Christianity, compared with non-incarnational forms of theism. But of course, our purpose here is not to discuss Murphy's most engaging account - and I mention it just to show how it represents one end of a spectrum at the other end of which lie views of broadly the kind to which Swinburne subscribes.
Whereas for Murphy, the divine discretion is very wide, so that there is no expectation at all that God will make the best possible universe, supposing there to be one, since only reasons of holiness are 'requiring' for God, for Swinburne, it seems, God will always perform the morally best or jointly best action, if there is one -or failing that, some other action that is consistent with God's perfect responsiveness to moral considerations. On this picture, moral reasons may seem to account for the sum of divine activity, so that God becomes as it were simply a cipher, whereby moral truths concerning what would constitute the best world, supposing there is one, or in general, a morally optimal world, are given effect. Here the divine discretion seems to be, to this extent, very limited.
We should add that Swinburne clearly supposes that there is no best possible universe: in his pithy formulation of the point, 'plausibly, the more stars the better' (P. 9). It follows, then, that on his view, moral truths do not fix precisely what God will do, since there is no uniquely best creative act. This position seems to restore a measure of divine discretion. But now the difficulty is perhaps that on this account, God's choice between various worlds that are by assumption equally deserving
This case is presented in detail and with great subtlety in Mark Murphy, [Murphy 2021].
of creation will be simply arbitrary. To be a worthy object of divine creation, a world will need no doubt to meet a certain minimum threshold of goodness: for instance, it will need to be such that all the bads in that world are defeated by goods, in the requisite way. But it appears that many worlds will cross this threshold, and on Swinburne's view, unless further elaborated, it seems to follow that God's choice of which of these worlds to make will be simply arbitrary.
In sum, as it stands, Swinburne's account seems to imply that God's choices when creating and sustaining the universe are either fixed precisely by reason (in the case where one action is uniquely best) or to one or another degree arbitrary. And like Murphy's position, this account seems to comport less than perfectly with the Christian revelation, which suggests that God's creation of human beings is neither morally required, as a precondition of God making the best possible world, nor a matter of arbitrary choice. In brief, on the Christian account of the matter, I take it, the divine choice in making, and loving, human beings cannot be understood by supposing that we are members of the best possible universe, and that God is obliged to bring about that universe. If that were the case, then it seems we could claim our existence from God as a matter of right, and that is hardly the view that is implied in Christian devotional practice, where God is thanked for gratuitously conferring existence upon us. But at the same time, from the point of view of Christian revelation, it seems that God's choice to create human beings is not simply arbitrary: we are invited to think that God rejoices in the existence of human beings in particular, and indeed in the existence of the very human beings who in fact exist. And on the traditional view, it is for this reason that we can speak of God's regard for us as one of love6.
To the extent that this is the right way to think of the divine love, as it is revealed in the Christian scriptures, it seems we need another way of understanding the divine choices in creation, to allow for the possibility that these choices need not be fixed precisely by the objective goodness of the relevant outcomes, or be merely arbitrary, since they may instead flow from divine preferences which do not simply track the objective goodness of various possible states of affairs, and orient the divine will in that way, but which are, rather, subjective in character7.
Of course, this position raises a question about whether theism has now become a 'gerrymandered' hypothesis, in rather the way that naturalism will need to be a gerrymandered hypothesis, it seems, if it is to have a reasonably high predictive power with respect to, say, the fine-tuning of the universe. I think the answer to this question is in brief: no, for the reason that any of a number of divine 'psychologies' can play the role that we have just delineated - all that matters is that the divine psychology that in fact obtains should favour one or another set of creatures, for reasons that do not simply track the objective goodness of their existing. By contrast, a naturalistic account of, say, the phenomena of fine-tuning is committed to explaining why the cosmological forces should have precisely these values. I take it that this account of the divine motivations is not inconsistent with Swinburne's view, as articulated in his paper, to the extent that there is no question here
For a very clearly argued formulation of a view of broadly this kind, see [Adams 1972]. Compare the view developed by Brian Leftow [Leftow 2017: Ch. 9].
of God acting contrary to the balance of reasons. But it is an account that, so far as I can see, goes beyond what Swinburne commits himself to in this discussion - and an account that is, so far as I can see, required if our understanding of the divine choices in creation is to cohere with the Christian narrative.
To conclude, Richard Swinburne's contributions to natural theology over recent decades have defined the field for our time. In this response to his essay for this volume, I have sought simply to probe some points of detail in his position, with a view to inviting him to elaborate, rather than retract, his account of various matters. In particular, I have focused on his understanding of the relationship between the idea of divine goodness and the a priori probability of theism, the predictive power of theism relative to naturalism with respect to various generic features of the world, and the question of why God should have made this world in particular.
References
Adams, M. Horrendous Evils and the Goodness of God. Ithaca, NY: Cornell University Press, 1999. 220 p.
Adams, R. "Must God Create the Best?", The Philosophical Review, 1972, Vol. 81, No. 5, pp. 317-332.
Leftow, B. "Two Pictures of Divine Choice", in: Hugh McCann, ed. Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology. Oxford: Oxford University Press, 2017, pp. 152-172.
Murphy, M. Divine Holiness and Divine Action. Oxford: Oxford University Press, 2021. 272 p. Swinburne, R. The Existence of God. Oxford: Oxford University Press, second edition, 2004. 363 p.
Swinburne, R. The Christian God. Oxford: Oxford University Press, 1994. 261 p. Swinburne, R. The Existence of God. Oxford: Oxford University Press, revised edition, 1991. 327 p.
