Katherin Rogers's book on Anselm has a threefold objective. She expounds his views on free will in the light of all the other major elements in his system. She also situates his work historically in a way that, if vindicated, will add a new dimension to his historical importance. And she attempts to show that his views concerning free will and related matters can make a major contribution to contemporary debates in the philosophy of religion, and are in fact superior to the main competing views on offer.

Rogers's historical narrative begins with Augustine. In contrast to much previous scholarship, she interprets Augustine as holding a compatibilist view of free will throughout his career, including the early work, On Free Will, which has often been read as libertarian. She makes a compelling case for her reading, and she observes that competing interpretations often read Augustine in the light of libertarian intuitions (such as the necessity of libertarian choice for moral culpability) that he seems not to have shared. In favour of her interpretation is the fact that those who find libertarianism in Augustine are also bound to find in his thought deep-seated tensions and confusions. That such confusions exist cannot be ruled out a priori, but with a profound and systematic thinker such as Augustine an interpretation that preserves an overall consistency is to be preferred, all else being equal.

According to Rogers, the centuries between Augustine and Anselm saw relatively little progress on the topic of free will. Church councils set doctrinal guidelines but understandably failed to clarify the underlying philosophical issues. The two considerable philosophical thinkers during this period were Boethius and Eriugena. Boethius, according to Rogers, was a compatibilist like Augustine, and his contributions, while significant, did not lead to a clarification of libertarianism. Eriugena was probably a libertarian, but he gave no detailed analysis of free will. (And it is unclear whether Anselm had read Eriugena's writings.) Anselm, who followed Augustine closely in most respects, found a need to correct his views on matters related to free will. In doing so, he became the first Christian philosopher, and perhaps the first philosopher of any description, to give a careful, detailed analysis of libertarian free will. Clearly this narrative, for which Rogers makes an impressive case, does a great deal to enhance Anselm's already massive historical significance.

As one might expect, most of the chapters are devoted to detailed analyses of Anselm's views on free will and related matters. Topics discussed include: Anselm's classical theism; the purpose, definition, and structure of free choice; alternative possibilities and primary agency; the causes of sin and the intelligibility problem; creaturely freedom and God as creator omnium; grace and free will; foreknowledge, freedom, and eternity; and the freedom of God. Even a modest sampling of these riches would exceed the limits of this review; readers are urged to consult the book for themselves. The detailed interpretations will need to be scrutinized by other Anselm scholars, but for the most part Rogers's readings seem to be plausible and well-supported by the texts she cites. (One possible exception will be discussed below.) Where typical modern intuitions conflict with those that motivated Anselm's work, she rightly insists that the medievals' ways of seeing things should not be dismissed without a fair hearing.

It is not always entirely clear how much of Anselm's thought Rogers wants to affirm as her own view, but it's evident that her Anselmianism is deep-dyed. She recommends ‘Augustinian until proven otherwise’, as a sound principle for Anselm interpretation; perhaps we can take ‘Anselmian until proven otherwise’ as a good guide to Rogers's own thought. For the remainder of this review, then, I will take Anselm/Rogers as a dialogue partner, and carry on with her/them the sort of discussion she wants to invite.

Much of Anselm's treatment of free will is admirable, and provides insights that a contemporary libertarian should welcome. There are a couple of points, however, at which there is more to be said that, if taken account of, might alter the overall tone of his view. Like many subsequent libertarians, Anselm focused narrowly on moral choice as the key to free will, to the exclusion of other sorts of choice for which freedom is also important. One realm in which such choice occurs is artistic creativity: one would be hard put to argue that there is some one ‘best’ way to write a symphony or to paint a landscape. The artist chooses from a vast range of open possibilities, and selecting one set of such possibilities in executing a work by no means implies that others that were not selected are inferior.

A key idea for the understanding of free will is that there are numerous values that make their appeal to us, and these values are often ‘incommensurable’, in the sense that there is no fixed ratio or proportion that determines how much of one value is worth trading off for how much of another. Anselm obscures this by lumping all motivations for choice into the broad categories of ‘justice’ and ‘benefit’. In reality there are far more motives for choice than two, and recognizing this would have enabled Anselm to avoid the conclusion, presented and defended in Rogers's final chapter, that everything God does He does of necessity. True, God cannot do evil (as most contemporary philosophers of religion have agreed), but His actions in creation and redemption cannot and need not be thought of, in quasi-utilitarian fashion, as a simple ‘maximization of benefit’. If there is no one best way to compose a symphony, why should we suppose that there is some unique best way to create a universe?

This leads to a somewhat broader point. The classical theism Anselm derived from Augustine was a view formulated with no concern for libertarian free will. All of the impetus was for a view which took God to be the source and determiner of all things, including all human actions. Anselm, moved by certain libertarian intuitions, rightly sought to make a place for free will in Augustine's worldview. But given the overall lineaments of the worldview, the place made for free will must of necessity be a very small one, and free will seems very much an anomaly in the overall system. No doubt it is needed, as Anselm says, to guarantee that the merits and demerits for which we humans are rewarded and punished are genuinely our own. But in order to secure this, the classical worldview has to be modified in ways that many of its sterner proponents regard as wholly unacceptable. For instance, God's knowledge in some instances is determined by human actions, not the other way around, and God has to react to human acts and omissions.

No doubt Anselm is to be applauded for devising a way to incorporate human freedom while retaining as much of the classical conception as it was possible to do. In doing this, he struck a balance between God as source and determiner of all – creator omnium – and humans as making a genuine contribution to the course of events. But while his ingenuity is commendable, it is not so clear that the balance has been struck in the best way possible. It seems entirely conceivable that a viewpoint which accords more scope to creaturely freedom and creativity than Anselm felt himself able to do, might in the end fit better with the Christian scheme of creation and redemption. And in the process, libertarian free will might come to seem more a natural part of the overall pattern of things, and less an expedient that has to be squeezed in to avoid otherwise intolerable consequences.

The topics on which Rogers is most concerned to recommend the Anselmian position are those comprised under the heading, ‘Foreknowledge, freedom, and eternity’. She has fundamental objections to all of the currently popular views on these topics. Augustinian/compatibilist solutions are rejected because, in spite of all denials, they entail in the end that God is the author of sin. Open theism has little appeal for a philosopher with Rogers's classical sensibilities. It is a bit ironic, however, that some of her objections to it are equally applicable to her own view! (It has been shown repeatedly, by Flint, Sanders, and myself, that the implications for divine providence of divine timeless knowledge without middle knowledge are exactly the same as for open theism.) Molinism is rejected because the postulated realm of ‘counterfactuals of creaturely freedom’ which constrain God's possibilities for creation is incompatible with the classical doctrine of God as creator omnium. One commonly associates this objection with Thomist and other deterministic theologies, but it has force from an Anselmian perspective as well. True, the Anselmian (and any other theistic libertarian) must admit that God's possibilities of action are constrained by the free choices of the rational creatures. But this is a limitation self-imposed by God in virtue of God's decision to create such beings and allow them to choose freely. The limitations implied by the counterfactuals of freedom, on the other hand, are not self-imposed; they limit God's options whether or not He actually creates any free and rational beings. It is as though God had been born inside a cage with bars, and could never escape from the cage no matter what He might choose or desire to do.

The Anselm/Rogers view begins with the nature of time. According to Rogers, Anselm was what we would now term a ‘four-dimensionalist’ – indeed, ‘He is the first clear and consistent four-dimensionalist’ (183). This means that all temporal objects and events are literally present, and available to be known, in God's eternity. There is indeed a kind of necessity involved; what God knows in eternity cannot fail to exist in time. But this is not a necessity of compulsion but rather what Rogers terms a ‘consequent necessity’, the necessity that results from the actual occurrence of an event; it is entirely compatible with free will. Anselm states,

Rogers backs this up with a typical Boethian appeal to the timeless nature of God's knowledge: ‘What to any given temporal perceiver at any given time is past, present, and future is all just there, equally real and directly present to God. God knows what is the case by “seeing” what is the case’ (173).

There is much in all this that calls for discussion – indeed, for a lengthier discussion than is possible in this brief review. To begin with, the description of Anselm as a four-dimensionalist is controversial and by no means obviously correct. Anselm certainly does hold that all temporal objects and events are present – literally present, not merely in representation – in eternity. So far, then, Rogers's interpretation is on solid ground. But more than this is needed for four-dimensionalism. Four-dimensionalists hold that all these objects and events are in every temporal respect on a par with each other; there is no such thing as a unique temporal present. ‘Now’ is an indexical term, used by a speaker to designate the moment at which she is speaking, just as ‘here’ is a spatial indexical, used to designate the point in space at which one is located. But there is not, cannot be, any objective, speaker-independent fact as to which place is ‘here’ or which time is ‘now’. The difficulty here is that Anselm repeatedly speaks as though in time there is indeed a fact of the matter concerning what has already transpired and what has yet to occur. Anselm comments on the difference between ‘eternal’ creatures, such as angels and human souls, which will never come to an end, and God: ‘And thus you are always beyond (ultra) [these “eternal” creatures], since you are always present “there”, or rather since it is always present to you, which for them has not yet arrived’ (179). The future is ‘there’ for God, but it has not yet arrived for the creatures; this certainly suggests that there is in fact some stage of their existence at which the creatures have arrived – that is, there is an objective temporal present. Similar quotations could easily be multiplied.

Now, a four-dimensionalist can, with some ingenuity, accommodate locutions such as these. The four-dimensionalist will explain that while, with reference to some particular point in time, a certain event ‘has not yet arrived’, there also exists the future temporal moment at which the event has arrived, and in which the same individuals for whom at the earlier moment it had not arrived experience it as present. So I am eternally just beginning to read Rogers's book, and you have eternally just finished reading my review, both of these events occurring at different temporal co-ordinates than the one at which I am writing these words. The problem is, however, that Anselm never gives these explanations. So there is a prima facie case that his references to a temporal present ought to be taken at face value, in which case he is not a four-dimensionalist. If we are going to characterize him in terms of present-day positions on the issue, he may be closest to the ‘moving spotlight’ view, according to which all events of all times do co-exist, but there is all the same a ‘moving present moment’ which determines which time is now.

Does the Anselm/Rogers view provide a solution to the problem of reconciling divine knowledge of the future with libertarian free will? Almost certainly it does not. It is clear that, for both Anselm and Rogers, all temporal objects and events coexist timelessly in eternity. But as Lawrence Sklar has pointed out, this negates the alternative possibilities that are crucial for libertarianism. (Rogers, following Anselm, repeatedly emphasizes the need for alternative possibilities for free will.) Consider Annie, an agent who is in the process of making a libertarian choice between X and Y. Suppose she, in fact, chooses Y; in doing this Annie brings about that her choice of Y exists eternally as part of the four-dimensional continuum. But in order for her choice to be free it must have been really possible, and really within Annie's power, to choose X instead of Y. And of course, her choice of X would bring about that it is her choice of X rather than her choice of Y that eternally exists as part of the continuum. But this means that, at the time when Annie makes her choice, ‘there are future actions of [Annie's] which timelessly exist in the divine eternity which are such that it is in [Annie's] power, now, to bring about that those actions do not exist in eternity’. Rogers quotes the italicized sentence from an earlier article of my own (‘The absence of a timeless God’, in Gregory E. Ganssle and David M. Woodruff (eds) God and Time (Oxford University Press, 2002), and responds, ‘Of course this condition cannot be met’ (171). I certainly agree that it cannot – but, pace Rogers, what this shows is that the Anselm/Rogers view does not, as they both claim, make it possible for libertarian free will to co-exist with the presence of all future events in God's eternity.

Whatever one's view on the points in contention, Rogers deserves our thanks for this well-researched and finely crafted study. And in view of the current popularity of the four-dimensionalist view of time, it is very much to the point for Anselm's views on these topics to become an active part of the current discussion of the relationship between God and time.