Currently, the classical doctrine of God is being challenged by one wing of the broad evangelical theological constituency. A major contribution from John Sanders, one of its representatives, has now appeared under the title The God Who Risks: A Theology of Providence. If Sanders is right, the first chapter of the Gospel According to Luke (vv. 26–38) should be read as follows:

In the sixth month, God sent the angel Gabriel to Nazareth, a town in Galilee, to a virgin pledged to be married to a man named Joseph, a descendant of David. The virgin's name was Mary. The angel went to her and said, "Greetings, you who are highly favoured! The Lord is with you." Mary was greatly troubled at his words and wondered what kind of greeting this might be. But the angel said to her, "Do not be afraid, Mary, you have found favor with God. You will be with child and give birth to a son, and you are to give him the name of Jesus." [Thinks: "That's the way I'll put it to encourage her. Actually, if she does not cooperate, she will not be with child and her high favor will turn into crushing personal failure. Still, if she does cooperate, it is surely most unlikely after this announcement that she will give him any other name except the name of Jesus.] He will be great and will be called the Son of the Most High [though there is no guaranteeing what he will do, and I suppose he could veer way off course]. The Lord God will give him the throne of his father David, and he will reign over the house of Jacob for ever; his kingdom will never end [at least that is the scenario we have every reason to expect, since God is incredibly resourceful at bringing these things about. Nor will you tax his resourcefulness too much, Mary, if you won't bear the child. Of course, if you refuse, we'll have to find someone else who will have some reason for going to Bethlehem for Jesus to be born, because that fits best with the prophecy of Micah. Alternatively, we could leave Bethlehem out of it and find some other way to make Micah's words prophetically meaningful. Still—it can be done]."

"How will this be," Mary asked the angel, "since I am a virgin?"

The angel answered, "[Well, strictly I can tell you only how it would be, if you were to consent, but not precisely how it will be, since that depends on you]. The Holy Spirit will come upon you, and the power of the Most High will overshadow you. So the holy one that will [in such an event] be born will be called the Son of God. [As it turns out] Elizabeth your relative is going to have a child in her old age, and she who was said to be barren is in her sixth month. [Now the ideal role of that is that he should become the Baptist, the prophetic forerunner to his cousin, Jesus, and God is very much hoping that you will not disappoint him and force a change of plans]. Nothing is impossible with God."

"I am the Lord's servant," Mary answered. "May it be to me as you have said." Then the angel left her [relieved].

If any think that this is a caricature, let them read the whole volume by Sanders, including the treatment of Mary (pp. 92–94). And if any think it exceptional and unrepresentative of the author's theology, let them reread it and dwell instead on the treatment of Judas, Gethsemane, and Cross (pp. 98–106). The Cross was a contingent event—it might not have happened and is not definitely predicted in the Old Testament. The options for Jesus and God, as regards executing the project of salvation, narrow as Jesus' ministry progresses until in Gethsemane, "Father and Son, in seeking to accomplish the project, both come to understand that there is no other way." But before then, if not right then, the options were not foreclosed. In that case, Jesus' predictions of his death were a fallible, though justifiable, prognosis; he strongly suspected his imminent death when he was in the upper room and presumed that a new covenant would be instituted through his blood. But the divine establishment of a sacrificial system in the Old Testament did not guarantee it, and at the time of its institution, God could not have been justifiably as confident that the death was prefigured as the writer of Hebrews apparently was.

It is important to appreciate what is at stake here. On the face of it, the thesis that one ought to examine is the claim that God takes risks. On the traditional view, which Sanders wants to revise radically, God is immutable, perfectly knows the future, and, on some standard accounts, foreordains it exhaustively. The questions of God's power and love, eternity and impassibility are all implicated. Sanders wishes to argue that God takes risks. He does not perfectly control the future. He could if he wanted to, but voluntarily enters into a free, loving, reciprocal relationship with his creatures. They, like him, are genuinely free. God has intentions and desires, but he risks their nonrealization. He is genuinely surprised and disappointed, as we learn from the very earliest chapters of Scripture (Genesis 6) and Jeremiah 3:7, for example. But he is infinitely resourceful, guarantees the occurrence of at least some things, and can be entirely trusted to bring the whole story of salvation to a resounding conclusion through the cooperation—or despite the noncooperation—of his creatures.

This is our God, and Sanders's reading of Bethlehem and Calvary follows from this. Despite the subsequent philosophical discussions, the basis for theology is putatively biblical, so Scripture is the first and important port of call. Its language about God repenting, being disappointed, changing his mind and not knowing, must be taken as it stands. We have no right to make it mean something else in light of some other principle for reading the texts. Scripture itself nowhere gives us permission to do so. The project of deterministic theology, to which Sanders opposes his divine risk theology, is driven by a philosophical tradition that has excessively influenced our reading.

On the face of it, we have the issue laid out before us here. And this is no maverick idiosyncrasy within current evangelicalism. While pointing out that differences exist among the theologians concerned, Sanders is participating in a project associated with others, such as Clark Pinnock, which presses for a truly relational theism, a nondeterministic outlook on the relation of God to humans, a relation that must be truly reciprocal and that cannot be so if God controls everything. Those who take different sides in the debate agree on its importance: the way we see God decisively shapes the ethos of our lives, our words, and our worship. We must try to get things right.

If this is what it is all about, then the natural response is a direct theological assessment of the claims. Back we go to those things that occupied Augustine and Pelagius, Calvin and Arminius, Wesley and Barth. True, some of us will also conclude that most of what can be usefully theologically said probably has been said; that renewed exegesis per se will not settle the question; that analytic philosophy will not solve it. Still, we'll join the fray.

Despite all appearances, however, these are not the things primarily at stake. Challenging classical beliefs in divine impassibility and immutability or arguing for libertarian freedom and divine nondetermination of all that occurs is one thing. Portraying God as Sanders does in this book is quite another. For the beliefs reported at the beginning of this article are expounded in statements that will cut across the religious sensibilities of many sympathetic to some of the project. There is much more going on here than the stark theological challenge of a theology of risk. Let us give examples of such statements.

1. God expected humanity not to fall into sin: while the expectation was mistaken, it was reasonable and justified (p. 45).
2. Human sin, as recorded in chapter 6 of Genesis, causes God regret and disappointment, shock and grief, such that, as Philip Yancey says: "God learns how to be a parent," and, as Sanders himself puts it: "Whatever God decides, he will never be the same again. God now knows what it is to experience grief" (p. 49).
3. After the flood, "the sign of the rainbow that God gives is a reminder to himself that he will never again tread this path. It may be the case that although human evil caused God great pain, the destruction of what he had made caused him even greater suffering. Although his judgment was righteous, God decides to try different courses of action in the future" (p. 50).
4. When God says, "I thought Israel would return to me but she has not" (Jeremiah 3), not only is he "explicitly depicted as not knowing the specific future," but also "God himself says that he was mistaken about what was going to happen" (p. 74).

Why will some of us feel in the marrow of our bones that all this must be wrong and that one does not need to adjudicate all the substantive issues surrounding risk in order to conclude that? According to Sanders, if we find these statements doing violence to our concept of God and to our religious and theological sensibilities, it is because we have read our Bibles informed by a misguided tradition. After all, is he not just following the scriptural text?

Now one certainly appreciates the force of this general point. Doubtless, we read the Bible through lenses that ought to be corrected; there is no telling how far back in the tradition deeply ingrained habits, that ought to be corrected, go; sola Scriptura commits us to a radical and often painful principle of regular scrutiny and preparedness to give up cherished ideas.

But having said that, one must also allow that a cumulative reading of the Bible may intuitively disallow Sanders' description. Such intuitions can be profoundly objective prior to the provision of a positive considered alternative, just as in the case where readers feel that, whatever else he was saying, Jesus was never enjoining emotional hatred toward parents and relatives (Luke 14:26), and proceed to guide their exegesis by that light. The point is not to institute comparison between the relative clarity of Scripture on that particular question and the questions that concern Sanders, but to remark on the propriety of referring to intuitions.

However, let's try to produce something with flesh on it. Behind the foregoing response is the conviction that the cumulative force of Scripture goes against Sanders's reading of the texts. The issue is hermeneutical, and one of the most surprising features in Sanders's account is the failure to tackle hermeneutical issues seriously while accusing his opponents (sometimes justly, no doubt) of hermeneutical prejudice. But take his claim that God is open-minded or changes his mind in response to petitionary prayer. My wife has just left the house in a tearing hurry, having executed the physically near-impossible task of answering around five different queries from two children, executing necessary duties both initiatively and responsively, and finally interrupting her husband as he penned this article with a request that he found it hard to get his mind to attend to instantly because of his deep concentration on this noble assignment. All this has happened in about two minutes and 45 seconds. If God had no one in the world to attend to except Abraham and Sarah, Ishmael and Lot, while he harkened to and responded to one of Abraham's prayers, perhaps he might have kept one of the dialogues going in exactly the way narrated by Scripture. But if, while Abraham prays and God ponders and answers, he is also engaged in running the entire world, all within our time frame, the risk to his health and well-being is infinitely severer than any risk courted on Sanders's account.

Now the riposte is obvious: of course, God cannot possibly be operating within our time frame. That is true, but its hermeneutical implications are completely ignored by Sanders. For it quite simply means that the narrative cannot be taken as it stands. This does not at all mean that we ignore what the narrative says. One is free to produce a scheme that reminds us that one day is as a thousand years for God, so that what for Abraham took two minutes for God to think about, from God's standpoint took several weeks, during which he was able to get other things done as well. The point is not to deny directly (nor to affirm) that God leaves his options open in relation to our petitions. The point is that the temporal scene for God must be radically different from the temporal scene for us. This immediately requires that we read the texts in terms of some hermeneutical principle of accommodation.

We are not remotely alerted to this by Sanders. Indeed, he challenges his opponents as to the provenance of their doctrine of accommodation when they interpret language of divine "repenting" or "changing his mind" in the traditional way. However, he himself must derive from somewhere the notion that God does not possess physical characteristics despite being portrayed as such. Of course, one might contend that this simply means that Sanders requires an appropriately refined hermeneutic in relation to the temporal and spatial characterization of God and that the fault is simply one of omission. But it is more than that: the ramifications are significant. For the whole morphology of surprise and disappointment, ignorance and risk takes on a different countenance once we dwell on the fact that in half an hour of Abraham's time or Jeremiah's time, God has spent an all but infinite number of half-hours with countless other individuals—let alone in all the uninhabited spaces of his world, all at the same time.

This response to Sanders admittedly ignores an important strand in his account. This has to do with metaphor, but his exposition is very unclear on this point. One reads his work for the most part as though he were referring literally to divine risk, repentance, or change of mind, and the logic of his discussion seems to vindicate such a reading. But actually he tells us that the terms are metaphorical. He even maintains that he is not committed to the temporality of God. In another context, we should inquire about the coherence on its own terms of an account that consistently seems to treat metaphors literally. However, I suspend judgment on that here, so as not to be deflected from pursuing what I have suggested is really at stake in this work. To get at this, let us move on to a second hermeneutical question.

Sanders's method is to take us through the Scriptures selectively according to the flow of the story of Israel and Christ. The result is that the narratives in Genesis and Exodus are hermeneutically central. But how can this be justified methodologically? It is rather as though Proverbs 8 were given hermeneutical control in Christology prior to the study of the New Testament data.[1] Sanders refers to passages such as we find in Job, in Psalm 139, and in Isaiah 40ff. that appear to announce a different concept of deity, one whose knowledge, foreknowledge, or control extends significantly beyond the one Sanders has limned. But why does the Wisdom literature at these points not distill our experience of God far better than the literal account of divine ignorance and surprise, particularly when we move on from it to prophets and apostles? Is not the impact of a canonically sequential reading to create in us a sense that the prophets, especially the major ones, present to us informed reflection on and understanding of the nature, ways, and purposes of God, Israel, and the peoples as they have been narrated to us in the preceding historical books?

These prophecies are hermeneutically significant. "Declare to us the things to come, tell us what the future holds, so that we may know that you are gods," says the Lord (Isa. 41:23), challenging his petty rivals to demonstrate their divinity by their foreknowledge. The divine announcement, where by "he foretold the former things long ago" (48:3) is not remotely like a "forecast" of what God "thinks might happen" (p. 131). Isaiah would not have recognized the Holy One of Israel in Sanders's depiction any more than Job would have taken him on as a replacement for the failed counselors. Whatever Sanders might make of the general matter of reading Scripture backward as well as forward (and he might either defend starting with Genesis or point out that he is not as dependent on the early accounts as I am suggesting), he agrees that Christology in particular must be hermeneutically pivotal, which is why we started the article there. Yet this is just where his account is hopelessly strained.

Other funny things happen in his hermeneutics. In attempting to defuse the argument that the narrative of Joseph shows how God controls everything, Sanders points out that God is absent from large stretches of the text. This is extremely curious; one wonders how he would resist the argument that the Book of Esther entails atheism. Still, one wonders whether hermeneutical shortcomings explain the religious sensibility evidenced in Sanders's work. Let the challenge to Augustinian-Calvinist "determinism" remain, for better or for worse. Let an account of divine self-representation in Old Testament narratives be offered, for richer or for poorer.[2] And let engagement with the philosophical issues at stake continue, unto increasing wisdom or confusion. When all is said, the real worry about Sanders's volume is the lack of any embarrassment at the emergent portrayal of the ways of God.

Perhaps this article will seem to be driven by a massive subjective reaction that bypasses the hard arguments. I do not for a moment believe they should be bypassed: in their place, the detailed biblical, theological, and philosophical discussions are important, and the point is not at all to come down for or against much that is argued or concluded by Sanders. But the business of delimiting boundaries for what is acceptable is equally as important as trying to answer theological questions directly. The latter has to do with the relative merits of theological moves. The former has to do with the validity of fundamental theological rules. Whether or not they are transgressed by Sanders's opponents, this article is meant to express the worry that they are transgressed here. It is to be hoped that this work will alert us to the need of staking out boundaries of permissible biblical interpretation in the context of discussion of this particular theme, as well as stimulate us to reach our own conclusions.[3]

Stephen N. Williams is professor of systematic theology at Union Theological College, Belfast, Northern Ireland.

1. This point is exaggerated, but it would be fruitful in this connection to revisit the hermeneutical principles of Athanasius in his Christological discussion with the Arians—Proverbs 8 was an important text in the controversy: see Jaroslav Pelikan, The Emergence of the Catholic Tradition (100–600) (Univ. of Chicago Press, 1971), pp. 191–200.

2. My point is not that "didactic" portions of Scripture take precedence over the "narrative" ones; it has more to do with reading Scripture according to the analogy of faith. Of course, knotty problems arise in relation to the early narratives, and failure to address them here is not a sign that they should be suppressed.

3. To use language such as "permissible" here seems arrogant: Who, pray, is seated in such a place of authority as to dispense permission? However, everyone involved in theological work has implicit convictions about what is credible and incredible, plausible and implausible, permissible and impermissible in biblical interpretation. And, of course, one must be sensitive to the distinction between what one regards as linguistically infelicitous and what is doctrinally controverted. I suspect that many who will reject the language of risk on the lips of Sanders will mind it less on the lips of C. S. Lewis, Pilgrim's Regress (London: Fount, 1998) p. 228. Finally, in the present climate, it ought to be said that we are no more concerned in this article with casting aspersions on people's Christianity than is Sanders himself when he deals with his theological opponents.

Copyright © 1999 by the author or Christianity Today/Books & Culture Magazine. Click here for reprint information on Books & Culture.