William Lane Craig is Research Professor of Philosophy at Talbot School of Theology in La Mirada, California. He lives in Atlanta, Georgia, with his wife Jan and their two teenage children Charity and John. At the age of sixteen as a junior in high school, he first heard the message of the Christian gospel and yielded his life to Christ. Dr. Craig pursued his undergraduate studies at Wheaton College (B.A. 1971) and graduate studies at Trinity Evangelical Divinity School (M.A. 1974; M.A. 1975), the University of Birmingham (England) (Ph.D. 1977), and the University of Munich (Germany) (D.Theol. 1984). From 1980-86 he taught Philosophy of Religion at Trinity, during which time he and Jan started their family. In 1987 they moved to Brussels, Belgium, where Dr. Craig pursued research at the University of Louvain until 1994.
In his “Perils of Paul” William Hasker intimates that our imaginary friend Peter needs to seek out a new philosophical advisor, given Paul’s missteps in counterfactual logic. But has Paul erred in the ways alleged by Hasker?
It is important to set the context here. Peter’s concern is that Paul’s proffered solution to the problem of Christian exclusivism—that is, that it seems incompatible with God’s love and power that some people should be damned due to historical and geographical accident—may itself involve people’s salvation or damnation’s being the result of the accident of our obedience or disobedience to the Great Commission.
Paul’s response is to express doubt that we have the power to act in such a way that if we were to act in that way, the balance between saved and unsaved would have been better as a consequence. That is to say, he attempts to undercut Peter’s objection by challenging the assumption that improvement of the salvific balance is counterfactually dependent upon our actions.
Why does Paul doubt this assumption? Because, he says, if there is such a thing as an optimal balance, the goodness of God would lead Him to prefer ceteris paribus a world exhibiting such a balance. So if there were any other world feasible for God which had a better balance between saved and lost than the actual world, God would have preferred that world over the actual world. But then that world would have been actual rather than this one. Given that our world is actual, it must be a member of that class of feasible worlds exhibiting an optimal balance between saved and lost. So, Paul, concludes, the counterfactual:
A. If we were to be more obedient to the Great Commission, the balance between saved and lost would not be improved.
Hasker contends that Paul is mistaken here. Not only does he think that Paul’s reasoning is confused, but he also thinks that from Paul’s own assumptions it clearly follows that:
A¢. If we were to be more obedient to the Great Commission, the balance between saved and lost would be improved.
Why does Hasker think Paul confused? Although he is not as clear as one might wish, I think that Hasker’s fundamental point can be expressed as follows:
Logically prior to His divine creative decree, God finds Himself confronted with a class of counterfactuals of creaturely freedom which we may, borrowing Thomas Flint’s terminology, call a creaturely world-type. Unlike counterfactuals of divine freedom, these counterfactuals of creaturely freedom lie outside God’s control and so determine for Him a galaxy of feasible worlds. Worlds constituting galaxies determined by other world-types are not actualizable by God and are, hence, infeasible for Him. Given God’s goodness, He prefers, ceteris paribus, worlds in which every created person comes freely to salvation. Unfortunately, given the creaturely world-type that is true, such worlds are infeasible for God. So God must choose from among worlds in the galaxy of worlds feasible for Him. Here he will prefer worlds exhibiting an optimal balance between saved and lost and select one of these to be actual. Thus, the actual world does have an optimal salvific balance within the galaxy of feasible worlds. Never mind that there are possible worlds exhibiting a better balance: such worlds are infeasible for God, and so He cannot be held responsible for not choosing such a world to be actual. We cannot bring it about that the galaxy determined by the true creaturely world-type should include worlds having a better balance than the class of optimally balanced worlds. This is just a point of logic, for if we were to act any differently in the same circumstances, a different world-type would be true and, hence, a different galaxy would confront God. This is the element of truth in Paul’s reasoning.
But Paul’s confusion becomes evident when we reflect that world-types are person-relative. The creaturely world-type is the world-type-for-God, containing no counterfactuals of divine freedom, since He has control over these. The world-type-for-Peter analogously does not include counterfactuals about his own free decisions, since he has control over these. Unlike God, Peter has the power to act in such a way that a different creaturely world-type would have been true. Specifically, Peter could act in such a way that he would be more obedient to the Great Commission than he actually is. Were he to do so, God would find Himself confronted with a different creaturely world-type and, hence, a different galaxy of worlds to choose from. In that galaxy the optimal balance might well be better than in the galaxy with which God is actually confronted. Thus, Peter may have the power to act in such a way that, were he to act in that way, the salvific balance would be improved.
Hasker is right: Paul is confused. He inferred rashly from the fact that the actual world has an optimal salvific balance among worlds feasible for God that (A¢) could not be true. Although (A¢) is not strictly a counterfactual of creaturely freedom, being of the wrong form, and so not technically part of the true creaturely world-type, still if it is true, it is so logically prior to the divine decree and so part of what we might call a “thick” creaturely world-type and so true in all worlds in the galaxy determined by that type. Paul therefore thought that God could weakly actualize the state of affairs described in the consequent by actualizing the relevant circumstances implied in the antecedent, circumstances under which the antecedent would be true. But in that case there would be worlds feasible for God in which a better salvific balance exists than in the actual world—which contradicts the hypothesis that the actual world exhibits such a balance.
But Hasker in effect holds open the possibility that God could not actualize the state of affairs described in the antecedent. No matter what circumstances He created us in, we should not have been more obedient to the Great Commission. Since Paul is willing to entertain scenarios not involving us as intermediaries to reach the lost, Hasker must say that there is no world feasible for God in which any persons are more obedient to the Great Commission than we actually are. This may seem far-fetched, but it is possible. Hence, Paul should not have concluded that it lies within God’s power to weakly actualize the antecedent of (A¢) and thereby to weakly actualize its consequent. Therefore, (A¢) might be true even though the actual world has an optimal salvific balance within our galaxy.
Still, what Paul said was that he doubted that we have the power to improve the salvific balance, and such a doubt seems rational. We do not know whether other creaturely world-types available to us are ones determining galaxies exhibiting a better optimal balance than our galaxy. That is sufficient to undercut, if not rebut, Peter’s objection.
But Hasker maintains that on Paul’s assumptions it clearly follows that (A¢) is true. Hasker lists the following assumptions:
1. Christ has given a Great Commission to his followers.
2. God has brought it about that, in times and places reached by the Gospel, there are numerous persons who would accept salvation through Christ if it were offered to them.
3. God has brought it about that, in times and places not reached by the Gospel, the inhabitants are persons who are transworldly damned and would not accept the Gospel even if it were preached to them.
From these assumptions, says Hasker, it “clearly follows” that:
4. The more obedient Christians are to the Great Commission, the more saved persons there will be and the fewer lost.
Now clearly (4) does not follow from (1)-(3). It is compatible with (1)-(3) that there should come a time at which, no matter how zealous our evangelistic outreaches, no matter how earnest our exhortations, no more converts would be won by our further obedience to Christ’s Great Commission. But that is incompatible with (4). Thus, (4) could be false and (1)-(3) true.
Perhaps we could avoid this problem by amending (2) to something like
2¢. God will bring it about that, as new places are reached by the Gospel, there will be persons there who will accept salvation through Christ
which might be thought to represent fairly Paul’s position. But even so amended, these assumptions do not imply (4), for one could infer with equal justification that
5. The more obedient Christians are to the Great Commission, the more widely geographically distributed will the saved be.
Paul does seem to infer something like (5), which is why Peter suggested that we should then just all go to the Bahamas, trusting that God would concentrate all the saved there. Therefore, Hasker has failed to show that (A¢) follows from Paul’s assumptions.
It seems that Hasker has not appreciated sufficiently that counterfactuals like (A) and (A¢) are just as beyond Peter and his fellow believers’ control as they are beyond God’s. He seems to think that because Peter and his fellow-believers can bring it about that they are more obedient to the Great Commission, they can by doing so bring about a more favorable balance of saved and lost. That this does not follow can be made plain by making explicit the different agents involved, as in:
A*. If we were to be more obedient to the Great Commission, proportionately more people would have freely embraced salvation and fewer freely rejected it.
(A*), if true, is part of God’s thick world-type and so beyond His control. Similarly, although Peter and his fellow-believers could bring about the truth of (A*)’s antecedent, whether (A*) is true will depend upon the free response those reached with the Gospel. Thus, the proposition expressed by (A*) is also beyond Peter’s control, since it summarizes propositions of the form:
A**. If S were in C (C including being evangelized by more obedient Christians), S would freely embrace salvation rather than reject it.
Propositions of the form of (A**) are counterfactuals of creaturely freedom which are pre-volitional for Peter and so part of the world-type-for-Peter and beyond his control. Peter can bring about the circumstances C of sharing the Gospel with S, but S’s free response is just as beyond Peter’s control as it is beyond God’s. It is the lost themselves who have control over whether (A**) is true.
The bottom line seems to be that we just do not know whether (A) or (A¢) is true, if either is true. That is enough to undercut the objection that Paul’s Molinist solution makes people’s salvation hang on the accident of our obedience to the Great Commission.
Still and all, one can only heartily agree with Hasker’s conclusion that a devout life is no protection against fallacious reasoning. Peter would do well to take Bill Hasker among his philosophical advisors, for one will find his advice inevitably stimulating and penetrating, even if, in the end, not always convincing.
 “The Perils of Paul: A Response to William Craig,” this journal.
 Hasker thinks that on Paul’s proposal there cannot fail to be an optimal balance of saved and lost in the galaxy of feasible worlds. But when one reflects that an optimal balance involves both the ratio of saved to lost as well as absolute numbers, this is not obvious. Perhaps as the numbers increase the ratio declines, with no intrinsic maxima. In that case an optimally balanced world would be like the chimaera of the best of all possible worlds.
 Notice that Paul treats worlds which are incompatible with God’s goodness as still feasible for God, even if rejected by Him. Thomas Flint’s approach is different. He asks:
“Could it be the case that, even within the feasible galaxy, there are certain worlds which God cannot actualize—because, e.g., doing so would be morally abhorrent? I think not. The Molinist will insist that, if a certain world is so atrocious that actualizing it would be inconsistent with God’s moral perfection, then the world in question is not genuinely possible, and so is not a member of the feasible galaxy in the first place” (Thomas P. Flint, Divine Providence, Cornell Studies in the Philosophy of Religion [Ithaca, N.Y.: Cornell University Press, 1998], p. 51).
Paul is presupposing a stricter modality than Flint. If we adopt Flint’s view, then every world feasible for God has an optimal salvific balance. Indeed, given God’s essential goodness, any galaxy of feasible worlds, regardless of which creaturely world-type is true, is composed exclusively of optimally balanced worlds. Such a conclusion would not materially affect the issue before us, however, since Paul recognizes that the optimal balance may be different from galaxy to galaxy, and Hasker’s contention is that we have it within our power to bring it about that a different creaturely world-type is true so that a different galaxy is feasible for God.
Notice, too, Paul’s ceteris paribus clause. This is a huge assumption, indeed, an arguably false assumption, since worlds having better salvific balances might have other overriding deficiencies that make them less preferable. Even if we can act so as to bring about a better salvific balance, those worlds in which we do so may be less preferable.
Thus, Paul is by no means confused, as Hasker alleges, when he asserts that whichever world is actual that world has an optimal balance between saved and lost. God’s essential goodness guarantees that only worlds belonging to that class of worlds exhibiting an optimal balance in their galaxy are candidates for actualization. So if the Bahamanian world b were actual, it would have an optimal balance of saved and unsaved.
 See, e.g., his comment that if we were to act in a certain way “the actual world would be part of a different galaxy than the galaxy to which abelongs.” I take it that he does not envision inter-galactic travel by worlds, but rather another galaxy’s being feasible than this one and so another world actual than this one. In Flint’s terminology, a different chosen world would be actual, and the question is whether it has a better salvific balance than this world. Hasker errs, moreover, in thinking that the reason that a would be infeasible if bwere actual is because the counterfactuals of freedom stating what Peter and others would do when confronted with the call to the mission field would have different truth-values. In fact, precisely the same counterfactuals concerning them could be true, since it is no part of the envisioned scenario that Peter and his fellow believers find themselves in the same circumstances in bas in a. Hasker notes that he is assuming that some persons in bwould reject the missionary call in exactly the same circumstances in which they actually accept it. Such an assumption is gratuitous. Hasker recognizes that because circumstances in bbecome increasingly divergent from those in a, the truth value of counterfactuals concerning subsequent choices might not have different truth-values. But he fails to appreciate that because C includes the whole history of the world, at the first difference between band a different counterfactuals immediately come into play, so that no different truth-values are required in bfor propositions about the choices which Peter and others like him make in a.
Counterfactuals of creaturely freedom have the form If S were in C, S would freely do A, where S is a free creature, C is a set of circumstances including the causal history of the world up until the time of S’s decision, and A is some action.
 In the case of (A) and its variants C will include various gifts of divine grace and solicitations of the Holy Spirit, but that fact does not make (A) or its variants only post-volitionally true, for C may include any envisioned circumstances, even circumstances concerning God’s actions; indeed, such truths are crucial to a Molinist doctrine of providence and predestination.
Since neither (A) nor (A¢) is a counterfactual of creaturely freedom, there is no guarantee that the Law of Conditional Excluded Middle holds for it. Indeed, given the multiplicity of situations they envision, it may seem intuitively true that:
A***. If we were to be more obedient to the Great Commission, the balance between saved and lost might or might not be improved.
 I myself thank Bill for stimulating discussion of this response.
©William Lane Craig. Originally appeared in Philosophia Christi 6 (2004): 273-8. Reposted with permission.