The triumph of the eternal decree of God over history is just as much a problem as the triumph of history over the eternal decree.
In an attempt to stave off Arminianism (a commendable task!) deniers of common grace have reasoned that God could in no way have any favor toward the reprobate. To say that God favors the reprobate is to introduce a contradiction between the eternal will of God and his works in history. God would be “two-faced.” He would will one thing, but then do another. Therefore, God is not gracious toward the reprobate nor does God genuinely desire for the reprobate to believe the Gospel and be saved. In this mode of theology, consideration of the eternal decree trumps how we are to understand God’s works in history.
One of the prominent proponents of this position was Herman Hoeksema. Van Til has much to say against his denial of common grace, but central to his critique is the following:
Hoeksema never answered adequately the charge that on his view the elect can never in any sense have been under the wrath of God and Christ need not have died for them in history. Hoeksema took no note. (Common Grace and the Gospel, 251).
Van Til’s point is that if we identify God’s attitude toward man in time with God’s predestination then we can never speak about the elect as ever having been under divine wrath. Hoeksema’s unqualified supralapsarianism has its center in the proposition that what happens last in the order of history comes first in the order of the eternal decree. This means that God chose the elect and the reprobate quite prior to his decree to create or ordain the fall. Each person’s eternal destination is determined apart from all the means that lead there onto. The means are swallowed up in and by the end.
This means that the reprobate cannot have any favor with God. But—and this is Van Til’s point—the elect can never be said to ever have been under God’s wrath. In other words, the person who becomes a Christian later in life can in no meaningful way be said to have transitioned from being under God’s wrath to being under his grace. On the terms of Hoeksema’s supralapsarianism, the elect person was always under God’s favor, even as an unbeliever.
This also means that when Adam fell there was no real transition from being in an estate of favor to an estate of wrath. What about the reprobate who were “in Adam” before the fall? Can we say, biblically, that the reprobate in Adam before the fall were under God’s wrath despite the fact that humanity was still innocent? Furthermore, when Christ came to die on the cross we cannot say that—relative to the elect—there was any real transition from an estate of wrath to an estate of grace. In fact, on Hoeksema’s presuppositions, there really was no need for Christ to come and atone for sins at all. Election is ultimate, and the elect were chosen quite irrespective of God’s decree to redeem in Christ. Christ and his work become somewhat of an unnecessary afterthought.
And so what Hoeksema ends up doing is making history somewhat of a farce. Historical dynamics are not real manifestations of moving from grace to wrath or wrath to grace. God does not have any real interaction with even his elect. His elect can never be under his wrath before their conversion, nor can they come under God’s Fatherly displeasure after their conversion. Likewise, the reprobate can never experience the true and genuine favor of God, nor ever hear a true and sincere call to repent and be saved. God does not really desire the repentance of the wicked.
So, how then are we to understand the relation between historical transitions and God’s eternal decree? Van Til proposes a Christian idea of “limiting concepts.” Limiting concepts, understood Christianly, has its basis in a Christian idea of mystery. In other words, there are things we simply do not know. In revelation, we are given knowledge of certain things, but not all things. God and his decree remain always incomprehensible to us. And where God’s revelation ends, there we must be content with mystery.
God does reveal to us that he elects some unto eternal life and some unto eternal reprobation. God does reveal to us that he really and genuinely interacts with history, and that there are real transitions of covenant status among men. Now, how exactly those two truths relate is a mystery. But each (God’s eternal decree and real transitions in history) are truths God gives to limit our thinking from going to one extreme or the other.
Unfortunately Hoeksema did not have his thinking limited by the truth of real transitions in history, and therefore fell into a form of rationalism by prejudicing God’s decree at the expense of history. Arminianism also falls into rationalism, by prejudicing history at the expense of God’s sovereign decree. Both have tried to pierce into mystery, and therefore they have surrendered one limiting concept for the sake of the other.
And this is why Bavinck is absolutely correct when he says that mystery is the lifeblood of all true theology. Forsake mystery—and its correlate, “limiting concepts”—and rationalism is the inevitable result.
 See Herman Hoeksema, The Heidelberg Catechism (Grand Rapids: Eerdmans, 1951), 6:148; cf. Cornelius Van Til, Common Grace and the Gospel, 241.