John MacArthur holds to Calvinism’s five points, including limited atonement (the “L” in TULIP). He departs from his generally literal hermeneutic in handling 1 John 2:2, arguing that ὅλου τοῦ κόσμου does not refer to the whole world “Jesus didn’t pay for the sins of Judas…or Adolf Hitler.” He actually said that. MacArthur explains that the verse is simply explaining that atonement was now available to the whole world, but that it does not mean that Jesus paid for the sins of the whole world. Appealing to John 11:52, MacArthur asserts Jesus only died on behalf of the children of God. But 11:51 describes that Jesus would die for the nation [of Israel.] Still, 11:51-52 makes no claim that Jesus would die for the children of God, but only that He died in order to gather them together. MacArthur’s presupposed limited atonement drives his (non) exegesis of 1 Jn 2:2. When John says “the whole world” he really means “the whole world…except for anyone in the whole world who would not believe in Him.”
John Piper agrees with MacArthur, and defends his own conclusion almost identically: “When Christ died on the cross paying the price for us…He decisively accomplished that for His own. His sheep. His elect…He didn’t just make it accomplishable, He accomplished it.” In other words, when Jesus declared that it was finished, He had completed the entire redemptive process for believers. For this reason, Piper prefers the term “triumphantly effective atonement” instead of limited atonement. One big problem, brother Piper: this goes a far cry further than regeneration preceding faith – it annuls the need for faith, if salvation for the elect was already accomplished the moment Christ died. Referring to John 3:16, Piper admits that “the cross is universal in that conditional sense,” that anyone believing will have life.
Jesus “decisively purchased with a dowry, His bride,” Piper says. In handling 1 John 2:2, Piper identifies Jesus as the “wrath remover” for the whole world, and like MacArthur, likens the passage to John 11:51-52 – a passage which again, does not limit His death to the children of God. Piper adds, “I think that is what he means by propitiation for the whole world, namely, as the gospel spreads around the whole world, the whole world becomes the object of His saving work in that He gathers children of God from out of every tribe and tongue.” But John didn’t say in 1 John 2:2 that Jesus died for people from the whole world, but that He was the ἱλασμός, the propitiation περὶ ὅλου τοῦ κόσμου – for or on behalf of the whole world.
Both Piper and MacArthur are playing a semantic game here, and neither can deal with the exact phrasing of 1 John 2:2. Both refer to a distant context that is dissimilar in wording (though Piper says “notice the almost exact wording here” referencing John 11:51-52, and I have no idea what he could be referencing), and both redefine the whole world to mean not the whole world.
The very simple problem they are trying to avoid – that Jesus’s propitiation must be a completed purchase, and thus must be fully efficacious – is a theological/philosophical problem, not an exegetical one. They are trying to resolve a philosophical conundrum that isn’t there by explaining away the passage that is there. In fact, there is no problem at all – they are assuming too much of the word ἱλασμός, without any exegetical warrant to do so. Romans 3:25-26 distinguishes between propitiation (ἱλαστήριον) and justification (δικαιοῦντα). There is no exegetical need to conclude that justification of the elect occurred at the cross. It didn’t. The justification is through faith, and has always been (Gen 15: 6; Rom 3:26).
Also, this argument favoring limited atonement takes, for example, the sheep/shepherd metaphor too far. Shepherds don’t create their sheep (Jn 1), nor do shepherds hold their sheep together in every way (Col 1), nor are shepherds priests on behalf of their sheep (Heb 4), nor do shepherds choose their sheep before the foundation of the world (Eph 1). The metaphor in John 10 simply serves to illustrate how the sheep enter the fold (10:1-6) – through the shepherd. The figure of speech doesn’t go much further than that. A sheep is one who enters through the shepherd (10:2), and His disbelieving audiences were not sheep because they didn’t come to the Father through Him (10:26). In other words, He puts it on their shoulders that they are not sheep – it is their fault. Jesus is not referring to election here at all.
Finally, let’s consider the formal quality of the limited atonement assertion. Piper’s and MacArthur’s appeals to John 10:11 (sheep) or 11:52 (children of God) is a valid (in form) modus ponens argument:
P1: If Jesus died for the sheep (or the children of God) then He didn’t die for those who weren’t sheep (or the children of God).
P2: Jesus died for the sheep (or the children of God).
C: Jesus didn’t die for those who weren’t sheep (or the children of God).
In a deductive argument, if the two premises are true, then the conclusion will necessarily follow. The problem is simply that P1 is not exegetically defensible (there is no passage which supports such a thing, and 1 John 2:2 seems to strongly assert the opposite view). If there is no exegetical evidence for the assertion, then it is not demonstrably true. In order to solve a nonexistent problem, these two intelligent men appeal to an exegetical assumption and state it as fact. This maneuver illustrates how tempting it can be to exegetically cheat when we are driven by theological presuppositions and traditions rather than simply allowing the Text to speak on its own terms.