I'll comment on this answer by Craig:
At face value, it seems incredible to think that Christ died only for the elect. You couldn’t get a much clearer repudiation of this view than I John 2.2: “he is the expiation for our sins, and not for ours only but also for the sins of the whole world.” Reformed thinkers are forced into exegetical acrobatics in order to explain away the prima facie meaning of such scriptural statements.
So what in the world would compel someone to re-interpret such passages in order to make them compatible with the view that Christ died only for the sins of the elect and not for the sins of every human being? The reason is a theological inference that forces one into such contrived exegesis. One is forced into this position by a theological argument that implies the limited extent of the atonement.
i) Evidently, Craig takes "the world"–or perhaps "the whole world"–to mean "every human being". Well, let's compare that to another statement in 1 John which uses the same compound phrase:
We know that we are from God, and the whole world lies in the power of the evil one (1 Jn 5:19).
If "the whole world" means "every human being", then it must mean the same thing in both verses. Yet it can't mean that in 5:19, for in 5:19, the second clause stands in contrast to the first clause. The first clause refers to Christians, in apposition and opposition to "the whole world". John doesn't think every human being is in the Devil's thrall, for he exempts Christians. Therefore, the scope of "the whole world" must be narrower than "every human being". Craig needs to be consistent. As one commentator observes:
John here returns to the duality between the world and God's children that is so characteristic of his thinking. The inclusive "we" refers to those who have been born of God and therefore are no longer of the world, a world that lies under the power of the evil one, the devil. The reason the devil cannot "touch" or take hold of one of God's children is that they are no longer within the realm of his power. K. Jobes, 1, 2, & 3 John (Zondervan, 2014), 238.
ii) Craig's response reflects the power of subliminal conditioning. John doesn't actually say the "world". John didn't write in English. Rather, that's simply a traditional English translation of kosmos. When translators render a text from one language to another, they attempt to find synonyms in the receptor language that correspond to the donor language. Sometimes that's straightforward. However, words in the donor language may have a different semantic range than words in the receptor language. So translators must sometimes settle for words with an overlapping semantic range. Even so, a word in the donor language may have different connotations than a word in the receptor language.
In addition, a writer may have his own idiolect. As standard works in Greek lexicography document, John often uses kosmos with a pejorative connotation.
In fact, Reformed thinkers themselves recognize this truth in distinguishing between redemption as accomplished and as applied. They will say that our redemption was accomplished at the cross but that it is applied individually when persons are regenerated and place their faith in Christ. This distinction is vital because otherwise the elect would be born redeemed! They would never be unregenerate sinners but would be justified and saved from the instant of their conception. But Scripture teaches that we once were “children of wrath like the rest of mankind” (Ephesians 2.3), and many of us recall our pre-Christian days. But how can such a distinction make sense if Christ won our actual redemption at the cross? If I was actually redeemed in AD 30 (never mind that I didn’t exist then!), how can I not be redeemed at every moment that I do exist? The undeniable distinction between redemption accomplished and applied makes sense only if we say that Christ’s death wins our potential redemption and that that potential is actualized in individual lives through repentance and faith.
i) In Reformed theology, the elect are, indeed, born redeemed. But although they are actually redeemed from the moment of conception–the full benefits of redemption don't accrue all at once. For instance, they aren't glorified at the moment of conception. There are stages in the application of the atonement. Craig's argument is fallacious.
ii) In addition, Craig doesn't bother to explain how faith actualizes a potential redemption.
I don’t see any problem of “double jeopardy” here. That is a convention of our human criminal justice system in the United States which cannot be automatically applied to God’s dealings with humanity.
It's as if Craig is ignorant concerning the history of the argument. It goes back to John Owen, a 17C English Puritan. So, no, I don't think Owen was influenced by the American jurisprudential system!
In any case, it is not as if the unrepentant person is being tried twice for the same crime. There is only one Judgement Day, and that is the only time a person is tried. If he has freely rejected the pardon Christ offers him, there is no one else to pay for his crimes.
That doesn't begin to engage the argument. Again, it's as if Craig is dependent on a truncated, secondhand version of the argument. The dilemma is how God can justly punish impenitence or unbelief if Christ made atonement for every sinner, or every sin, or the guilt of every sin. In that event, what are the just grounds for condemning an unbeliever? Is unbelief culpable? Is impenitence culpable? But if Jesus paid the price for your sin, then unbelief and impenitence are covered.
Isn’t the view I suggest biblical? The Old Testament sacrifices availed for nothing unless they were conjoined with a contrite and repentant heart on the part of the person for whom they were offered.
That's terribly confused. The OT sacrifices were merely emblematic placeholders. They didn't lay the basis for a sinner's forgiveness. Animal sacrifice didn't really contribute anything to a sinner's forgiveness. They never had a latent power to remit the guilt of sin in combination with a contrite heart.
But suppose you do think that Christ dies only for the elect. Does that imply that “most people couldn't even possibly be saved”? I don’t think so. There are two ways in which salvation could be universally accessible. First, if we take election to be primarily corporate, then it is up to us whether we want to be part of that corporate body which is the object of Christ’s redemption. Christ died only for the elect, but anyone can be part of the elect by repentant faith.
If election is after the fact, what difference does it make to the outcome?
Or, second, we could adopt a middle knowledge perspective, holding that God knew who would freely receive God’s grace and be saved, and so He sent Christ to die for them alone but not for those persons who He knew would freely reject Him.
That depends, in part, on how you define grace. What if grace is, in part, like a psychotropic drug that restores sanity to a mental patient? The patient is in no condition to accept or reject it. Unless and until a person is in a right state of mind, he lacks the mental competence to rationally consider a proposal.
If someone who remains unrepentant were to place his faith in Christ, then God would have included him in Christ’s atoning death.
That counterfactual scenario is true in Calvinism.
Once again, we see the astonishing power of the doctrine of middle knowledge to open up unexpected options theologically.
There's nothing astonishing about the ability to toy with hypothetical scenarios.