Warning: session_start(): open(/opt/alt/php56/var/lib/php/session/sess_r0iuoac8cp6bdd098j7chkuha1, O_RDWR) failed: Disk quota exceeded (122) in /home/q85ho9gucyka/public_html/wp-content/plugins/counterize/counterize.php on line 16
Calvin and Calvinism » Blog Archive » Romans 8:32 and the Argument for Limited Atonement (Revisited)


Part 1. Introduction

In Romans 8:32, Paul says,

“He who did not spare his own Son but gave him up for us all, will he not also give us all things with him?”

Paul uses a basic a fortiori argument to demonstrate that if the Father delivered up Christ for us, how much more, then, will he not give us all things. If he did that much for us, how then will he not also do the lesser thing of bringing us to final salvation?

The argument for limited satisfaction:

Paul bases the certainty of our inheritance on the death of Christ. He says, “God will most certainly give you all things because he did not spare his own Son but gave him up for you.” If Christ is given for those who do not in fact receive “all things” but who are, instead, finally unsaved, Paul’s argument is voided. If God gave his own Son for unbelievers who in the end are lost, then he cannot say that the giving of the Son guarantees “all things” for the those for whom he died. But this is exactly what he does say. If God gave his Son for you, then he most certainly will give you all things. The structure of Paul’s thought here is simply destroyed by introducing the idea that Christ died for all men in the same way.[1]

Here is the simplest way I can think of to respond to this argument in order to show why its invalid.

1) The hasty term generalization: “us” (from the text) is converted into a general term, “all,” which is “all” irrespective of faith, when all along, in Romans 8, the “us” explicitly presupposes believers.

E.g., at its simplest:

We for whom Christ died will be given all things


All for whom Christ died will be given all things

The first sentence is true to the text, the second is not. It is a non sequitur.

2) If the term conversion is invalid, then the subsequent modus ponens and modus tollens arguments based upon it are invalid.

E.g., at its simplest:

Modus ponens: If A, therefore B.

Modus tollens: Not B, therefore not A.

That is, on the assertion that “All for whom Christ died will be given all things,” the following arguments are normally constructed:

a) If Christ died for a man, that man must be saved [If A, then B]
b) If that man is not saved, Christ did not die for that man [not B, therefore not A]

From which the general conclusion becomes: “Christ has not died for any man not (actually) saved

So the three flaws in this argument, are:

1) The invalid term conversion
2) The unsound claim that if Christ dies for a man, that man cannot fail to be saved.
3) The invalid inference to a universal category negation (discussed below).

The first assumption, being invalid on its face shows that the conclusion to limited satisfaction is flawed as it engages in unjustified term conversation (equivocation). The second false assumption needs to be proved on grounds other than Romans 8:32 etc., as the text, itself, does not imply it. At this point it just begs the question, formally.

Part 2. Rebuttal in Detail

The is argument is, if it is possible that Christ could die for a person and that person fail to be saved, Paul could not assure believers that if Christ died for them, it cannot fail that they will be given all things, namely final and complete salvation.

Paul’s point is being missed here. Paul’s assurance is to the “we” who have been given Christ, should not doubt, but be very assured that “we” will be given all things. This is exactly the same sentiment behind or entailed in 8:28, 31, and, 34, for example. The faithful can and do have this assurance.

At most all that can be inferred is perseverance of the saints or the effectual intention to apply the death of Christ to believers bringing about their complete salvation, and not limited satisfaction for the sins of the elect alone. The logic of the text will just not give one grounds for inferring a limited satisfaction. The textual data does not sustain any generalization outside of its own stated limitation (namely the “us” of 32b). Here now is the explanation for my counter.

Paul also uses the same form of the a fortiori argument in Romans 5:8-10:

But God demonstrates His own love toward us, in that while we were yet sinners, Christ died for us. Much more then, having now been justified by His blood, we shall be saved from the wrath of God through Him. For if while we were enemies, we were reconciled to God through the death of His Son, much more, having been reconciled, we shall be saved by His life.

The same structure is present. There is a bi-conditional argument here. For example: having been died-for, and having now been justified, much more then shall we be saved from the coming wrath.

Verse 10, Paul again sets out two conditions, “while we were died for, as enemies, AND now that we have been reconciled . . .” both need to be present for the conclusion to hold: “much more then shall be saved from the coming wrath!”

Paul is saying, because Christ died for us, AND because we have now been justified and reconciled, much more then, can we be assured of complete salvation. The result is the same, perseverance yes–limited satisfaction for limited sins, no.

There are no grounds to infer a limited satisfaction in Romans 5:8-10, any more than there are in Rom 8:32.

Looked at from another angle, this point becomes even more apparent. If we go back to Romans 8:32. Paul’s actual structure is very interesting.

Romans 8:32:

He who did not spare His own Son, but delivered Him over for us all, how will He not also with Him freely give us all things?

Given the structure of Rom 8:32 we have, an “us all” and an “us,” so there are only so many possible relevant interpretations here. For explanation, in the following, let “all mankind” refer to all who have lived, live, and shall live. And “all the elect” refer to all elect who have lived, live, and shall live, irrespective of whether any given elect person is a believer.

1) us all refers to all the elect, and us refers to all the elect.

2) us all refers all mankind and us refers all mankind

3) us all refers to all believers, us refers to all believers

4) us all refers to all the elect, us refers to believing elect

5) us all refers to all mankind, us refers to believers.

6) us all refers to all mankind, us refers to all the elect

7) us all refers to all mankind, us refers to all believing elect

Only 2) is unacceptable, as all mankind will not be given all things.

5) and 7) are in practical terms the same, but I will include them separately so the point is clear.

Exegetically, I can say 1) and 6) are not exegetically probable.

4) may be possible, but only given one’s wider systematic assumptions.

3) is possible.

7) is probable.

Either way, it’s a “in the eye of the beholder” sort of thing. My take is probably 5) aka 7).

That means 1, 3-7 are all logically and biblically acceptable (broadly speaking).

Whichever one of the acceptable outcomes we plug in, universalism is not the outcome, as the conclusion is always regulated by the second “us.”

Like a simple bi-conditional statement: If A and B, then C. C only holds if both conditions are present.

Paul’s structure is actually not a simple if A, then C (as is often suggested in popular literature on limited satisfaction).

Ironically, however, all this analysis is rather unnecessary.

If, us all, stands for all mankind, and us, stands for believers or the elect or the believing elect, the argument for limited satisfaction (even by way of the reductio entailment of universalism) is invalid, because in each case, the us regulates the conclusion and it delimits it.

Defining the us all, is actually irrelevant to defeating the argument for limited satisfaction based on Romans 8:32.

To recap the argument.

The (invalid) term conversion must convert the us all and/or the us, into “all for whom Christ died” (irrespective of election or belief). However, there is no textual warrant to do so, which is exactly what is being done here, as it is alleged that: “All for whom Christ died will infallibly be given complete salvation.” What is happening is the attempt to force the invalid term conversion upon others to support the claim that text data necessitates this premise:

All for whom Christ died will be given all things

All that needs to be pointed out is that the term conversion is, itself, invalid. End of discussion.

We, for our part, do not need to establish who exactly who the us all are. That can remain a completely open question. In short, whoever the us is, the conclusion of any modus ponens argument is defined by it.

For example,

Let X stand for “us” (8:32b) according to any of the acceptable permutations above.

All that could be constructed is this:

All X for whom Christ died will be given all things.

That is:

All the elect for whom Christ died will be given all things
All the believers for whom Christ died will be given all things
All the believing elect for whom Christ died will be given all things

In each case, universalism is not entailed.

And this also leads us to the very important fact that one cannot infer category exclusion from a simple positive statement. That is, nothing is being said about any non-X. Paul is not saying that Christ did not die for any non-X at all. Such and inference would be invalid on its head. He is saying, we (i.e., X) for whom Christ has died, we will be given all things. Nothing is entailed or implied, negatively or positively, about any other class, such as the class of unbeliever, or the class non-elect, for whom Christ did or did not die for.

The alleged conclusion that: “Since it is assumed that both sides of the argument for limited versus unlimited satisfaction deny universalism, this text therefore teaches limited atonement” is just false, and categorically so.

The possibility that Christ could have died for the non-elect, or non-believers, or non-believing elect cannot be categorically excluded. Such a category exclusion formally begs the question (petitio principii).

Part 3. Follow-up

The approach here has to be exegetical first and foremost. The question is, who are the us in 8:32b?

The are two principal candidates,

1) Elect as a class, including elect who are not yet converted, that is, all the elect who have lived, live and shall live.

2) Believers. Even if the believers are believers who are elect, it is still believers.

The question is, then, which one best fits the context. I would say that the contexts fits better with 2) because the whole chapter references believers. That is, the predications in the chapter can only apply to believers.

8:26 In the same way the Spirit also helps our weakness; for we do not know how to pray as we should, but the Spirit Himself intercedes for us with groanings too deep for words; 27 and He who searches the hearts knows what the mind of the Spirit is, because He intercedes for the saints according to the will of God.

Clearly these verses reference believers. Does the Spirit intercede for unbelievers even unbelieving elect? I think not.

8:28 And we know that God causes all things to work together for good to those who love God, to those who are called according to His purpose. 29 For those whom He foreknew, He also predestined to become conformed to the image of His Son, so that He would be the firstborn among many brethren; 30 and these whom He predestined, He also called; and these whom He called, He also justified; and these whom He justified, He also glorified.

Again the called are believers, as those who love God are believers. These are believers who have been elected, foreknown and predestined. These are not the class “elect” simply considered, apart from belief. Paul speaks of the “called” in this manner: the elect are considered as they are “called” elect (elect qua their calling), not simply elect as they are elect (elect qua elect).

8:31 What then shall we say to these things? If God is for us, who is against us? 32 He who did not spare His own Son, but delivered Him over for us all, how will He not also with Him freely give us all things?

God will give us all things, namely give us who believe. Exegetically, the context and text clearly indicates believers are the subject of the discourse. That being so, the modus ponens conclusion has to be limited by the scope of the premise.

So now we come back to my earlier taxonomy:

All the believing elect for whom Christ died will be given all things.

Or stated slightly differently again for emphasis: To all X for whom Christ died, all things will be given. X is now defined as believers who are elect.

That is the most one can get out of this verse.

Part 4. Enthymematic Premises in Everyday Language

A preacher says to his Church, “God loves us, and will certainly save us.” He can say that because there is an unstated premise, called in logic an enthymematic premise, which in this case is the supposition that he and his audience are “believers,” or the unstated conditional assumption, “if we believe.” On the supposition that we believe, it is absolutely true that as God loves us, and as Christ has died for us, he will certainly save us. That, in essence, is what is going on in Paul discourse in Romans 8.

The preacher says: “If the Father delivered up his own Son for us, how much more will he not give us all things?”

Again, the enthymematic premise is assumed exactly because such a statement could not be true when expressed before the local atheists association. Why is that?

The same statement said in different contexts can be true in one but become false in the other.

To the Church:

If the Father delivered up us his own Son for us, how much more will he not give us all things?

Or more simply:

If Christ died for us, he will give us all things.

Both would be true.

To a crowd of Atheists:

If the Father delivered up us his own Son for us, how much more will he not give us all things?

Or more simply:

If Christ died for us, he will give us all things.

Both would be false. Why? Because of the unstated assumption or condition is missing in the second situation; that is, the second bi-conditional is absent. The conditional or supposition “if” we believe” is absent in the lives and personal commitment of the atheists present. And without personal faith in Christ, no man can be saved. Salvation, and “all things” are not given apart from or irrespective to faith in Christ.

Furthermore, in the example of the preacher speaking to his congregation, no one would imagine the preacher was attempting to imply a category exclusion or negation, namely, Christ died only for the preacher and his singular congregation. Or that he did not die for the Baptists down the road. All his hearers would naturally not make such a conclusion or imagine that it was ever intended.

Now if we apply this to Paul’s argument in Romans 8:32 the problem should now be clearer, as Paul is not saying “all for whom Christ died, irrespective of faith, will infallibly be saved.” Like the preacher before his congregation, he says “We for whom Christ has died, we will be given all things, and that infallibly so.” Nor is it the case that Paul is attempting to communicate a category exclusion or negation. Paul’s readers would never had made such an inference. The only natural inference would have been to apply Paul’s argument to all other believers by the principle of “all things being equal.”[2]

And just to be sure, my point is, Paul’s predications and assertions speak to his readers who are believers “who have been elected,” not to the “elect” as a total class irrespective of faith and repentance, which is what most strict TULIP advocates tend do with this text in my experience.

Thus, in conversational language, one can assume a condition, which is not stated, but which the context of the proposition (to whom it is addressed) sustains the true assumption of an enthymematic premise.


For all the above reasons, Romans 8:32 cannot be used to prove a limited satisfaction for sins on the basis of either the original Pauline a fortiori argument or on the basis of any extra textually constructed modus ponenss or modus tollens arguments. Paul here is actually silent on the “extent” question. All he can be taken to speak to is the intent to effectually apply the benefits of Christ’s death to the believer or believing elect.

[1] John Piper, For Whom Did Christ Die?

[2] A predication made to one believer holds good for another and so on indefinitely for all believers based on the principle of representation, for this reason the present day believers can read their Bible and predicate its promises and blessings to themselves.

This entry was posted on Thursday, October 25th, 2012 at 7:02 pm and is filed under Short Essays, Notes, and Comments. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

One Trackback/Ping

  1. Traduções Crédulas: Romanos 8:32 e o Argumento pela Satisfação Limitada | credulo    May 18 2013 / 12pm:

Leave a reply

Name (*)
Mail (will not be published) (*)