Negative Inference Fallacies of Acts 2:38, Matthew 19:9, and 1 Corinthians 11:5

by Paul S. Dixon

  1. The Negative Inference Fallacy of Acts 2:38 (and Mark 16:16)
  2. The Negative Inference Fallacy of Matthew 19:9
  3. The Negative Inference Fallacy of 1 Corinthians 11:5
  4. Conclusion

If logic is "the set of relationships (nicely formulated by Aristotle and others) that must apply if any knowledge is possible and if any communication of propositional knowledge is possible," (1) and if such logic is universally applicable, then it is imperative that its rules be strictly observed, especially in the interpretation of Scripture. D.A. Carson warns the exegete against eighteen logical fallacies. (2) One of these is the fallacy of the negative inference.

Simply put, the negative inference fallacy says if a proposition is true, it does not follow that a negative inference from that proposition is also true. It may or may not be true, but if it is true, it is not so by inference from the original proposition. In conditional format, (3) "If A, then B," does not imply the negation, "If not A, then not B." For example, "if a man is a resident of Oregon, then he is a resident of the United States," does not imply "if a man is not a resident of Oregon, then he is not a resident of the United States."

Most interpreters do not have difficulty with the simple conditional. Inferring "if not A, then not B" from "if A, then B" would be too blatant an error. However, when multiple conditions exist (If A and B, then C), then the situation becomes somewhat more treacherous, and the tendency to infer the negation (If not both A and B, then not C) increases significantly. (4) We will demonstrate this to be the case in interpretations of Acts 2:38, Matthew 19:9, and 1 Corinthians 11:5.

I. The Negative Inference Fallacy of Acts 2:38 (and Mark 16:16)

Those who insist that baptism is required for salvation rely heavily upon passages such as Acts 2:38 and Mark 16:16. (5) In doing so, however, they fall into the negative inference trap.

The conditional thought in Mark 16:16 is simply, "If a man believes and is baptized, then he will be saved." The negation, often erroneously deduced, runs something like this, "If a man does not both believe and receive baptism, then he can not be saved." While the second half of the verse does affirm the negation for belief, "he who believes not is condemned already," it does not affirm the negation for baptism. It does not say, "he who is not baptized will not be saved." To infer such is to infer the negation.

Likewise, in answer to the question, "what shall we do?" (Acts 2:37) Peter responds, "repent and let each of you be baptized in the name of Jesus Christ for the forgiveness of your sins." The conditional thought is, "if you repent and receive baptism, then your sins will be forgiven." This does not imply, however, that if they did not both repent and receive baptism, then their sins would not be forgiven. This is the erroneous conclusion of both F.F. Bruce and John Stott. Bruce asserts that "As John's converts were required to receive baptism in water, so Peter's convicted hearers were now required to submit to it." (6) According to Stott:

"Peter replied they must repent ... and be baptized in his name, submitting to the humiliation of baptism, which Jews regarded as necessary for Gentile converts only, and submitting to it in the name of the very person they had previously rejected." (7)

Bruce's use of "required" and Stott's use of "must" suggest that both repentance and baptism are necessary for the forgiveness of sins. This is the error. Peter's words do not make both repentance and baptism requirements for forgiveness. He does not say "must," or "required." Neither does he say that if these Jews do not both repent and receive baptism, then their sins cannot be forgiven. To affirm this is to infer the negation.

There is a difference between conditions for salvation and requirements for salvation. It is the difference between the conditional and its negation. Plenty of verses declare the simple conditional truth, "If a man believes, then he will be saved" (e.g. John 3:16; Acts 16:31). Belief, then, is a condition for salvation. But belief is also a requirement, for numerous scriptures affirm the negation that if man does not believe, then he will not be saved (John 3:18, 36, 6:53-54, 8:24; Mark 16:16b). However, no such negation is to be found for baptism. (8) It cannot, then, be demonstrated that baptism is a requirement for salvation.

If a man is saved by faith alone, then we can add any number of conditions to the protasis and he will still be saved. If a man believes and is baptized, then he will be saved. If a man believes, is baptized, goes to church, takes communion and gives a tenth of his income to the Lord, then he will be saved. We err, however, if we conclude he has to do all these things in order to be saved.

Why, then, do both Acts 2:38 and Mark 16:16 link baptism so closely with belief and repentance as conditions, if not requirements, for salvation? It is probably to stress the importance of baptism as a command to be obeyed. It is so important, in fact, that if one were to claim he had faith, yet refused to submit to baptism, then there would be good reason to doubt his salvation. Commenting on Mark 16:16, Barnes adds:

"It is worthy of remark that Jesus had made baptism of so much importance. He did not say, indeed, that a man could not be saved without baptism, but he has strongly implied that where this is neglected knowing it to be a command of the Savior, it endangers the salvation of the soul. Faith and baptism are the beginnings of a Christian life: the one the beginning of piety in the soul, the other of its manifestation before men, or of a profession of religion. Every man endangers his eternal interest by being ashamed of Christ before men." (9)

Scripture never says that if a man is not baptized, then he cannot be saved. Furthermore, we cannot rightfully deduce it. Yet water baptism is vitally important, not as a means to salvation, but as an expression of the believer's love and devotion to the Lord. (10) Let us next turn our attention to a similar construction in Matthew 19:9.

II. The Negative Inference Fallacy of Matthew 19:9

Suggested interpretations of Matthew 19:9 and the exception clause abound. Almost all interpretations, however, are guilty of the negative inference fallacy. The conditional thought of the verse is, "if a man divorces his wife for any reason except immorality and remarries, then he commits adultery." The negative usually assumed is, "if a man divorces his wife because of immorality and remarries, then he does not commit adultery."

Even Murray, who takes great pains to argue that the exception clause modifies both divorce and remarriage, then assumes the negative by assuming the clause also modifies adultery. He writes:

"The question then is: does this exception, by way of right or liberty, extend to the remarriage of the divorcing husband as well as to the putting away? Obviously, if the right extends to the remarriage, the husband in such a case is not implicated in the sin of adultery in the event of his remarriage." (11)

Obviously? How does negating the protasis necessitate negating the apodosis? (12) Murray has fallen into the negative inference trap. In order to get the negation, Murray must show 'if not P, then not Q.' All he has done is affirmed 'not P.' He has failed to show 'not Q' follows from that. To assume it is to erroneously infer the negation.

The basic rationale for assuming the negation seems to involve the following question: why would Christ bring up this exception or additional condition if He did not mean to imply the negation? This, of course, is an argument from silence. Just because we cannot imagine another explanation does not mean this must be the right one, especially if it causes us to make invalid inferences.

Further attempts to justify the inference of the negation have focused on linguistics, but only as a cover for the real reason (the assumption of the negation). The English translations almost always render the Greek mn in Matthew 19:9 as "except," in spite of the fact that it is never translated like that anywhere else in the New Testament in similar constructions. (13) Some, apparently aware of this problem, have suggested an ellipsis of the Greek particle ei or ean in order to get the desired "except." (14) But even if an ellipsis is understood, it still does not deal with the problem, for that would give us only the first half of the desired negation. The silent second half of the negation still screams for assertion. We cannot simply infer it.

The translation "except" is not only lexically without merit, but it is especially unfortunate if it conjures up the negation to the English reader. But that is precisely why the translators render it such, and it gives away their assumption of the negation. Porter and Buchanan, however, have shown that even the English "except" does not necessarily imply the negation. As an example, they say: "All centers, except those over 6 feet tall, will fail in the NBA." (15) Clearly, this is saying nothing about the success or failure of centers over 6 feet tall. That consideration is excluded from discussion. The point, rather, is to assert something about centers under 6 feet tall.

There is no good reason why mn in Matthew 19:9 (mn epi porneia, "except for immorality") should not be translated by its normal "not." Literally, the translation would be something like, "not for immorality," or "setting aside the matter of porneia," (16) the idea being to exclude porneia or immorality from consideration at this point, but certainly not to imply its negation.

Of the many suggested interpretations of the so-called "exception" clause, only one does not erroneously assume the negation. Suggested by Augustine over 1500 years ago, (17) and essentially the view of the early church, (18) the preteritive view sees the clause as merely excluding from discussion the case of the wife who commits fornication. It has also been referred to as the "no comment" view. (19) The situation regarding the wife who commits fornication is simply not being considered, and no inference may be drawn regarding such. According to this verse alone, the man who divorces his wife for fornication and remarries may or may not be committing adultery.

But why would the Lord exclude this case from consideration in verse 9? In part, because he has just discussed it in the immediately preceding verses. If the exclusion clause does not refer back to those verses, then it remains syntactically unrelated to anything in the text. If it does refer back, however, then the clause would serve to sharply separate the two discussions, making them mutually exclusive. Accordingly, if one wishes to justify divorce and remarriage on the basis of fornication, then he must do so from verses 4-8 where justification for divorce is the hardness of heart of the apparently unforgiving husband. But even such justification is offset by the clear teaching of Christ that God's intent for marriage from the time of creation has always been and remains permanency.

A second reason for the exclusion clause is the hardheartedness of those questioning Jesus. As was His custom, Jesus limited His revelation accordingly. Full disclosure would have meant nothing to them. It would be enough for them to be told the that God's purpose of permanence in marriage remains intact (vss. 4-7, 8b), that the divorce of an immoral wife was due to a hardness of heart (v. 8a), and that the divorce of a moral wife and remarriage constitutes adultery (v. 9). Their hardness of heart precluded their being able to handle the greater and more difficult statement as recorded in Mark and Luke. It was not until later and in private to His disciples only that the Lord more fully revealed the matter so as to include all divorce and remarriage. Fleming comments:

"... such discussion would have been worse than useless with the "hardhearted" Pharisees. As a good pedagogue, He is content in the circumstances with a partial revelation of the truth, a principle on which He also acted in the gradual revelation of His divinity ... he thus answers the query originally proposed to Him ... His reply condemns the lax doctrine of Hillel and is more strict than the teaching of Shammai. The legal-minded Pharisees see that He has answered their question ... and they do not press the point any further. The disciples, however, are not satisfied with the partial answer given to the Pharisees, so they seek fuller information." (20)

Just as we cannot deduce that a man who believes but is not baptized is not saved, so we cannot deduce that a man who divorces his wife because of fornication and remarries does not commit adultery.

III. The Negative Inference Fallacy of 1 Corinthians 11:5

At first glance it might appear that Paul's total prohibition against women speaking in the churches (1 Corinthians 14:34-35) is contradicted by his previous teaching regarding the shame women incur by praying and prophesying with their heads uncovered in the church (1 Corinthians 11:5). Most interpretations, however, resolve the tension by interpreting Paul's rather severe instruction in 1 Corinthians 14 in light of the less severe instruction in 1 Corinthians 11:5 where it appears "Paul is quite prepared for women to pray and prophesy, albeit with certain restrictions." (21)

When they do so, however, they fall into the negative inference trap. The conditional thought of 1 Corinthians 11:5 is simply this: if a woman prays or prophesies with her head uncovered, then she acts shamefully. The negation, usually assumed, is: if a woman prays or prophesies with her head covered, then she does not act shamefully. The text, of course, does not say this, and it is not valid to infer it. If 1 Timothy 2:12 and 1 Corinthians 14:34-35 forbid a woman from teaching or holding authority over a man, and from even speaking in the assembly, then her praying or prophesying in the assembly would be a shameful thing in itself. Thus, the shame is not necessarily just in the doing so with her head uncovered. If a man drinks to excess and drives his car while doing so, then he acts shamefully. Does this imply if he drinks to excess and does not drive his car while doing so, then he does not act shamefully? Is he not acting shamefully by merely drinking to excess? Calvin argues to the point:

"But it seems unnecessary for Paul to forbid a woman to prophesy bare-headed, since in 1 Timothy 2:12 he debars women from speaking in the church altogether. Therefore they would have no right to prophesy, even with their heads covered, and the obvious conclusion is that it is a waste of time for Paul to be discussing the question of head- covering here. The answer can be given that when the apostle disapproves of the one thing here, he is not giving his approval to the other. For when he takes them to task because they were prophesying bare-headed, he is not giving them permission, however, to prophesy in any other way whatever, but rather is delaying the censure of that fault to another passage (chapter 14.34 ff.)." (22)

Some think the negation is implied because, as they reason, Paul most certainly would have said something otherwise. But Paul is under no obligation here to address the propriety of women praying and prophesying irrespective of the head covering. Not only does he address it later in general (1 Corinthians 14), but the form of the argument does not require him to do so here. In 1 Corinthians 11 Paul is assuming the praying and prophesying by the women for the sake of argument without passing judgment on it. This is typical Pauline argumentation. In 1 Corinthians 7:11, for example, Paul says that if a wife leaves her husband, then let her remain unmarried, or else be reconciled to her husband. We cannot infer from this that Paul approves of a wife leaving her husband, for he had just stated that a wife should not leave her husband (v. 10). In verse 11 He is merely assuming for the sake of argument that the wife has left. In 1 Corinthians 15 Paul assumes many things for the sake of argument, things which, in fact, he does not believe or endorse. He assumes that there is no resurrection of the dead (15:13), that Christ was not raised (15:14), and that we have only hoped in Christ in this life (15:19). Just as Paul felt no compulsion to re-teach the truth on these points in the fifteenth chapter, neither does he here in the eleventh chapter. He would, after all, address the matter in the fourteenth chapter.

In summary, there is no conflict between a total silence interpretation of 1 Corinthians 14 and 1 Corinthians 11:5. To assert otherwise is to affirm the negation of 1 Corinthians 11:5.

IV. Conclusion

The conditional does not imply its negation. If we are to be logically consistent and true to scripture, then we must resist the temptation to infer the negation, especially when multiple conditions exist and the temptation seems highest. When the Bible asserts the negation, then so should we. When the Bible does not affirm the negation, however, then neither should we.

We looked at three specific passages where interpreters have deduced the negation, but scripture has not. From Acts 2:38 we cannot deduce that if a man repents, but is not baptized, then he cannot be saved. From Matthew 19:9 we cannot deduce that if a man divorces his wife for fornication and remarries, then he does not commit adultery. From 1 Corinthians 11:5 we cannot deduce that if a woman prays and prophesies with her head covered, then she does not act shamefully. In each of these cases, the trap of the abused conditional would lead us to inferences unwarranted by the text.


  1. D.A. Carson, Exegetical Fallacies (Grand Rapids: Baker Book House, 1984), 93-94.

  2. Ibid., 94-126.

  3. From here on we will use the conditional format "if A, then B" instead of the syllogistic format "All A are B." There is no difference in meaning.

  4. Technically the negation of "if A and B, then C" is "if not both A and B, then not C." This may appear as "if A and not B, then not C," as "if B and not A, then not C," or as "if not A and not B, then not C."

  5. Das, for example, says that Acts 2:38, "has been a pivotal verse for the Lutheran, sacramental position." A. Andrew Das, "Acts 2:38: Water, Baptism, and the Spirit," Concordia Journal 19 [April 1993]: 108.) Beasley-Murray says, "the confession made in baptism receives the salvation of God." (G. R. Beasley-Murray, Baptism in the New Testament [St. Martin's Press; New York, 1962]: 103.) The belief that baptism was required for salvation goes back as far as Justin (Dial. 44.4), Irenaeus (Adv. haer 13.12.2; Dem. 41), Barnabas (Ep. 11.1) and Tertullian (De bapt. 4).

  6. F.F. Bruce, Commentary on the Book of Acts (Grand Rapids: Wm. B. Eerdmans, 1973), 76.

  7. John Stott, The Spirit, The Church and the World (Downers Grove, Illinois: InterVarsity Press, 1990), 78.

  8. John 3:5 does affirm the negation, "unless one is born of water and the Spirit, he cannot enter the kingdom of God." It is unlikely, however, that this refers to water baptism. Rather, the parallel with v. 3 and the background of Ezekiel 36:25-27 suggests that being "born of water" refers to the spiritual cleansing from sin at the time of regeneration. In 1 Pet 3:21 Peter says, "baptism now saves you," but he explains this as meaning not physical, but spiritual baptism, "not the removal of dirt from the flesh, but an appeal to God for a good conscience."

  9. Albert Barnes, Notes on the New Testament; Matthew and Mark (Grand Rapids: Baker Book House, 1972), 392.

  10. Not all have the opportunity to obey the Lord in baptism. The thief on the cross surely did not have the opportunity to be baptized, yet the Lord assured him of his salvation, "today, you shall be with Me in paradise."

  11. John Murray, Divorce (Philadelphia: The Presbyterian and Reformed Publishing Co., 1961), 35.

  12. The parts of the conditional, referred to as the protasis and apodosis, are the statements following the words "if" and "then" respectively. Sometimes the "then" is understood.

  13. In over 500 occurrences of mn by itself, it is translated "not". Only when it occurs with particles such as ei or ean is it translated "except". No such particles, however, are found in Matthew 19:9.

  14. B. Witherington, "Matthew 5.32 and 19.2 - Exception or Exceptional Situation?"," NTS 31 (1985): 576.

  15. Stanley Porter and Paul Buchanan, "On the Logical Structure of Matthew 19:9," JETS 34 (Sept., 1991): 335-339.

  16. Bruce Vawter, "The Divorce Clauses in Matthew 5:32 and 19:9," Catholic Biblical Quarterly 16 (1954): 164.

  17. De conjugiis adulterinis I; 9ff.: PL 40, 456ff.

  18. William Heth points out that the standard view of the church until the 16th century was that immorality of the wife (usually understood to be adultery) did not dissolve the marriage bonds, and so that even though separation was expected, remarriage was forbidden. In recent days, however, due largely to the writing of John Murray, Divorce (Philadelphia: OPC, 1953), the thinking of the church has been affected to such an extent, that the majority view today seems to be that fornication is just cause for divorce and remarriage, and that doing so does not result in adultery. William Heth, "The Changing Basis for Permitting Remarriage After Divorce for Adultery: the influence of R.H. Charles', Trinity Journal (1990): 143-159.

  19. Heth and Wenham, Jesus and Divorce (Toronto: Hodder and Stoughton, 1984), 181.

  20. Thomas V. Fleming, "Christ and Divorce," TS 24 (1963): 106-120.

  21. D.A. Carson, Recovering Biblical Manhood and Womanhood (Wheaton: Crossway Books, 1991), 140.

  22. John Calvin trans John W. Fraser, Calvin's Commentaries: The First Epistle of Paul The Apostle to the Corinthians (Grand Rapids: Wm. B. Eerdmans, 1960), 231.

Top of Page Churches of Christ Oneness Pentecostalism Toxic Doctrines