← All essays
New Testament Exegesis4,641 words23 min read

Let Your Yes Be Yes

Oaths, Evasion, and the Word of a Man in Matthew 5:33-37

Abstract

Jesus says to make no oath at all, and the church has never agreed on whether he meant it. This paper argues that the target is the graded system of oath formulas his contemporaries used to sound binding while keeping an escape open, rather than the solemn oath itself, and it rests the case on Matthew's own later chapter instead of on the grammar. It concedes at five points that the absolutist reading has arguments this position does not fully answer, and it ends where the passage does, on why a man needed the scaffolding in the first place.

Introduction

Jesus says “make no oath at all”, and then the church spent two thousand years deciding whether he meant it.1 Anabaptists went to their deaths with oath refusal among the charges against them. Quakers lost property and were barred from testifying in court. Meanwhile most of the church has taken judicial oaths, sworn ordination vows, and married people who said “I do” under God without believing any of it violated the Sermon on the Mount.

Both groups are reading the same four verses. Only one of them is reading them the way they were meant.

This paper argues that Jesus forbids the oath system his contemporaries had built, a graded scheme of formulas designed to sound binding while leaving an escape open, and that he does not forbid the solemn oath as such. It argues this on internal evidence, chiefly from Matthew's own later chapter on the subject, and it concedes at five points that the absolutist reading has arguments this position does not fully answer. It also declares its own interest: I swore an oath of enlistment, and a man defending a reading that happens to justify what he already did should say so out loud rather than let a reader discover it.2

Where the Passage Sits

Matthew 5:33-37 is the fourth of six units in which Jesus quotes something his hearers have been taught and then states his own demand against it. Murder, adultery, divorce, oaths, retaliation, love of enemies. The units are governed by a paragraph that comes before all of them and cannot be skipped: “Do not think that I came to abolish the Law or the Prophets; I did not come to abolish but to fulfill”, followed by the requirement that the disciple's righteousness surpass that of the scribes and Pharisees.3

That governing paragraph rules out one reading immediately. Whatever Jesus is doing in verses 33 to 37, he is not repealing Scripture, because he has just said he did not come to do that. A reading that has Jesus cancelling Deuteronomy is a reading at war with the paragraph four verses earlier.4

The literary neighbors matter too. Units three and four sit together and share a subject: the binding force of words. Unit three concerns a certificate of divorce, unit four concerns an oath formula, and both concern a legal instrument used to secure a self-serving outcome while staying inside the letter of the law. That pairing is not accidental, and it is an early signal of what Jesus is attacking. In both cases the machinery was working exactly as designed, and that was the problem.

What the Formula Introduces

Everything turns on what “you have heard” introduces, and there is a decisive piece of evidence two units later. In the sixth unit Jesus says, “You have heard that it was said, 'YOU SHALL LOVE YOUR NEIGHBOR and hate your enemy.'”5 The first half is Leviticus. The second half is not in the Old Testament anywhere. No text commands hatred of enemies, and the closest analogue in Second Temple literature is sectarian rather than scriptural.6

So the formula is not a citation formula. It introduces what the hearers have received, which may be Scripture, Scripture with an addition, or the standard interpretation of Scripture that has grown up around it. Once that is established, verse 33 stops being a problem and becomes evidence.

Verse 33 reads, “Again, you have heard that the ancients were told, 'YOU SHALL NOT MAKE FALSE VOWS, BUT SHALL FULFILL YOUR VOWS TO THE LORD.'” This is not a quotation of any single verse. It is an accurate summary of a body of legislation: the prohibition on swearing falsely by God's name, and the requirement that a vow be performed.7 That is a fair summary and Jesus does not dispute it.

What he disputes is what has grown on top of it. Note what the summary leaves out: the law regulated false swearing and unperformed vows. It said nothing about a man whose plain word could not be trusted. A system built strictly on those two rules can produce a perfectly law-abiding liar, provided he never puts the lie under oath.

Make No Oath At All

Verse 34 is the sentence the absolutist stands on, and it deserves to be stated at full strength before it is answered. The Greek reads mē omosai holōs. Matthew includes holōs, an adverb meaning wholly, altogether, or at all, which is not required by the sense and which is doing emphatic work. Under a negative its most natural force in Greek is precisely the absolutist's: not at all. NASB95 renders the phrase “make no oath at all”, and that is a fair translation. If Jesus meant “do not swear evasively”, holōs is a strange word to reach for.8

Calvin's answer is that the adverb attaches to the form rather than the substance, meaning neither directly nor indirectly, which is to say by no formula whatsoever.9 That reading is available, and it is worth being clear about what kind of claim it is. Calvin is not reporting a rule of Greek. He is making a contextual argument in grammatical clothes, and it stands or falls with the context rather than with the adverb.

So the honest position is this. The grammar permits the qualified reading and does not establish it. Context has to do that work, and the argument that follows is a contextual argument, not a lexical one. It is also worth naming a temptation and refusing it: the tense of the infinitive is aorist, and some have tried to extract a punctiliar sense from that, as though Jesus forbade only individual acts of swearing rather than the practice. That is precisely the kind of aspectual over-reading that a competent reader will strike, and nothing here rests on it.10

The Four Formulas

Verses 34 to 36 do not prohibit swearing in the abstract. They name four things and prohibit swearing by each: heaven, earth, Jerusalem, and one's own head. Each gets a reason attached, and the reasons are the argument.

Heaven is refused because it is “the throne of God”. Earth is refused because it is “the footstool of His feet”. Jerusalem is refused because it is “THE CITY OF THE GREAT KING”.11 In every case the logic is identical: you thought this formula avoided God, and it does not, because the thing you swore by is his.

The fourth is different and it is the one that shows what is being attacked. Swearing by your own head is refused “for you cannot make one hair white or black”. That is not an argument about God's ownership of creation. It is an argument about your own incompetence as collateral. A man who swears by his head is pledging something he does not control, which means he is pledging nothing at all.12 He has produced the sound of a guarantee without the substance of one.

Read together, the four reasons are not four instances of a general ban. They are a demolition of a specific technique. Somebody was using these formulas, and using them for a reason, and Jesus is closing them one at a time and explaining why each one fails.

The Decisive Evidence: Matthew 23

This is the argument that carries the paper, and it comes from Matthew himself.

In chapter 23 Jesus returns to oaths and describes the system directly: “Woe to you, blind guides, who say, 'Whoever swears by the temple, that is nothing; but whoever swears by the gold of the temple is obligated.'”13 There it is, named. A graded scheme in which some formulas bind and others do not, and a man who knows the grades can sound solemn while retaining an exit.

Now watch what Jesus does with it, because this is the point an absolutist reading has to explain. He does not say the formulas are all void. He says they all bind: “whoever swears by the altar, swears both by the altar and by everything on it. And whoever swears by the temple, swears both by the temple and by Him who dwells within it. And whoever swears by heaven, swears both by the throne of God and by Him who sits upon it.”14

That is reasoning conducted from inside an oath-taking framework. Jesus tells men who are swearing that their oaths are more binding than they think, and he tells them so in order to shut down evasion. Nowhere in the woe does he add that they should not have been swearing in the first place. That silence is suggestive rather than probative, since the woe is addressed to opponents rather than to disciples and its subject is their evasion, not their practice.

Add that this is the same Gospel, the same author, and the same subject, and a reading of 5:34 as a blanket prohibition becomes harder to hold: eighteen chapters later Matthew records Jesus arguing about which oaths bind rather than teaching that none should be sworn. This is the load-bearing argument of the paper, and it has a standard reply which is taken up below rather than left for a reader to supply.

The dating supports the reading rather than undermining it. The formulas in chapter 23 presuppose a functioning temple with gold in it and an altar with offerings on it, which points toward a practice current before AD 70 and makes this contemporary evidence rather than a later reconstruction.15

Anything Beyond These

Verse 37 gives the alternative and the reason: “But let your statement be, 'Yes, yes' or 'No, no'; anything beyond these is of evil.” Two things need saying about this sentence, one of which cuts against the position argued here.

The phrase rendered “of evil” is ek tou ponērou, and the form is identical in the masculine and the neuter, so morphology cannot decide whether Jesus means evil in general or the evil one. Matthew's two undisputed uses of the substantive, at 13:19 and 13:38, are both personal.16 Of the five singular substantival uses in the Gospel, two are clearly personal and three are contested in varying degrees, so the personal reading carries some weight here without being settled by the count. The personal reading is the sharper one: the elaborate formula is not merely regrettable, it has a source.

The point that cuts the other way should be conceded now rather than buried in the objections section. Jesus does not say that anything false is of evil, or that anything evasive is of evil. He says anything beyond yes and no. The category is excess, and an oath is by definition an excess over a plain yes. That is the strongest single sentence in the absolutist's case and it is in the text.

The reply available is that verse 37 states an ideal rather than a boundary, and that Matthew's own subsequent narrative shows the boundary lying elsewhere. It is a real reply. It is not a refutation, and it should not be dressed as one.

James 5:12

James supplies the only other New Testament witness to this tradition, and it is verbally independent of Matthew rather than a quotation of it, which makes it genuinely corroborating testimony.17 He writes, “But above all, my brethren, do not swear, either by heaven or by earth or with any other oath; but your yes is to be yes, and your no, no, so that you may not fall under judgment.”

Two features favor the absolutist. James generalizes explicitly with “any other oath”, closing the door on the argument that only listed formulas are in view, and he attaches a sanction, which the Matthean version does not.

Two features favor the reading here. The first is the setting: James places this at the end of a passage about speech under pressure, sandwiched between endurance under suffering and the instruction to call the elders. The man in view is not a witness in a courtroom. He is a man under strain whose speech is inflating. The second is that James, writing to Jewish believers who knew the formula system firsthand, lists heaven and earth, which are the first two of the four Jesus named. He is invoking the same target, not opening a new one.

Neither set of features is decisive. James 5:12 is best described as a text that both readings can hold without embarrassment, and the honest way to write it is to say so.

Where This Reading Is Weakest

Five objections, in descending order of force.

The first goes at the argument this paper leans on hardest, and it is the one an absolutist should raise. Matthew 23:16-22, the reply runs, is ad hominem. Jesus argues from his opponents' own premises in order to trap them inside a system he rejects, and this Gospel shows him doing exactly that. When he asks, “If I by Beelzebul cast out demons, by whom do your sons cast them out?”, he reasons from a premise he plainly does not hold.18 Inside chapter 23 itself he tells the scribes they should have tithed mint and dill and cummin without neglecting the weightier provisions, affirming an obligation he is not thereby establishing as a standing rule for disciples.19 On that reading his point in 23:16-22 is that the scribes cannot keep their own scheme straight, and it tells us nothing about whether he would have his disciples swear. This is the standard absolutist reply and the one a reader of the critical commentaries will reach for.20

The answer available is real and it is narrow. Matthew 12:27 is a genuine parallel and it is also instructive, because of what Jesus does not do there: he turns the charge back on his accusers without telling their exorcists that their work carries more authority than they supposed. In 23:20-22 he does the opposite. He escalates obligation. He does not say the formulas bind nobody, he says each one reaches God and binds more than his opponents assumed, and escalating the force of an oath presupposes that an oath can carry force. The tithing case in 23:23 does not undercut this, since a Mosaic requirement still in force during his ministry is not a premise he rejects. The reply does not dissolve the objection, and a reader who finds the ad hominem framing persuasive will not be moved by it. It does mean the objection carries a cost of its own.

The second has already been given: verse 37's category is excess, not falsehood. Every qualified reading has to soften a sentence that is not written softly.

The third is that the four items in verses 34 to 36 may be a merism rather than a list of loopholes. Heaven, earth, the holy city, and one's own body cover the heights, the ground, the sacred center, and the self. That is arguably the whole available universe of oath objects, and closing all of it is functionally identical to closing oaths.

The fourth is Chrysostom, and it is the objection a New Covenant reader should feel most. He does not argue that oaths were always wrong. He argues that Mosaic permission was a concession to immaturity, “the precepts then uttered had reference to the weakness of them who were receiving the laws”, and that after Christ the concession is withdrawn: “Yes, indeed it is altogether of the evil one; that is, now, after so high a rule of self-restraint; but then not so.”21 That is a redemptive-historical argument, and it is the same shape as arguments I have made elsewhere about what the new covenant changes. It cannot be dismissed as wooden literalism. It has to be answered on its own ground, and the answer is given below.

The fifth is a suspicion of motive, and it is fair. The qualified reading became dominant as the church acquired responsibility for civil society and needed oaths for courts and magistrates. Calvin's rebuttal in the Institutes is aimed at a persecuted minority by a man running a city.22 The reply is not that the suspicion is unworthy. It is that the Particular Baptists, who had no state church, no magistracy, and every reason to resent an oath imposed by authority, kept the qualified reading anyway.

That argument has to be made carefully, because the obvious version of it does not work. Chapter 23 of the Second London Confession is a revision of the Savoy Declaration, itself a revision of Westminster 22, and most of its language and most of its proof texts are inherited rather than independently arrived at.23 Pointing at what they retained shows only that they retained it.

What they changed is the evidence. Westminster and Savoy both state that it is a sin to refuse an oath touching anything good and just when it is imposed by lawful authority. The Baptists deleted that clause.24 They kept the lawfulness of the oath, and they kept the duty to take one lawfully imposed, but they declined to call refusal a sin. That is a smaller edit than it first looks, and it is still an edit running one direction. Whatever the framers' reasons, and they recorded none, the change moves toward the conscience and away from the magistrate. If the qualified reading were merely what magistrates find convenient, this is the group and the moment where it should have given way, and it did not.

What the Rest of the New Testament Does

If the apostolic church held oaths to be sin, several passages become difficult, and they should be graded honestly rather than piled up.

The strongest is Hebrews. The author states, as an unargued premise, that “men swear by one greater than themselves, and with them an oath given as confirmation is an end of every dispute”, and reasons from the validity of that practice to explain why God swore by himself.25 He is not criticizing the custom. His argument requires it to be sound. The Anabaptist answer is precise and must be quoted rather than paraphrased: what God may do, we may not, “for everything is possible for Him”, while “we can do nothing, as is said above by Christ, to keep or perform (our oaths): therefore we shall not swear at all.”26 That answer covers God's oath in verse 13. It does not cover verse 16, which is a statement about what men do, offered without censure.

The trial scene is often overstated and should be handled carefully. The high priest says, “I adjure You by the living God, that You tell us whether You are the Christ, the Son of God”, and Jesus, who had been silent, answers.27 Jesus does not swear an oath and it is an error to write that he did. What the text reports is narrower: he had been silent, he was adjured, and he answered, obliquely. The inference available, and it is an inference rather than a report, is that he did not treat a formal adjuration in God's name as an evil to be refused. That is the same Gospel again, twenty-one chapters on.

Paul's asseverations are the weakest tier and should be labeled as such. “But I call God as witness to my soul” is an invocation of God as witness, not an oath formula, and an absolutist can reasonably say Paul stayed on the right side of the line.28 The useful move is not to assert that these are oaths but to note that Augustine, reading the same texts, classed “God is my witness” as a form of swearing and concluded that Paul's practice shows how “swear not at all” is to be understood.29

Answering Chrysostom

Chrysostom's argument deserves its own answer because it is the serious one, and because the framework this paper works in makes it more rather than less plausible.30

The Mosaic legislation on vows was covenant-specific administration. It governed vows made to the LORD within a theocratic nation with a functioning sanctuary, and much of it, the valuation schedules and the household provisions, has no application outside that polity.31 That body of law ended with the covenant it administered. So far Chrysostom and I agree.

Where we part is on what replaced it, and the text of verse 33 decides it. Jesus does not withdraw a permission. He addresses a practice. His four reasons in verses 34 to 36 are reasons about ownership and incapacity, not reasons about a season having passed. If the argument were “this was allowed then and is not allowed now”, the reasons would be redemptive-historical, and they are not. They are theological in a way that would have been just as true in Moses' time. Heaven was God's throne then too.

The direction of change is upward, and the law itself pointed that way before Christ did. Deuteronomy already says that “if you refrain from vowing, it would not be sin in you”.32 The safest vow was the one never made. What Christ does is take that hint and make it the rule, moving the regulation off the oath and onto the speaker.

Notice the asymmetry that proves the standard rose rather than relaxed. Under Moses a man could keep the letter of the law while being a man whose bare word meant nothing, provided he performed his formal vows. Under Christ that man is precisely the one being addressed. The Mosaic administration supplied legal machinery for oath-keeping. Christ removes the need for the machinery by dealing with the man.33

This also explains why an absolute prohibition is the harder position for a reader who takes the Old Testament seriously. Swearing by God's name is not merely permitted there. It is commanded, and truthful swearing is named as part of Israel's witness to the nations.34 Whatever else is true, an act God commands and performs is not intrinsically evil.

A Distinction Worth Keeping

Two kinds of oath are usually distinguished, and the distinction sharpens what is at stake. An assertory oath confirms a claim about the past or present. A promissory oath binds future action.

The Mosaic background in verse 33 is entirely promissory: vows to be performed. The formulas in Matthew 23 appear to cover both. A courtroom oath is assertory. Marriage vows, ordination vows, church membership covenants, and the enlistment oath are promissory.

That last group is worth pressing, because the Anabaptist and Mennonite traditions read Matthew 5:34 absolutely and still administer marriage vows. A man and a woman stand in front of a congregation and bind themselves for life, before God, in language nobody mistakes for a casual promise. If verse 37 forbids everything beyond yes and no, the wedding is a problem before the courthouse is. It should be said that the Quakers did draw the line, and drew it by rebuilding the thing itself: a Friends marriage is a declaration the couple make before the meeting rather than a vow administered to them. That is what consistency on this question actually costs, and it is more interesting than the inconsistency. What remains of the objection therefore bites only where vows are administered by traditions that read verse 34 absolutely, which is a narrower claim than it first appears.

The Westminster divines identified the same abuse Jesus identified, and said so in a sentence that could serve as a summary of this paper: “An oath is to be taken in the plain and common sense of the words, without equivocation, or mental reservation.”35 Equivocation and mental reservation are exactly what the graded formula system existed to permit. Westminster 22.4 and Matthew 5:34 are the same doctrine in two idioms.

What This Asks of a Man

None of this is finally about courtroom procedure, and a paper that ends on the judicial question has missed the passage.

The oath system existed because men wanted to sound binding while remaining free. That is the condition Jesus is diagnosing, and the tell is not the oath. It is the elaboration. When a man says he swears on his mother's grave, he has told you what he thinks his plain word is worth, and he has told you without meaning to. The scaffolding is a confession.

So the question verse 37 puts is not whether a Christian may take a legal oath. It is why a man needed the scaffolding in the first place. If your yes has to be reinforced to be believed, something has already gone wrong upstream of the oath, and it is not going to be fixed by a better formula. Matthew has a line for this: “For the mouth speaks out of that which fills the heart.”36

There is a version of this that has nothing to do with a courtroom. A man tells his wife he will handle it. He tells his son he will be there. Six months later the words have been said so many times without following through that they have stopped carrying weight, and he starts reinforcing them. He says he promises. He says he swears. The reinforcement is not strength. It is the sound a word makes when it has been spent.

The repair is not a vocabulary change. It is doing the thing you said. A man whose yes has been reliable for ten years does not have to swear, because nobody in his house is asking him to.

The Cost of the Other Reading

It would be easy to close by declaring the Anabaptists and the Quakers simply wrong, and that would be both ungenerous and inaccurate.

Quakers refused all oaths on a literal reading of Matthew 5:34 and James 5:12, and they paid for it in prison, property, and the right to testify. Parliament eventually granted relief, first in a 1696 act permitting a solemn affirmation, which many Friends rejected because its wording still invoked Almighty God, and then in a 1722 act granting a form with no reference to God at all. Neither act restored what the opening of this paper described: affirmation remained excluded from criminal cases, jury service, and office-holding, and Friends could not give evidence in criminal proceedings until 1828.37

The same accommodation appears in the American founding documents, and it appears repeatedly. The Constitution prescribes “Oath or Affirmation” for the president, for senators sitting on an impeachment, and for legislators and for executive and judicial officers of the United States and of the several states, and the federal enlistment statute still offers both forms.38

So the legal system I swore into is built on the assumption that the reading I have argued against might be right, and must be accommodated if it is. That is worth sitting with. A minority held a reading at cost for two centuries, and the result is a protection for conscience that my reading never had to produce because it never needed one. Being right is not the only thing a position can be, and being wrong is not the only thing it can cost.

Conclusion

Matthew 5:33-37 attacks an oath system, not the oath. The evidence is Matthew's own: the formula in verse 33 introduces received teaching rather than bare citation, which verse 43 shows is possible; the four prohibitions come with four reasons that dismantle a technique rather than a practice; and in chapter 23 Jesus argues that the same formulas bind more than his opponents thought, which escalates the force of an oath rather than voiding it.

The reading has costs and they have been named. The heaviest is that Matthew 23 may be ad hominem, and the answer offered here narrows that objection rather than removing it. Verse 37's category is excess rather than falsehood. The four objects may be a merism. Chrysostom's redemptive-historical case is serious, and the answer to it is an argument rather than a proof. On all of these the absolutist has real ground.

What is not in doubt is the demand. The standard rose. Under Moses a man had to perform his vows; under Christ he has to be the kind of man whose word does not require reinforcement. That is harder, not easier, and it cannot be satisfied by refusing to swear. A man can decline every oath in his life and still be a man nobody believes.

Let your yes be yes. Then go be the man that sentence describes.

Notes

  1. Matthew 5:34. All Scripture quotations are from the New American Standard Bible, 1995 edition.
  2. The federal enlistment oath, 10 U.S.C. § 502. The statute itself provides both forms, “solemnly swear (or affirm)”, which is a point this paper returns to at the end. The permission to omit the closing words “So help me God” rests on service regulation and department policy rather than on the statute, which prescribes no exception.
  3. Matthew 5:17, 5:20.
  4. On the governing function of 5:17-20 over the six units, see R. T. France, The Gospel of Matthew, New International Commentary on the New Testament (Grand Rapids: Eerdmans, 2007), on 5:17-20. It should be said that whether the units correct the Law itself or the received handling of it is genuinely disputed. This paper takes the second, which is the more common evangelical reading and the one 5:43 supports, but the first has serious defenders and the argument here depends on the choice.
  5. Matthew 5:43.
  6. Compare the Qumran Community Rule, 1QS 1:9-10, which directs love for the sons of light and hatred for the sons of darkness. The point is not that Jesus is quoting Qumran but that the sentiment existed as received teaching without existing as Scripture.
  7. Leviticus 19:12; Numbers 30:2; Deuteronomy 23:21-23. The NASB sets verse 33 in small capitals, the convention it uses for Old Testament quotation, which is correct. The material is Scripture. It is Scripture in summary rather than in citation.
  8. For the absolutist case set out at strength, together with the Jewish background and the history of interpretation, see W. D. Davies and Dale C. Allison Jr., A Critical and Exegetical Commentary on the Gospel according to Saint Matthew, vol. 1, Matthew I-VII, International Critical Commentary (Edinburgh: T&T Clark, 1988), on 5:33-37.
  9. John Calvin, Commentary on a Harmony of the Evangelists, on Matthew 5:34. Calvin's stronger argument in the same place is not grammatical: he charges his opponents with pressing a single word while passing over the scope of the passage with closed eyes.
  10. On the general danger of loading aspectual freight onto the aorist, see Daniel B. Wallace, Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament (Grand Rapids: Zondervan, 1996), 554-565.
  11. Matthew 5:34-35, drawing on Isaiah 66:1 and Psalm 48:2.
  12. On the force of the hair image and its place in the sequence, see John Nolland, The Gospel of Matthew: A Commentary on the Greek Text, New International Greek Testament Commentary (Grand Rapids: Eerdmans, 2005), on 5:36.
  13. Matthew 23:16. He repeats the structure at 23:18 with the altar and the offering on it.
  14. Matthew 23:20-22.
  15. Charles L. Quarles, “The Oath Formulas of Matthew 23:16-22 as Evidence for a Pre-70 Date of Composition for Matthew's Gospel,” Tyndale Bulletin 72 (2021): 1-24. Quarles is careful that the formulas cannot establish the date conclusively, and the claim here is correspondingly modest. This point matters because the fullest rabbinic evidence on oath formulas sits in the Mishnah, redacted around AD 200, and an argument resting on it alone would be anachronistic. The internal evidence of Matthew 23 does not have that problem.
  16. The five singular substantival uses are 5:37, the verse in question, along with 5:39, 6:13, 13:19, and 13:38. The last two are clearly personal. 5:39 is most often taken as human, though it has defenders on the other side, and 6:13 is disputed, with the NASB margin offering “the evil one” as an alternative rendering. Matthew 13:39 is sometimes cited here in error: it reads “the enemy who sowed them is the devil” and uses a different word. The plural at 13:49 is a substantival adjective of a different order and is left out of the count. See BDAG, s.v. ponēros.
  17. Matthew 5:33-37 has no Lukan parallel and is Matthean special material. James 5:12 differs from Matthew in structure and in its sanction while sharing key vocabulary, which most read as independent access to the same dominical tradition rather than as literary dependence on Matthew. See Douglas J. Moo, The Letter of James, Pillar New Testament Commentary (Grand Rapids: Eerdmans, 2000), on 5:12, who treats the verse as among the closest contacts between James and the Jesus tradition.
  18. Matthew 12:27. This is the clearest Matthean instance of an argument run from a premise the speaker rejects, and it is the first text an absolutist should cite here.
  19. Matthew 23:23.
  20. On the absolutist reading of 5:33-37 as the more natural one, and on its long life in the church, see Ulrich Luz, Matthew 1-7: A Commentary, trans. James E. Crouch, rev. ed., Hermeneia (Minneapolis: Fortress, 2007), on 5:33-37 and in his history of interpretation.
  21. John Chrysostom, Homilies on the Gospel of Saint Matthew 17, in Nicene and Post-Nicene Fathers, first series, vol. 10.
  22. John Calvin, Institutes of the Christian Religion 2.8.26-27. Calvin charges the Anabaptists with making a stumbling-stone of Christ by setting him against the Father, which is a serious charge leveled at people who were being executed.
  23. Compare The Baptist Confession of Faith (1689), 23.2, with Westminster Confession of Faith 22.2. The proof texts at 23.2, which include Matthew 5:34 and 5:37, James 5:12, and Hebrews 6:16, largely carry over. Citing them as evidence of independent Baptist exegesis would prove nothing except that the framers copied a source.
  24. Compare Westminster Confession of Faith 22.3 and Savoy Declaration 23.3, which read, “Yet is it a sin to refuse an oath touching anything that is good and just, being imposed by lawful authority”, with The Baptist Confession of Faith (1689), 23.3, which rewrites the paragraph and omits the clause. The omission should not be overstated: 23.2 retains Westminster's “so a lawful oath being imposed by lawful authority in such matters, ought to be taken”. The duty survives. The charge of sin for refusing does not.
  25. Hebrews 6:13-18.
  26. The Schleitheim Confession (1527), Article VII. The same article states that Christ “prohibits all swearing to His (followers), whether true or false”, and grounds the prohibition in Matthew 5:36, human incapacity to change the color of a hair.
  27. Matthew 26:63-64. The usual reconstruction supplies Leviticus 5:1, under which a person placed under adjuration who withholds testimony bears guilt, as the legal background to the silence being broken. That is a reconstruction and not something the passage states.
  28. 2 Corinthians 1:23; compare Romans 1:9, Philippians 1:8, Galatians 1:20.
  29. Augustine, De sermone Domini in monte 1.17.51-52, in Nicene and Post-Nicene Fathers, first series, vol. 6. Augustine also observes that the Jews thought oaths by heaven, earth, Jerusalem, or one's head were not binding, and that Christ shows all creation stands under God so that such oaths bind after all. That observation anticipates the argument made here. Augustine's own resolution of the prohibition does not: he treats swearing as necessary rather than good, a concession to hearers slow to believe, and dangerous chiefly as a habit. On the main question he stands nearer to Chrysostom than to this paper.
  30. For the qualified reading defended from within conservative evangelical and Baptist scholarship, see D. A. Carson, “Matthew,” in Matthew-Mark, rev. ed., Expositor's Bible Commentary 9 (Grand Rapids: Zondervan, 2010), on 5:33-37; and Charles L. Quarles, Sermon on the Mount: Restoring Christ's Message to the Modern Church, NAC Studies in Bible and Theology (Nashville: B&H Academic, 2011), on 5:33-37.
  31. Leviticus 27; Numbers 30. These are the clearest cases of legislation that cannot be transposed out of the covenant that housed it.
  32. Deuteronomy 23:22.
  33. It would be an overstatement, and a common one, to say that Moses regulated externals while Christ addressed the heart. Deuteronomy 6:5, 10:16, and 30:6 address the heart directly, as does Leviticus 19:17-18. The difference is not interiority. It is that the machinery becomes unnecessary.
  34. Deuteronomy 6:13, 10:20; Jeremiah 4:2. God himself swears, at Psalm 110:4 and Hebrews 6:13, and an angel swears at Revelation 10:5-6.
  35. Westminster Confession of Faith (1646), 22.4. Compare 22.1, which defines a lawful oath as an act in which the swearer calls God to witness and to judge him according to the truth of what he swears.
  36. Matthew 12:34. The immediate context, 12:33-37, ends with the warning that men will give account for every careless word, which is the same subject arriving from a different direction.
  37. 7 & 8 Will. III c. 34 (1696), sometimes dated 1695 by session; 8 Geo. I c. 6 (1722). The 1722 form reads, “I, A.B., do solemnly, sincerely, and truly declare and affirm.”
  38. U.S. Const. art. I, § 3, cl. 6; art. II, § 1; and art. VI, cl. 3. The phrase also appears at Amend. IV, in the unrelated context of warrants. 10 U.S.C. § 502. Whether the American provisions descend causally from the English relief acts is commonly argued and is not demonstrated here.

Bibliography

  • Augustine. Our Lord's Sermon on the Mount. In Nicene and Post-Nicene Fathers, first series, vol. 6, edited by Philip Schaff. Buffalo: Christian Literature Publishing, 1888.
  • Bauer, Walter. A Greek-English Lexicon of the New Testament and Other Early Christian Literature. 3rd ed. Revised and edited by Frederick William Danker. Chicago: University of Chicago Press, 2000.
  • Calvin, John. Commentary on a Harmony of the Evangelists, Matthew, Mark, and Luke. Vol. 1. Translated by William Pringle. Edinburgh: Calvin Translation Society, 1845.
  • Calvin, John. Institutes of the Christian Religion. Edited by John T. McNeill. Translated by Ford Lewis Battles. Library of Christian Classics 20-21. Philadelphia: Westminster, 1960.
  • Carson, D. A. “Matthew.” In Matthew-Mark. Rev. ed. Expositor's Bible Commentary 9. Grand Rapids: Zondervan, 2010.
  • Chrysostom, John. Homilies on the Gospel of Saint Matthew. In Nicene and Post-Nicene Fathers, first series, vol. 10, edited by Philip Schaff. Buffalo: Christian Literature Publishing, 1888.
  • Davies, W. D., and Dale C. Allison Jr. A Critical and Exegetical Commentary on the Gospel according to Saint Matthew. Vol. 1, Matthew I-VII. International Critical Commentary. Edinburgh: T&T Clark, 1988.
  • France, R. T. The Gospel of Matthew. New International Commentary on the New Testament. Grand Rapids: Eerdmans, 2007.
  • Luz, Ulrich. Matthew 1-7: A Commentary. Translated by James E. Crouch. Rev. ed. Hermeneia. Minneapolis: Fortress, 2007.
  • Moo, Douglas J. The Letter of James. Pillar New Testament Commentary. Grand Rapids: Eerdmans, 2000.
  • New American Standard Bible. 1995 ed. La Habra, CA: The Lockman Foundation, 1995.
  • Nolland, John. The Gospel of Matthew: A Commentary on the Greek Text. New International Greek Testament Commentary. Grand Rapids: Eerdmans, 2005.
  • Quarles, Charles L. Sermon on the Mount: Restoring Christ's Message to the Modern Church. NAC Studies in Bible and Theology. Nashville: B&H Academic, 2011.
  • Quarles, Charles L. “The Oath Formulas of Matthew 23:16-22 as Evidence for a Pre-70 Date of Composition for Matthew's Gospel.” Tyndale Bulletin 72 (2021): 1-24.
  • The Baptist Confession of Faith. 1689.
  • The Schleitheim Confession of Faith. 1527. Translated by John C. Wenger. Mennonite Quarterly Review 19 (1945): 243-253.
  • The Westminster Confession of Faith. 1646.
  • Wallace, Daniel B. Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament. Grand Rapids: Zondervan, 1996.

Written by Jake. If you want to push back on any of it, write me. I would rather be corrected than comfortable.

← All essays