A Review of Jonathan Bernier's Rethinking the Dates of the New Testament

Dr. Jonathan Bernier recently published a book (Rethinking the Dates of the New Testament; Baker Academic, 2022) defending an early chronology for the New Testament corpus, the first study of its kind since John A. T. Robinson's 1976 book Redating the New Testament. With the exception of some of the Pauline epistles, Bernier argues for dates of composition "twenty to thirty years earlier than is typically supposed by contemporary biblical scholars". His text is systematic, thorough, and approachable, though the strength of the arguments varied depending on the NT book in question. In this review, I will discuss the pros and cons of the book, overview the arguments adduced for an early date, and assess the strength of his conclusions.

Introduction

These are the dates Bernier defends:

Matthew: 45-59
Mark: 42-45
Luke: 59
John: 60-70
Acts: 62
Romans: winter of 56/57
1 Corinthians: early 56
2 Corinthians: late 56
Galatians: 47-52
Ephesians: 57-59
Philippians: 57-59
Colossians: 57-59
1 Thessalonians: 50-52
2 Thessalonians: 50-52
1 Timothy: 63-64 if Pauline, 60-150 if pseudo-Pauline
2 Timothy: 64-68 if Pauline, 60-150 if pseudo-Pauline
Titus: 63-64 if Pauline, 60-175 if pseudo-Pauline
Philemon: 57-59
Hebrews: 50-70
James: prior to 62
1 Peter: 60-69
2 Peter: 60-69 if Petrine, 60-125 if pseudo-Petrine
1 John: 60-100
2 John: 60-100
3 John: prior to 100
Jude: prior to 96
Revelation: 68-70

1 Clement: 64-70
Didache: 60-125
Epistle of Barnabas: 70-132
Shepherd of Hermas: 60-125

Perhaps the best thing about the book is the systematic treatment of each topic. Each section begins by listing and numbering each piece of evidence to be considered, which correspond to the bold-faced section titles. At the end of each section is a conclusion recapping the evidence considered and stating the current possible date range. There is also a cumulative conclusion at the end of each chapter giving a higher-level recap. The result is an argument that's extremely easy to follow.

My disagreements were principally regarding the evidential weight assigned to various data. I will discuss some of these below.

Overview of the Book

The book is divided into an introduction and five parts. I will briefly discuss each of them.

Introduction

Bernier explains the three main chronologies defended by New Testament scholars as so:

• Lower (defended in this volume)—most of NT prior to 70
• Middle (majority view)—NT generally between 70-100, with the exception of the undisputed Pauline epistles
• Higher—dates much of the NT to the second century
  

Next, Bernier surveys the history of scholarship on the compositional dates of the New Testament and notes that this is the first book-length defense of the lower chronology in almost fifty years, while the middle and higher chronologies haven't been defended at such length since the Victorian era. Accordingly, he argues that three books are needed, each rearticulating a robust argument for one of the three dating schemes. With this volume, he is attempting to do that for the lower chronology.

The next section discusses some of the shortcomings of A.T. Robinson's infamous book. Specifically, Bernier argues 1) that Robinson relies too much on arguments from silence, and 2) that while Robinson is correct to reject NT references to persecution as necessarily referring to the Domitianic persecution, he makes many of the same errors with the Neronic persecution—failing to establish that many references to persecution in the NT refer specifically to the Neronic persecution and not something else.

The next section discusses criteria that will be used for generating hypotheses: synchronization, contextualization, and authorial biography. Bernier uses these three criteria to structure each discussion of dating. Synchronization deals with whether certain portions of a text are more intelligible before or after certain dates. This is often used with reference to the destruction of the Temple in the year 70. For example, a description of the Temple that uses the present tense is more intelligible before the year 70, and thus should incline us toward a pre-70 date. Contextualization deals with the themes of the book in relation to early Christian development in areas like Christology or ecclesiology. Authorial biography deals with what we know about the author. It can yield the most precise dates of the three criteria, such as dating Romans to the winter of 56/57.

Part 1 - The Synoptic Gospels and Acts

Matthew, Mark, and Luke-Acts are the first texts considered by Bernier. He discusses several data that have figured prominently in discussions of the dating of the Synoptics and considers most of them to be nonprobative for purposes of establishing a compositional date. A notable exception is the reference to the desolating sacrifice in the "texts of desolation", which Bernier argues makes the most sense if the Synoptics were written prior to 70.

Bernier also discusses some concerns specific to Luke-Acts, such as the unity of Luke-Acts, whether Luke knew Josephus, the relevance of Marcion's Gospel, Luke's (non)use of Paul's letters, and the enigmatic ending of Acts. He puts a lot of weight on this last piece of evidence, citing Karl Armstrong's new book on the subject, Dating Acts in its Jewish and Greco-Roman Contexts.

Chapter 2 seeks to narrow down the pre-70 date by providing evidence from contextualization and authorial biography. Bernier argues that Maurice Casey's work on Markan translation of Aramaic is probably nonprobative for the purposes of dating the Gospel. He then discusses Crossley's argument from Mark's silence on the Gentile mission, concluding that it lends some support to pre-45 dating of Mark, but "this hypothesis is sufficiently tentative that it should be supported by additional argumentation". (71) Unfortunately, it seems to me that Bernier never supplied this additional argumentation.

Bernier also discusses the relationship between Mark and Peter. He argues that Peter was present in Rome ca. 42 but notes that the evidence for this is late and thus not decisive. He concludes that Mark was written sometime between 42 and 45. 

Pivoting into a discussion of Luke, Bernier argues that the famous "we-passages" are best explained if the author of Acts was present for the events in question. As a key piece of evidence, he cites the fact that the we-passages begin in Philippi and trail off when the narrative returns to Philippi. He argues that the best time for Luke to have written his Gospel would be ca. 59, with Acts being written sometime between 60-62.

Lastly, in discussing Matthew, Bernier argues that a date can't be nailed down with as much precision. He concludes that Matthew must postdate 45, since it postdates Mark, and predate 59, since it predates Luke. 

Part 2 - The Johannine Tradition

The synchronization section of chapter 3 covers several lines of evidence. Most of these are deemed to be nonprobative or otherwise of limited evidential value. Of particular interest was Bernier's discussion of the Birkat Haminim and the tenuity of the argument that it necessitates a post-80 date for John.

The lone piece of positive evidence given for a pre-70 date is the use of the present tense to refer to the pool of Bethsaida in John 5.2: "in Jerusalem by the Sheep Gate there is a pool...". Accordingly, Bernier spends the better part of six pages defending its evidential value. I think he succeeds in showing that it should be taken as evidence for an early date. However, since this is the only significant factor that weighs in Bernier's analysis of the dating of John's Gospel, it should be taken with caution. Of course, we can't always expect a multi-pronged cumulative case for dating a specific text, but I don't think agnosticism on the matter of John's date is a particularly irrational position.

In the next sections Bernier critiques further arguments for a later date of John, such as Johannine Christology. Arguing that John's Christology is approximately as high as that of Paul, he correctly asserts that the Christological factor doesn't have much bearing on the date of John's Gospel.

There is little evidence bearing on the dating of the epistles of John. Bernier concludes that 1 and 2 John were written sometime between 60-100 and 3 John was written no later than 100, as he was unable to establish a lower bound for the latter.

Bernier discusses several factors regarding the date of Revelation:
1. external attestation
2. the succession of kings in Revelation 17:9-11
3. the death of Nero in Revelation 13:1-18
4. the passages about the temple in 11:1-2 and 11:13
5. food sacrificed to idols
6. Rome as "Babylon"
7. Revelation's Christology
8. Revelation's ecclesiology
9. Irenaeus' ostensible testimony to a mid-90s date for Revelation

Bernier places the most weight on point #4. Revelation 11:13 reads: "At that moment there was a great earthquake, and a tenth of the city fell; seven thousand people were killed in the earthquake, and the rest were terrified and gave glory to the God of heaven."

Josephus reports that all of Jerusalem fell in 70 AD (Jewish Wars 7.1.1 §§1-4), and moreover, it was not an earthquake that caused the fall but Roman soldiers. Bernier argues: "It is difficult to imagine that an author writing after 70 would employ the fall of Jerusalem for symbolic effect and yet grossly misrepresent the nature of that fall, and more to the point, underestimate its extent." (122)

Concerning point #9, Bernier offers no alternative interpretation but rather argues that the internal evidence should win out. The problem with this, of course, is that Bernier gave only two pieces of internal evidence for a pre-70 date, which were at least partially epistemically dependent, so the counterevidence they can sustain is limited.

Part 3 - The Pauline Corpus

Part 3 is structured differently than the other parts. Instead of dating texts in both chapters, Bernier spends Chapter 5 discussing how to date the Pauline corpus and Chapter 6 actually dating it.

Of particular note is his discussion of the importance of Acts in dating Paul's letters. He defends an interesting argument for the chronological accuracy of Acts (e.g., whether Acts narrates events in the order they happened): he adduces four key events that can be dated independently (the crucifixion of Jesus, the death of Herod Agrippa, Gallio's tenure in Corinth, and the succession from Felix to Festus) and demonstrates that Acts narrates these all in the order that they occurred. Obviously, this isn't sufficient to demonstrate beyond reasonable doubt that Acts is chronologically accurate, but it certainly points in that direction.

Here are Bernier's dates for the Pauline corpus, with brief explanations:

1 and 2 Thessalonians
On the basis of Paul's Corinthian sojourn, Bernier argues that 1 and 2 Thessalonians were written between the years 50 and 52.

1 and 2 Corinthians
On the basis of Paul's Ephesian sojourn and the data contained in 1 Corinthians 16.8 ("But I will remain in Ephesus until Pentecost") Bernier argues that 1 Corinthians was written in early 56. On the basis of correlation with Acts, Bernier argues that 2 Corinthians was written in late 56.

Romans
On the basis of correlation with Acts, Bernier argues that Romans was written in the winter of 56/57. The integrity of Romans is considered to be nonprobative for purposes of establishing the compositional date.

Galatians
Bernier gradually narrows the lower bound for Galations from 31 to 40 to 42 and finally to 47. On the basis of correlation with Acts, the upper bound is argued to be 52.

Ephesians, Colossians, and Philemon
The primary datum for dating these three epistles is Paul's imprisonment in Caesarea from 57-59. Bernier argues that they were all written from this time period. Notably, Bernier does not seem to entertain serious doubts as to the authenticity of Ephesians and Colossians.

Philippians
Bernier judges the integrity of Philippians to be more probable than not, arguing that it was written during Paul's imprisonment in Caesarea from 57-59.

1 Timothy, 2 Timothy, and Titus
Given the disputes as to the authenticity of these letters, Bernier adduces two date ranges for each of them: one if Pauline, the other if pseudo-Pauline. He argues from multiple lines of evidence that these letters, if authentic, would postdate Paul's career as recorded in Acts. He also argues that Paul plausibly had a post-Acts career from ca. 62-68. He concludes that 1 Timothy and Titus were written no later than 150 and 175 (respectively), if pseudo-Pauline, and from 63-64 if Pauline, and that 2 Timothy was written no later than 150 if pseudo-Pauline, and from 64-68 if Pauline.

Part 4 - Hebrews and the Letters of James, Peter, and Jude

Bernier dates Hebrews between the years 50 and 70. The majority of his analysis consists of refuting arguments for a later date. To establish the lower bound of 50, Bernier cites Hebrews 13.23, specifically the phrasing "our brother, Timothy". He argues that this makes most sense after Timothy had entered into ministry, ca. 50. To establish the upper bound of 70, Bernier cites Hebrews 10.1-3 ("would not the sacrifices have ceased?"). This makes more sense if the Temple hadn't been destroyed. Otherwise, the point of the argument could be lost: "If in 85 I were to ask, 'Would not the sacrifices have ceased being offered?' you likely would respond, 'But they have!'" (191)

Bernier dates James no later than 62, the date of James' death. His primary reason for doing this is his acceptance of traditional authorship. Bernier discusses and refutes many arguments for a later date, though in some cases he didn't condemn those arguments as strongly as could be warranted, such as the claim that James 3.9-10 betrays knowledge of the Birkat Haminim.

Bernier's lower bound for 1 Peter is the year 60. This is reached on the basis of 1 Peter's knowledge of Romans and Ephesians, as argued by Ora Delmer Foster. The authorial biography section includes a discussion of objections to the traditional authorship view. Bernier's upper bound is 69, which he judges the last possible year of Peter's death. Notably, Bernier does not even provide a date for if the epistle is pseudo-Petrine, despite his doing so for 2 Peter, as he believes 1 Peter is likely authentic.

Bernier argues that "2 Peter is probably the strongest candidate for pseudonymous authorship in the NT corpus." (228) If authentic, he gives the same 60-69 date range as 1 Peter. If pseudonymous, the upper bound is pushed out to 125.

There is little evidence for the dating of Jude, due to the brevity of the letter and the questions surrounding its authorship. Bernier concludes that Jude was probably dependent on 2 Peter, setting a lower bound somewhere in the 60s. He argues for an upper bound of 96, based on the fact that Jude had probably passed away by the time of the Domitianic reign.

Part 5 - Early Extracanonical Writings

References to the deaths of Peter and Paul solidly exclude a date for 1 Clement earlier than 64. Narrowing down this range, Bernier refers to 1 Clement 40-44, which uses the present tense to refer to the temple administration. Judging most other lines of evidence to be nonprobative, Bernier argues that this present tense should incline us to prefer a pre-70 date for the letter, when the temple was still standing. The final date range is 64-70.

The Didache is argued to date between 45-125. the upper bound is established by the contextual relevance of concerns about traveling teachers, apostles, or prophets and Gentile inclusion, which Bernier argues are most intelligible in the 40s-60s but look increasingly out of place the later one dates the text.

Very little can be said to narrow down the date range of the Epistle of Barnabas. The reference to the destruction of the Temple in 16.3-4 necessitates a post-70 date. Bernier argues that the verses also tell against a date after the Bar Kokhba revolt, as that would have made the prophecy about the temple being rebuilt less likely. Thus, Bernier settles on a wide date range of 70-132.

The Shepherd of Hermas also resists restriction to a narrow date range, in part due to its composite nature. On the basis of ecclesiology, Bernier gives a date range of 60-125.

Conclusion

Rethinking the Dates of the New Testament is a must-read for anyone interested in the central topic of the book—namely, dating the New Testament. I would contend, however, that the discussion is a bit too esoteric for general reading in New Testament reliability. Bernier has done New Testament scholarship a great service by presenting anew the case for earlier dates of the NT corpus. Due to numerous disagreements about the strength of the evidence as well as a desire to withhold my final conclusions until I've researched the relevant issues further, I was ultimately unconvinced that I should confidently embrace an earlier chronology, but Bernier has certainly shown the flaws in a position of epistemic dogmatism with regard to the consensus dating. I agree with Bernier that more books are needed on this subject, particularly defenses of middle and high chronologies, so that the merits of each case can be more easily judged.

Reviewing the Ehrman-Gathercole Debate on Early Christology

This will be a brief review of the second 2014 debate on the "Unbelievable?" radio show between Simon Gathercole and Bart Ehrman on the topic of Christology, specifically a discussion of the earliest texts in the New Testament such as Romans 1.3-4 and Philippians 2.5-11. The discussion was sparked by the publication of Ehrman's How Jesus Became God and a response volume to which Gathercole contributed titled How God Became Jesus. Timestamps refer to the recording here; note that the timestamps might vary depending on how your browser loads the recording and other factors. The debate starts after the third minute, moderated by Justin Brierley.

In his initial comments, Ehrman says that this issue (who Jesus was) is "fundamentally important for everybody, whether they're a Christian or not a Christian." (4:55). This is a good take. Despite Ehrman's dubious conclusions on the historical Jesus, this remark is a refreshing point of agreement.

Clarifying his views on the evolution of early Christology, Ehrman goes on to say that "Christians didn't come out and call Jesus God for decades after his resurrection." (6:50) Later, he comments about how he changed his mind on the subject of Jesus' burial, saying that the traditions of Jesus having a tomb that was discovered empty are "probably not historical" (8:20). This was not the subject of the debate, so I will not be discussing it here. Those interested in a response to arguments from Ehrman and others concerning the historicity of Jesus' burial can consult this post from jobapologetics or this video from InspiringPhilosophy. Both links cite a plethora of scholarly sources for further research on the topic.

At the 12 minute mark, Ehrman summarizes the consensus dating of the Gospels, placing Mark around 65 or 70. Gathercole agrees with the chronology fleshed out by Ehrman. I don't think the evidence for these dates is strong enough to command the consensus they enjoy, but this is irrelevant to the topic of the debate so I will not discuss it here. I will post a review soon of Jonathan Bernier's new book on the topic.

The discussion of early Christology begins in earnest at 14:18, when Brierley introduces Philippians 2 as a purported example of early high Christology. For reference, here is the text from the NASB:

5 Have this attitude in yourselves which was also in Christ Jesus,
6 who, as He already existed in the form of God, did not consider equality with God something to be grasped,
7 but emptied Himself by taking the form of a bond-servant and being born in the likeness of men.
8 And being found in appearance as a man, He humbled Himself by becoming obedient to the point of death: death on a cross. 
9 For this reason also God highly exalted Him, and bestowed on Him the name which is above every name,
10 so that at the name of Jesus every knee will bow, of those who are in heaven and on earth and under the earth,
11 and that every tongue will confess that Jesus Christ is Lord, to the glory of God the Father.

This passage is regarded by some scholars to be pre-Pauline. At 18:00 Gathercole answers a question about how this conclusion is reached. He emphasizes the unusual vocabulary as one factor, though he qualifies that by saying that the difficulty with this approach is that the corpus of Paul's letters is tiny. This is generally a good principle: while unusual vocabulary is some evidence for non-Pauline origin, since the corpus is sufficiently large to get an idea of Paul's verbal and grammatical inclinations, writers use unusual vocabulary frequently so caution is warranted.

Returning to around the 16 minute mark, the conversation turns to various translation-related issues in the first section of the poem. Both Gathercole and Ehrman think that translating the first part of verse 6 as "being in very nature God" is dubious. Gathercole says that the more controversial part is "over the grasping bit" (17:00). Gathercole notes that translating it as "grasped" suggests that Jesus didn't already have equality with God, but the alternative translation of "exploited" suggests that he did have it but didn't abuse it. At 17:40 Ehrman comments that the word for "grasped" (harpagmos, from arpazō) isn't used very much. Indeed, this is the only occurrence of this verb in the New Testament. Due in part to time constraints, neither scholar lays out a case for preferring one translation over the other.

At 18:50, Ehrman summarizes his interpretation of the Christology of this creed as "Before Jesus became a human, he was a divine being with God in heaven, but he didn't want to seek equality with God... Then, God exalts him more highly than he was before." At 19:40, Ehrman notes that the language of every knee bowing and every tongue confessing is a quotation from Isaiah applied exclusively to YHWH. Thus, he argues, Jesus became equal with God after the resurrection.

At 19:33 Gathercole explains his interpretation of the Philippians passage. He notes that his primary disagreement is over the translation of harpagmos; he finds it convincing that this should be translated as saying that Jesus had equality with God but didn't want to hold onto it. He then comments as to the difficulty of treating a small hymn like this in isolation and suggests going elsewhere in Paul to find his high Christology. Specifically, he cites 1 Corinthians 8.6: "yet for us there is only one God, the Father, from whom are all things, and we exist for Him; and one Lord, Jesus Christ, by whom are all things, and we exist through Him."

Seeing as the primary difference between these two interpretations of the Philippians passage depends on the translation of harpagmos, some comments about this are in order. Scholars have long debated the meaning of this word and the matters of syntax are complex. For those interested in reading more about this term, perhaps the best place to start is Roy Hoover's 1971 landmark study. [1] On the basis of his survey of other uses of the word, Hoover suggested a translation of Philippians 2.6 as "he did not regard being equal with God as something to take advantage of". This went without serious challenge until 1988, when J. C. O'Neill published a brief critique of Hoover's conclusions, arguing that harpagmos is best understood as robbery. [2] Gordon Fee took issue with O'Neill's critique, saying "by turning Hoover's findings into a 'rule', O'Neill eliminates the 'rule' by noting the exceptions. But that is not the same thing as eliminating Hoover's understanding of the idiom." [3] Most recently, Michael Wade Martin published a rebuttal to O'Neill's comments and proposed some refinements to Hoover's thesis. [4] Martin's primary conclusion is that the use of harpagmos itself is ambiguous and its meaning should be settled by context.  This latter position seems safest. I think the context provides a strong case for the equality of Jesus with God by 1) the wording "in the form (morphē) of God" and 2) the theological impracticality of translating harpagmos as "robbed". These considerations are not conclusive, though, so Pauline Christology should not be settled on the basis of this text alone.

There is little to comment on from the 30 minute mark to the 40 minute mark. Ehrman and Gathercole primarily discuss the divisions between creator and created in first-century Judaism, where they generally agree. Gathercole's central contention is that the worship of which Paul declares Jesus worthy would be unfitting for a created being: "If Jesus is a creature, then you jolly well shouldn't worship him." This conclusion comes from passages like Romans 1 ("worshiped and served the creation rather than the Creator"). Thus, Paul viewed Jesus as uncreated.

At 39:30, Gathercole explains that a central problem with an exaltation or adoptionist Christology is that it would make a creature into a non-creature. Ehrman responds by saying that Jesus is still a creature, but now one who has been exalted to the level of God. He uses the analogy of a Roman emperor adopting a son. Further, Ehrman comments that Paul's condemnation in Romans 1 is irrelevant because 1) Paul is talking about worshiping idols, and 2) Jesus has now been exalted to the level of God, an uncreated thing. Later, Gathercole responds to 1) by agreeing that Paul is indeed condemning a specific brand of idolatry, but in doing so he reflects his belief that only the creator is worthy of worship. 

This is an interesting contention from Ehrman. One objection is that it's difficult to view a created Jesus as being equal to God, because the creature/creator distinction precludes any notion of metaphysical equality—the creator will always be greater than the creation.

Ehrman goes on to defend his view that Paul viewed Christ as a created being by saying that "from whom are all things" in 1 Corinthians 8.6 includes Christ. Gathercole references the second part of the verse saying that "through [Christ] are all things". Here I think he drops a good rebuttal. If Ehrman wants to argue that "from [God] are all things" necessarily implies that Christ is a created being (interpreting the "all" literally), then saying that "through [Christ] are all things" would imply that God exists through Christ. I doubt this is a position Ehrman would want to affirm, given his belief that Paul viewed Christ as a created being. Further, if God's existence depends on Christ, it would suggest that Christ is higher than God in some way. Thus, Ehrman would have to concede that God is excluded from the "all" in the phrase "through [Christ] are all things". This significantly weakens his argument that "from [God] are all things" necessarily includes Christ—if the second phrase permits an exception, the first one can as well.

At 44:00 begins the discussion of the "tunnel period". This is the period of about 20 years from Jesus' death to our first Pauline letter. Ehrman says that the best way to figure out what was going on Christologically during the tunnel period is to look at pre-Pauline creeds and hymns in the letters of Paul. While this is probably our best bet for elucidating early Christological beliefs, I don't think it can yield much certainty. Creeds are often condensed, for one thing, and leave out a lot of qualifications that might impact our interpretation. Another concern is that Paul is widely argued to have redacted Christologically inadequate creeds to make them compatible with his own Christology, but the speculation employed to reconstruct the original protocreeds is often quite tenuous and significantly hampers our ability to derive confident conclusions.

At 47:05 Ehrman brings up Romans 1.3-4 for the first time during the debate. Ehrman lays out a basic case for finding an adoptionist Christology in this text. In a few months I will be publishing a series of posts addressing various interpretations of Romans 1.3-4, including a comprehensive analysis of various redaction theories. That series will be reasonably exhaustive. Thus, I will limit my comments here to those of direct pertinence to the debate.

Here is the NASB translation of the text:

3 concerning His Son, who was born of a descendant of David according to the flesh, 
4 who was declared the Son of God with power according to the Spirit of holiness by the resurrection from the dead, Jesus Christ our Lord.

"With power" can also be translated "by power" or "in power". I prefer the latter.

Gathercole's first response, at 48:30, is to emphasize the level of speculation involved in determining whether a given text is a creed. He does, however, think that there's a good chance Romans 1.3-4 is a creed. Where he disagrees with Ehrman is that the creed presents something radically different than what Paul believed. 

At 49:40 Gathercole says that it's much more speculative to say that Paul added "in power" to the creed. Ehrman's initial response seems to indicate that the inclusion of "in power" tells strongly against an adoptionist reading—Jesus became the "son-of-God-in-power" at his resurrection. Thus, if the creed originally taught an adoptionist Christology, "in power" would have to be absent. This is precisely what Ehrman argues: "in power" was probably a Pauline interpolation. Around the 51 minute mark, Ehrman lays out his principal argument for that conclusion—it's the only part of the second section that doesn't correspond to anything in the first section.

To see this, consider how Ehrman contrasts the elements of the creed in his book How Jesus Became God: [5]

A1 Who was descended
A2 from the seed of David
A3 according to the flesh,

B1 who was appointed
B2 the Son of God in power
B3 according to the Spirit of holiness by his resurrection from the dead

Since "son of God" corresponds to "seed of David", Ehrman argues, "in power" is left hanging.

Against Ehrman's argument we may adduce six responses. I will discuss these all more fully in my planned series on Romans 1.3-4:
1. We have to be careful what we conclude from parallelism arguments—it's highly speculative to assume everything will line up perfectly.
2. The strength of the argument is dependent on how we structure the creed. Scholars like Matthew Bates and Gordon Fee structure the various sections of the creed differently than Ehrman, making "in power" no longer unnecessary.
3. There are other parts of the creed that don't line up perfectly. For example, consider the phrases "according to the flesh" and "according to the Spirit of holiness". Technically, one could omit "of holiness" and still have an understandable creed. Someone might respond that the emphasis is on corresponding phrases, not word-for-word similarities—in that case, if "in power" modifies "Son of God" (as many scholars argue) it is no longer out of place. The same parallel dooms the response that "of holiness" is a necessary qualification for the office of "the Spirit"—"in power" could also be a necessary qualification for "Son of God".
4. Even if we accept Ehrman's structuring of the creed, James Dunn argues that "seed of David" and "Son of God" might have been regarded as insufficiently contrastive, because the seed of David was already considered to be the Son of God. Thus, "in power" preserves the antithetical parallelism.
5. There is precedent for this formulation in other Christological hymns (1 Tim 3.16, Heb 1.3).
6. "Son of God in power" is an expression not used by Paul elsewhere. This is not a particularly strong argument, but it's presence is slightly more probable if the expression originated elsewhere than if Paul used it himself.

Gathercole responds that scholars differ as to whether "in power" was there originally. He cites James Dunn's commentary arguing that "seed of David" and "son of God" don't contrast enough, and thus the "in power" is necessary to preserve the antithetical parallelism (point #4 above). At 54:13 Ehrman briefly responds: "My case doesn't rest on whether two words in Romans 1.3-4... there are other preliterary traditions that all point in the same direction." (ellipses indicate Ehrman's incomplete sentence, not an omission). While possibly due to time constraints, it's notable that Ehrman drops the issue after Gathercole provides Dunn's response. He then refers to his book, indicating that he makes a case there for viewing an adoptionist Christology as one of the earliest views.

Ehrman's comment that "there are other preliterary traditions that all point in the same direction" references passages in the speeches in Acts, though he doesn't bring them up in the debate. For engagement with these passages, see C. Kavin Rowe's paper "Acts 2:36 and the Continuity of Lukan Christology" and Michael Bird's book Answering Adoptionist Christology. 

In conclusion, I think Gathercole had the upper hand in this discussion, though both scholars provided robust defenses of their views. The debate covered a lot of ground, though it would have been interesting to see Ehrman defend his redactional thesis for Romans 1.3-4 a bit more. Ultimately, this debate did what all good debates should: offer an introduction to two competing viewpoints and provide resources for further research.

References and Notes
[1] R. W. Hoover, "The Harpagmos Enigma: A Philological Solution," HTR 64 (1971), 95-119
[2] J. C. O'Neill, "Hoover on Harpagmos Reviewed, with a Modest Proposal concerning Philippians 2:6," HTR 81 (1988), 445-49
[3] Gordon Fee, Pauline Christology: An Exegetical-Theological Study. Hendrickson Publishers, 2007, 382
[4] Michael Wade Martin, "ἁρπαγμός Revisited: A Philological Reexamination of the New Testament's 'Most Difficult Word'," JBL 135 (2016), 175-94
[5] Bart Ehrman, How Jesus Became God: The Exaltation of a Jewish Preacher from Galilee. HarperOne, 2014. 166 (eBook)

Setting a Date for the Second Coming: Mark 9:1, 13:30, and Matt. 10:23

This post is the third in a series on the so-called “delay of the parousia”. See my previous posts here and here.

Three texts in particular have been used by many scholars to argue not only that Jesus thought the end of the world was “near”, but also that he more boldly set a generational deadline for it. Or, to borrow A. L. Moore’s word, these scholars argue that Jesus “delimited” the parousia within one generation. [1] 

The three texts given in support of this notion are:

Mark 9:1: “Truly, I say to you, there are some standing here who will not taste death until they see the kingdom of God after it has come with power.”

Mark 13:30: “Truly, I say to you, this generation will not pass away until all these things take place.”

Matthew 10:23: “When they persecute you in one town, flee to the next, for truly, I say to you, you will not have gone through all the towns of Israel before the Son of Man comes.”

Mark 9:1

In her commentary on Mark, Adela Yabro Collins argues that Mark 9:1 refers to the parousia (the Second Coming of Jesus), and thus those to whom the verse is addressed (“some standing here”) are those who will “live until the coming of the Son of Man”. [2] She points to the perfect participle “has come” (ἐληλυθυῖαν) as signifying the full manifestation of God’s kingdom, the second stage after the kingdom has drawn near (1:15). She also notes that the kingdom’s coming in “power” resembles the language of the coming of the Son of Man described in 13:26 (“with great power and glory”). Both of these points are questionable. The use of ἐληλυθυῖαν need only signify the inauguration of the kingdom, not its completed arrival. Jesus elsewhere describes the kingdom having already come (e.g., Matt. 12:28/Luke 11:20), and the early Christians continued this already-not-yet theme (e.g., 1 Cor. 15:25 [the Messiah is already reigning]). And the while Mark does describe the parousia in terms of “power and glory”, even Paul, writing earlier, could use the same kind of kingdom-and-power language to refer to something that has already occured: “and he was marked out as the Son of God in power according to the Spirit of holiness by his resurrection from the dead” (Rom. 1:4). (The same phrase, ἐν δυνάμει, used). The enthronement scene of Psalm 2:7-8 is alluded to here, and Rom. 1:5 (“to bring about the obedience of faith for the sake of his name among all the nations”) echoes the language of the Son inheriting the nations and taking the earth for his possession in Psalm 2:8. For this reason N. T. Wright remarks, 

Thus, though the word “kingdom” is not explicitly mentioned, the way in which this extract… claims the biblical language of kingship for Jesus and declares that this kingship is “with power” and “for the sake of his name,” already indicates an answer to the question of Mark 9:1: yes, the kingdom of God has already come with power, when Jesus was raised from the dead and began to commission his emissaries to summon the nations to allegiance. [3]

(Wright’s point also weakens Bart Ehrman’s argument that Matthew 16:28 and Luke 9:27 omit the phrase “in power” from Mark 9:1 to soften the imminent expectation implied by it). [4]

What does Mark 9:1 refer to, if not the parousia? I think the standard alternative interpretation, which views the Transfiguration as the (or a) fulfillment of the coming of the kingdom, is correct. All the Synoptic Evangelists for this reason place the event immediately after the saying. Even Dale Allison, who argues that saying in its original context referred to an imminent parousia, concedes that the Evangelists have framed it this way. [5] Collins protests that 9:1 has nothing to do with the Transfiguration, because the verse is connected with the parousia described in 8:38, and the time marker in 9:2 (“after six days”) indicates that a new section. [6] Certainly 9:1 is connected with 8:38. But 9:1 follows the focus on the completed arrival of the kingdom in 8:38 with a focus on the foretaste of that future state in 9:1. In other words, there is a progression in Jesus’s words — “The kingdom of God will come in the future,” Jesus says, “and some standing here will be able to experience its inauguration even now.” As Craig Keener comments, 

This verse points to the future glory mentioned in the preceding verses by way of an anticipatory revelation of that glory they are to experience in 9:2-13. Because the future Messiah had already come, the glory of his future kingdom was also already present. [7]

And while Collins is correct that the time marker in 9:2 begins a new scene, by no means does it follow that this new scene can have no thematic continuity with the preceding one. Moreover, Collins herself concedes that 9:1 “is set off from the preceding sayings by the introductory phrase ‘And he said to them’ (καὶ ἔλεγεν αὐτοῖς).” Could this introductory phrase function as a transition between the theme of final judgment in 8:38 and the Transfiguration scene starting in 9:2?

Mark 13:30 

Scholars who see Jesus predicting the end of the world within a generation here have simply misread. The phrase “all these things” does not refer to the parousia (vv. 24-27), only the signs leading up to it (vv. 5-23). 

   

The structure of Mark 13 makes this clear. The disciple’s question in v. 4 pertains to “these things” (πάντα) and “all these things” (ταῦτα… πάντα). [8] This question is followed by Jesus’s answer in vv. 5-23. These verses are held together by an inclusio structure. The πάντα of verse 23 bookends the πάντα of verse 5. The phrase “I have told you all things beforehand” (v. 23) recalls the phrase “Jesus began to say to them” (v. 5). And the exhortation to “be on guard” (v. 23) mirrors the exhortation, “See that no one leads you astray” (the same Greek word is used, βλέπετε). Since vv. 4-23 are bookended by πάντα and ταῦτα… πάντα, it is only natural that ταῦτα πάντα of v. 30 should refer to the material within that section, not to the parousia material in vv. 24-27. [9]

   

Moreover, the parable of the fig tree in v. 29 should put to rest the idea that ταῦτα πάντα in v. 30 includes the parousia itself. The most natural referent of “all these things” in v. 30 is “these things” in v. 29. And in v. 29, Jesus does not say that the occurrence of “these things” means that the Son of Man is here, but that he is near. Indeed, if v. 30 included the parousia itself within “all these things” it would render v. 29 absurd: “when you see all these things (including the parousia) taking place, you know that the parousia is near, at the very gates”! [10]

   

George Beasley-Murray also points out that the saying in v. 30 resembles the saying preserved in Matthew 23:36/Luke 11:51. [11]

Matthew: “Truly, I say to you, all these things will come upon this generation.”

Luke: “Yes, I tell you, it will be required of this generation.” 

In context, both sayings refer to the doom that it to fall upon Israel for its rejection of God’s messengers. As Beasley-Murray comments, “If Mark recognized this meaning he would have specifically related it to the ruin of the temple and all that is bound up with in within the discourse.” [12] That Mark 13:30 is so close to this logion supports the interpretation already suggested by the structure of the discourse.

Lastly, the interpretation of Mark 13:30 I am putting forward removes a tension with v. 32: “But concerning that day or hour, no one knows, not even the angels in heaven, nor the Son, but only the Father.” If one wishes to “harmonize” these two passages by suggesting that Jesus did predict the parousia within his generation without specifying the date more exactly, he is free to do so. But the broadness of time — “day” and “hour” in v. 32 and “time” in v. 33 — fits more naturally if 13:30 only refers to the events of vv. 5-23. Note as well the close relation these verses have with Acts 1:6: “It is not for you to know the times or seasons that the Father has fixed by his own authority.” [13] 

Jesus apparently understood the parousia’s timing to be a matter of God’s providence, and therefore focused more on moral exhortation (Mark 13:33-37) rather than eschatological speculation (cf. Didache 16:1, 2 Clement 12:1 [the time in unknown]; Shepherd of Hermas 9.5.2, Justin Martyr, 1 Apology 45 [it is a matter of God’s providence]). 

Matthew 10:23

Matthew 10:23 was the passage which Albert Schweitzer famously called “the first postponement of the parousia.” [14] Anticipating some scholarly interpretations of the passage today, he saw in it a failed prediction on Jesus’s part. 

On the other hand, some have suggested that the Son of Man’s coming in Matt. 10:23 is not even eschatological. Witherington, for example, suggests that Matthew could mean “that the disciples shall not have completed the missionary work in Israel that the early Jesus sent them out to do before he rejoins them.” [15] In support of this de-eschatologized interpretation he cites Matt. 11:19 (cf. Luke 7:34), which shows “that Jesus is capable of speaking of his present activities using the phrase ‘Son of Man’ and the verb ‘to come’ in its more common sense.” [16] He also points out that Matt 10:23 lacks any mention of the stock parousia imagery — angels, clouds, etc.. 

Another interpretation sees the Son of Man’s coming as fulfilled, or at least partially fulfilled, in the resurrection. This is the view, for example, of Leopold Sabourin. [17] In Matthew 10, Jesus sends out the disciples to preach the good news only to the towns of Israel. “Go nowhere among the Gentiles [ἐθνῶν] and enter no town of the Samaritans,” Jesus says, “but go rather to the lost sheep of the house of Israel” (10:5-6). But this temporary command is exactly reversed at the end of the Gospel:

And Jesus came and said to them, “All authority in heaven and on earth has been given to me. Go therefore and make disciples of all nations [πάντα τὰ ἔθνη], baptizing them in the name of the Father and of the Son and of the Holy Spirit, teaching them to observe all that I have commanded you. (28:18-20)

Jesus now commands his disciples to evangelize to the “nations” (or “Gentiles”); before he told them not to. And Jesus in this passage tells us what accounts for this reversal: “All authority in heaven and on earth has been given to me.” Matthew’s language here is an allusion what is said about the “son of man” in Daniel 7:14. (Compare Matthew’s Ἐδόθη μοι πᾶσα ἐξουσία ἐν οὐρανῷ καὶ ἐπὶ τῆς γῆς to Daniel’s καὶ ἐδόθη αὐτῷ ἐξουσία καὶ πάντα τὰ ἔθνη τῆς γῆς.) Thus Matthew shows his readers that the Son of Man’s “coming” in 26:64 was inaugurated in Jesus’s resurrection.

Yet one more interpretation holds that in Matt. 10:23 Jesus is not cutting short the disciples’ mission but describing a continued mission. A. L. Moore notes the composite structure of the discourse:

1. 10:5-15: The immediate mission of the disciples, limited to the towns of Israel (v. 5). 

2. 10:16-23: A more general mission to the disciples, where a Gentile mission is implied (v. 18). Matthew has transferred most of these sayings from the Olivet Discourse in Mark 13. 

3. 10:26-42: Various sayings of Jesus.

It is also notable that each section ends with ἀμὴν λέγω ὑμῖν (“truly, I say to you”; v. 15, 23, 42). [18]

Moore then points out that the saying in v. 23 must be interpreted within the more widespread mission of vv. 16-23, not the immediate mission of the disciples in vv. 5-15. This frees verse 23 from a having a delimited meaning — “V. 23b is neutral in respect of the duration of the work involved, simply affirming that it will not be completed before the parousia; and if v. 23a is understood in connection with v. 23b and the entire mission charge, this too is undelimited.” [19]

Stephen Witmer similarly argues for this view, noting that a) Matt. 10:16-22 describes an “extended period of time”, with the mention of the disciples before governors and kings, b) Jesus’s command not to go among the Gentiles in 10:5 contradicts his claim that they will bear witness before Gentiles in 10:18, indicating “that 10:16-22 refers to events beyond the immediate trip the twelve will make”; and c) Matthew omits the return of Twelve as found in Mark 6:30. [20] 

Of these options, Sabourin’s view and Moore’s view are best. I do see an eschatological meaning in the “coming” of the Son of Man here, contra Witherington’s suggestion. Sabourin’s view requires that vv. 5-23 be intended to portray one extended mission, limited to the towns of Israel, and that Matthew overlooked the discrepancy between v. 5 and v. 18. This, of course, is possible. But Moore’s view irons out this difficulty. 

In any case, Schweitzer’s insistence that the saying of Jesus in Matt. 10:23 is a false prophecy and represents the “first postponement of the parousia” is weak. 

Notes 

[1] A. L. Moore, The Parousia in the New Testament. (Leiden: E. J. Brill, 1966). 

[2] Adela Yarbro Collins, Mark: A Commentary. Ed. Harold W. Attridge. (Minneapolis: Fortress Press, 2007) 412-413 (quotation from 413). 

[3] Wright, N. T. “Hope Deferred? Against the Dogma of Delay.” Early Christianity 9 (2018): 58-59. 

[4] Bart D. Ehrman, Jesus: Apocalyptic Prophet of the New Millenium. (New York: Oxford University Press, 1999) 130, calling this difference “huge”. 

[5] Dale C. Allison, Jesus of Nazareth: Millenarian Prophet. (Minneapolis: Fortress Press, 1998) 168. 

[6] Collins, Mark, 412. 

[7] Craig Keener, The IVP Bible Background Commentary: New Testament. Second Edition. (Downers Grove, IL: Baker Academic, 2014) 149. 

[8] It is not clear what to make of Mark’s two questions. Matthew more clearly separates the two, with the first pertaining to the temple destruction and the second pertaining to the end of the age (24:3). Robert Stein, Jesus, the Temple and the Coming Son of Man. (Downers Grove, IL: IVP Academic, 2014) 64-69, argues that both questions in Mark refer to the temple destruction, and cites 11:28 as a parallel. Edward Adams, “The Coming of the Son of Man in Mark’s Gospel.” Tyndale Bulletin 56.2 (2005): 55, argues that the second question in Mark alludes to Daniel 12:6-7 and therefore refers to the final consummation.

[9] For this the argument for this inclusio, see Stein, Jesus, 67, 72. Cf. Ben Witherington III, Jesus, Paul and End of the World: A Comparative Study in New Testament Eschatology (Downers Grove, IL: InterVarsity Press, 1992) 42-43, who similarly argues that the phrase “these things” characterizes vv. 5-23, as opposed to “those days”, which characterizes the parousia material. Therefore, he argues, Mark 13:30 (when you see “these things”) refers to material in vv. 5-23. See also C. E. B. Cranfield, “St. Mark 13.” Scottish Journal of Theology 6.3 (1953): 291.

[10] Stein, Jesus, 123-124. 

[11] George R. Beasley-Murray, Jesus and the Last Days: The Interpretation of the Olivet Discourse.  (Vancouver, British Columbia: Regent College Publishing, 1993) 447-448. 

[12] Ibid., 448. 

[13] Thanks to Jason Engwer for making this connection.

[14] Albert Schweizter, “The Solution of Thoroughgoing Eschatology.” Pages 6-49 in. The Historical Jesus in Recent Research, ed. James D. G. Dunn and Scot McKnight. (Winona Lake, Indiana: Eisenbrauns, 2005). 

[15] Witherington, End of the World, 40-41 (quotation from 41). 

[16] Ibid., 41. 

[17] Leopold Sabourin, “‘You Will Not Have Gone Through All the Towns of Israel, Before the Son of Man Comes’ (Mat 10:23b).” Biblical Theology Bulletin 7.1 (1977): 5-11. 

[18] Moore, Parousia, 143-144, see esp. n. 3. 

[19] Ibid., 145. 

[20] Stephen Witmer, “Keeping Eschatology and Ethics Together:

The Teaching of Jesus, the Work of Albert Schweitzer, and the Task of Evangelical Pastor-Theologians.” Themelios 39.3 (2014): 494.