Is Martha an Interpolation into John’s Gospel?

One of the readers of our blog, and a very promising PhD student at Duke University, Elizabeth Schrader, is currently working on examining the textual transmission of John 11:1–12:2, specifically the presence (or absence) of the two sisters, Martha and Mary in the story. I have invited Elisabeth to share her research in three consecutive posts. Based on her observations in the manuscripts, she proposes the bold thesis that Martha was interpolated into the Fourth Gospel in the second century. I know she looks forward to response and debate. Personally, I think her findings are very significant, although I disagree with her overall explanation of the data.

---

A Problem around Martha: Introduction

In a 2017 article published in the Harvard Theological Review (Open Access version here), I argued that the character Martha is likely to be a second-century interpolation into John’s Gospel. In recent weeks, this research has received increased attention and discussion (Duke Today, Religion News Service, Religion for Breakfast). Tommy Wasserman has kindly invited me to introduce my work to this blog, and I was very happy to discuss it in more detail here.

In my research, I have demonstrated that Martha’s presence is consistently unstable throughout the textual transmission of John. Such instability is found in nearly every verse where Martha appears in John 11:1–12:2, in witnesses as early as Papyrus 66 and as late as the 1611 King James Bible. My conclusion is that approximately one in five Greek manuscripts and one in three Old Latin manuscripts has some problem around Martha. I define a “problem around Martha” according to the following five criteria:

1. the unexpected omission of Martha’s name
2. the initially transcribed name “Mary” altered to “Martha”
3. the name “Mary” appearing instead of an expected “Martha”
4. an unexpected singular noun, verb, or pronoun to describe the Bethany sisters
5. a different person named as the first of those Jesus loved in John 11:5

My most recent compilation of the relevant manuscript and patristic data is available here for those who are interested; I will happily approve a read-only view of the spreadsheet for anyone that asks.

Of course we must now ask, why is there such a remarkable textual problem around Martha in John’s Gospel? My position is that Martha was added to John’s Gospel in order to discourage readers from identifying Lazarus’ sister Mary as Mary Magdalene, an identification that was happening in patristic and extracanonical literature as early as the third century. Martha’s presence in the Lazarus story is important: she is now the woman who speaks the central Christological confession “Yes, Lord, I believe that you are the Messiah, the Son of God, the one coming into the world” [John 11:27, NRSV]. This confession is often compared to Peter’s similar Christological confession in Matthew (“You are the Messiah, the Son of the living God” [Matt 16:16, NRSV]) to which Jesus replies that Peter is the rock upon which the church will be built [Matthew 16:18]. Since according to Matthew, the Christological confessor is designated as the foundation of the church, the identity of the Christological confessor in John’s Gospel is of paramount importance. It should give us pause to learn that Tertullian, writing in 206 AD, believed that Mary gave this confession – and that this Mary was understood by many early readers to also be Mary Magdalene, the first person to see the risen Jesus and to receive an apostolic commission in John’s Gospel. Indeed, if it was the Evangelist’s intent for Mary (Magdalene) to give the Christological confession in John 11:27, this would suggest that her leadership role in John was akin to Peter’s in the Gospel of Matthew.

One need not give authority to the Gospel of Thomas, the Gospel of Mary, the Gospel of Philip, or the Pistis Sophia today to notice that they offer evidence of early objections to Mary’s stature as “apostle to the apostles” (her status among Orthodox Christians even now). Composed by diverse Christian writers over the course of more than a century, these four documents each portray Jesus’ disciples—particularly Peter—objecting to Mary or to special status given to her by Jesus. Perhaps the changes we see from “Mary” to “Martha” are not so different from whatever process led to a textual variant at Luke 2:33, where Ἰωσὴφ is often copied instead of ὁ πατὴρ αὐτοῦ. We know that there were early debates about the doctrine of Jesus’ virgin birth. So, when we see textual instability around the suggestion that Joseph was Jesus’ “father,” it is reasonable for textual critics to infer that this textual problem could be connected with the virgin birth debate. Similarly, we know that there was early controversy about Mary Magdalene’s authority, especially vis-à-vis Peter. So when we see that the presence of Martha, the Christological confessor, is unstable throughout John 11–12—and since we know that many early readers identified Lazarus’ sister Mary as Mary Magdalene—our scholarly antennae should go up.

Announcing the Sacred Words Conference

On February 21–22, 2020, plan to attend the Phoenix Seminary Text & Canon Institute’s inaugural history of the Bible conference: Sacred Words: History of the Bible Conference.

About

Brief description from the conference website:

The Bible is the best-selling book of all time and its influence on Western culture beyond compare. But how did this group of ancient books written and then copied over millennia become the Bible we now know?

Join us for the Text & Canon Institute’s inaugural conference and learn from internationally known speakers about how the Bible has been copied, collected, and confessed as God’s sacred words.

The Text & Canon Institute seeks to foster academic study of the Bible and to serve the church and impact the culture by providing expertise and instruction on the history of the Bible. The Sacred Words conference brings this expertise directly to the church and wider culture.

More Info & Tickets

Speakers

We are delighted to announce that Daniel Wallace, Peter Gentry, and Stephen Dempster will be the plenary speakers for the event. We’re also very excited to have some top-notch breakout presenters from the region: Jeff Cate, Anthony Ferguson, Darian Lockett, and Timothy Mitchell. Each of these speakers and presenters will be speaking on the topics in their own wheelhouse and each has a desire to see the church and honest seekers learn more about the history of the Bible.

Schedule

Friday, Feb. 21

Arrive/Registration (6:00 pm) Welcome (6:30 pm)
Plenary: Dr. Peter Gentry – Chaos Theory and the Text of the Old Testament
Breakouts:

• Dr. Jeff Cate – The Stories They Tell: A Look at Interesting Variants in the Transmission of the New Testament
• Dr. Darian Lockett – The Catholic Epistles: What Do James, Peter, John, and Jude have in Common?
• Dr. Anthony Ferguson – Listening to the Dead Sea Scrolls
• Timothy N. Mitchell – Where Inspiration is Found: Putting the New Testament Autographs in Context

Dismiss (9:00 pm)

Saturday, Feb. 22

Coffee and pastries (8:30 am)
Opening remarks (9:00 am)
Plenary: Dr. Stephen Dempster – How the Bible Became the Bible
Break
Plenary: Dr. Daniel B. Wallace – Is What We Have Now What They Wrote Then?
Panel discussion
Dismiss (12:00 pm)

If you are in or around the Phoenix area next February, we hope you will join us for what will undoubtedly be some of the best teaching on the history of the Bible in the USA in 2020 and maybe for a long time to come.

The Purpose of Catchwords

In Medieval manuscripts it’s not uncommon to find the first word of each page written at the bottom of the previous page. These repeated words are called “catchwords” and they continue well into the age of print. My vague sense is that they stopped being used around the 18th/19th cent. in printed works.

Catchwords in Erasmus’s 1516 edition

I had always assumed that these catchwords were a type of reader’s aid. I assumed that the idea was that you can have the next word, in whole or in part, at the ready as you turn the page. In this, it’s a bit like a pianist who needs to know what the notes are on the next page so that the music doesn’t stop during the page turn. This explanation, however, doesn’t make  sense of why these catchwords are found on the verso where no page turn is needed. But I just assumed they were included there since the eye has to move from the bottom to the top of a page.

But Michelle Brown has cleared up the reason for me. She says that catchwords are actually a printer’s aid not a reader’s (see Understanding Illuminated Manuscripts, p. 36). They are meant to help facilitate the arrangement of quires during book binding. Brown says they were introduced in Europe by way of Spain, Italy and France around 1,000 and that they may have entered through Islamic influence.

So, there you have it. I’d still like to know more about how they developed and why they became obsolete in printed books. If anybody knows more about the history of this little device, let me know.

The Textual Optimism of the UBS4

The grading system for variants in the UBS editions is probably its most distinctive feature. These ratings rank from A to D and indicate the relative level of certainty the editors felt in their decision (A being the most confident). I suspect this system has been a help to many translators over the years. For myself, the system is a rare and welcome peek behind the editorial curtain to one of our most important (and bestselling?) editions.

Over time, however, there is a noticeable and well-documented shift in these ratings without any additional explanation or justification. For example, the reading of Eph 5.22 is given a C in the UBS3 and a B in UBS4, but the textual commentary for this decision remains word-for-word the same. Kent Clarke called this “textual optimism” and his work on this is well worth consulting. (See the helpful summary from Mark Ward here.)

What I didn’t realize until today is that several of the editors owned up to this “textual optimism” and wrote about it even before the UBS4 was in print. Kurt and Barbara Aland say as much in their Text of the New Testament. On p. 45, they write about the UBS3 that “The only question is whether the editors have not been too cautious in applying the classifications, so that a B should often be replaced by an A, a C by a B, and a D by a C (a thorough reexamination has led to a revision of these for the fourth edition of GNT).” It does lead one to wonder how much responsibility for the increased textual optimism is the responsibility of the Alands.

Asa or Asaph in Matthew 1: A Teaser for the THGNT Textual Commentary

Yesterday morning, I was reading a sermon by Charles Haddon Spurgeon (1834–1892) titled “Concealing the Words of God.” No, the sermon is not about textual criticism, but Spurgeon made a remark that sent my mind back to the arguments surrounding Asa or Asaph in Matt. 1:7–8. Spurgeon said (emphasis mine),

Lay no embargo upon any form of truth; demand no toll for the commodities of heaven. Let your mind be an open port, carrying on a free trade in the treasures of the gospel. Believe whatever God says, because God says it, though you may not always see its why and wherefore or perceive its internal consistency. Be prepared, and even anxious, to know the whole truth as far as you can know it, and let it pervade your entire being with its holy influence. It will be a terrible thing if one of these days you shall have to say “I rejected a great truth. I had a suspicion that it was so, but I did not wish to believe it, and so I shut my ears to its evidence. I had a leaning towards the opposite view, and I felt committed to it, and so refused to alter.” —Metropolitan Tabernacle Pulpit 25 (1879): 245.

Of course, Spurgeon is talking about theology here, but his remarks have relevance to arguments I’ve come across regarding a textual variant in Matt. 1. Since Spurgeon was not opposed to textual criticism, I’ll use his remark to leap into the Asa/Asaph variant at Matt. 1:7–8. I derive parts of what follows from the forthcoming Textual Commentary for the Tyndale House Greek New Testament that Dirk Jongkind and I are working on.

The variant in question is at Matt. 1:7–8. There are two instances of a name in Matthew’s genealogy—Ἀσά or Ἀσάφ, right at the end of v. 7 and beginning of v. 8. A sampling of the textual evidence breaks down thus, derived mostly from the UBS5 (with languages separated by commas):

Ἀσάφ· Ἀσὰφ] P1(^vid ασ̣α̣[φ ασα]φ̣) ℵ B C 700 f¹ f¹³, aur c g1 k q vg^mss, sa bo mae, arm, eth, geo, (sy^hmg) 

Ἀσά· Ἀσὰ] K L W Δ Σ 33 565 1241 892 1424 Byz, (a), f ff1 vg, sy^c sy^s sy^p sy^h sy^pal, slav

The Tyndale House edition, UBS, NA and SBL editions all go with Ἀσάφ. The earliest manuscripts support Ἀσάφ, as well as much of the Old Latin tradition, all of the Coptic tradition, and most of the ‘minor’ versions. On the other hand, the majority of Greek, Latin and Syriac manuscripts support Ἀσά, as it is supported by the Byzantine Greek text and the Vulgate, as well as some Old Latin and most Syriac versions.

Textually, if one’s preference is for early manuscripts, this decision is easy. There is a clear ‘winner’ among early witnesses in the Greek tradition. Still, the explanation for the variation has caused concern for some. The issue is that in 1 Kings 15:9, this individual is named Asa, not Asaph. Copyists were not unaware of this difference—in GA 1582 (which is close to the archetype of fam. 1), the text has Ἀσάφ, but a marginal note mentions that “Ασα” is the name according to 1 Kings (see Amy S. Anderson, The Textual Tradition of the Gospels: Family 1 in Matthew. NTTS 32. [Leiden: Brill, 2004], 62.)

The question, then, is whether copyists changed Ἀσά to Ἀσάφ, presumably because of the familiarity with the Psalmist, or did copyists see Ἀσάφ, perceive an error and ‘correct’ it, writing Ἀσά and conforming the text to 1 Kings?

Metzger writes that Ἀσάφ likely came about because “the evangelist may have derived material for the genealogy, not from the Old Testament directly, but from subsequent genealogical lists, in which the erroneous spelling occurred” (Textual Commentary, 2nd ed. [Stuttgart: German Bible Society, 1994], 1). If one believes that the original text is inerrant, one would rightly be suspicious of Metzger’s explanation here. That being said, one must not fall into the fallacy of a false dilemma and think that one must accept Metzger’s reason in order to accept the same text as Metzger.

In the current draft of the forthcoming Textual Commentary for the THGNT, we write the following to support the decision to print Ἀσάφ:

The text is adopted on the basis of its early evidence. The king’s name is given as Ἀσά thoughout 1 Kings 15, which could have led scribes to think the reading Ἀσάφ is an error, even as some modern scholars have suggested.[1] The similarity of the names Ἀσάφ (also note the Psalmist of the same name) and Ἀσά could have led to confusion as well.  

The name used in the Hebrew text of 1 Kings 15 is אָסָא (Asa), and a legitimate interpretation of this name is that it is a hypocorism (a shortened name)—an alternative form of the same name,אָסָף  (Asaph).[2] An alternative explanation for the use of Ἀσάφ might be that Matthew knows Hebrew better than his later copyists, and for the king who is called Asa in 1 Kings 15, Matthew uses the longer form of his name, Asaph.

---

[1] For example, Bruce M. Metzger suggests that Matthew may have copied Ἀσάφ from “genealogical lists, in which the erroneous spelling occurred,” in Bruce M. Metzger, A Textual Commentary on the Greek New Testament, 2nd ed. (Stuttgart: Deutsche Biblegesellschaft, 1994), 1. W. D. Davies and Dale C. Allison also suggest that Ἀσάφ is an error, corrected by later scribes in The Gospel According to Saint Matthew: Volume 1, International Critical Commentary (Edinburgh: T. & T. Clark, 1988), 175.

[2] On אָסָא as a hypocorism (a shortened form) of אסף and the divine name but אָסָף as a hypocorism of the divine name and אסף, see Wilhelm Gesenius, Hebräisches und Aramäisches Handwörterbuch über das Alte Testament: Gesamtausgabe, ed. Herbert Donner, 18th ed. (Heidelberg: Springer, 2013), 83, 85. We are indebted to Pete Myers for this reference.

If one wanted to push the evidence a bit further, one might conclude that having the right text (Ἀσάφ) supports traditional views about Matthean authorship better than the majority reading, Ἀσά. If the author of Matthew’s Gospel was a Hebrew-speaking Christian, wouldn’t it make sense that he knew how Hebrew names worked better than his Greek- and Latin-speaking copyists centuries later?

As Spurgeon said, Believe whatever God says, because God says it, though you may not always see its why and wherefore or perceive its internal consistency. Our priority should be for the text and what it is, trusting that it is internally consistent even if we do not always perceive it so. When I was discussing this passage with a friend, he said something helpful to me once we found the explanation of hypocorism. “The thing you’ve got to remember,” he said, “is that the text [i.e. Asaph] was already perfect and without error long before we found an explanation for why it is perfect and without error.”