Some Notes on the Earliest Manuscript of Paul’s Letters

The publication of P46 in 1935–37––then, and now, the oldest extant manuscript of Paul’s epistles––has not ceased to pique the interest of biblical scholars. Beginning with the plates and text published by Kenyon (1936, 1937), and continuing with the virtual birth of reasoned eclecticism with Zuntz’s magisterial The Text of the Epistles: A Disquisition on the Corpus Paulinum (1953), and reconsiderations of its date (Young Kyu Kim, “Palaeographical Dating of P46 to the Later First Century,” Biblica 69 [1988] 248–57), this priceless document has made its way to the front lines of biblical scholarship for a long time. Though Kim’s suggestion that Chester Beatty Biblical Papyrus II was written before the reign of Domitian (81–96 CE) has been refuted, the consensus continues that it was produced c. 200 CE.

Where Are the Pastoral Epistles?
One curiosity of this papyrus is that, in its current state, it lacks the pastoral letters. With 86 of the original 104 leaves still extant, scholars have a good amount of material to work with as to whether it would have originally contained the pastorals.

How do they know that it originally contained 104 leaves? Two features are used to infer this. First, the manuscript is a single quire codex. This means that all the double leaves (or bifolia) were laid down on top of each other, then folded and sewn into the binding. The fact that it was a single-quire codex can be detected by size of the pages: moving from the beginning to leaf 52, the pages get increasingly narrower. Then, from leaf 53 to the end, the pages get increasingly wider. This can only mean that all the bifolia were laid down in one stack, folded, then trimmed on the outside so that all the leaves were relatively flush. (The later, standard quire was eight leaves [or four double leaves], used throughout the early to middle ages.)

Second, each page is numbered by the original scribe. This is unusual for manuscripts of any age. Usually the quires are numbered, and frequently the leaves were numbered (on the front page), but not the pages. Since the first extant page is numbered 17 (folio 8), and starts with Romans 5.17, we can extrapolate that the manuscript began with Romans 1 and is missing the outer seven double leaves.

What is extant are nine ‘Pauline’ letters: Romans, Hebrews (almost always included in the Pauline corpus as far as ancient manuscripts are concerned), 1 Corinthians, 2 Corinthians, Ephesians, Galatians, Philippians, Colossians, and 1 Thessalonians. Missing are 2 Thessalonians, Philemon, and the pastoral letters.

Almost surely 2 Thessalonians followed 1 Thessalonians, as it does in all other manuscripts which have both letters. But what about Philemon? Most scholars are of the opinion that Philemon was also included in this codex.

But the pastorals? This is where opinions have varied. Kenyon started the ball rolling with the suggestion that the last five leaves were left blank and did not include the pastorals. The argument is that there was simply not enough room to include the pastorals and Philemon, which would require about twice as many leaves. Why would this be the case? Was the scribe unaware of them? If he was, this might suggest that they are not authentic.

Although Kenyon’s view held sway for many decades, and is even now occasionally affirmed, Jeremy Duff’s article in New Testament Studies 44 (1998), “P46 and the Pastorals: A Misleading Consensus?” challenged this hypothesis.

One of the points made by Duff is that the letters are more compressed at the end of the codex than in the middle. He noted that there are half again as many letters per page in the last leaves than in the middle leaves. But this is partially due to the fact that the outer leaves are wider than the inner leaves. Nevertheless, there are more letters in the back outer leaves than the front outer leaves, showing that at least some compression did take place. And this seems to suggest that the scribe was aware of the problem he had created for including the pastorals and he began to compensate upon realizing his mistake.

There are problems with Duff’s analysis, however. The scribe seemed to understand to some extent how much space it would require to produce his codex. This is seen in his notes at the end of each book of how many lines he wrote. This was customary for professional scribes; it was in essence a bill for services, since they were paid by the line. He may have been working from a manuscript that had already indicated the number of lines, and thus would have known how many leaves he needed for his manuscript. Of course, the fact that his words were more compressed at the end of the codex seems to show that as careful as he was in his calligraphy (and he was), this tells us nothing about his math skills.

A Fresh Examination of the Codex
This week I had the privilege of examining P46 in the flesh. Fifty-six of the 86 leaves are housed at the Chester Beatty Library in Dublin. I have spent the last three days examining them in some detail, comparing them to the editio princeps by Kenyon. Sir Frederic Kenyon produced the plates of this famous papyrus in 1937. Each image has been assumed to be exactly the same size as the actual leaf it represents. I measured each leaf against Kenyon’s plates and noticed some interesting phenomena.

Of the 88 plates (44 leaves) I was able to measure, the plates in Kenyon’s volume are off at least 61 times. Most of these are within 1 mm or so, but a few are fairly significant. At least three are off by 3 mm, and one is off by as much as 5 mm.

In addition, I noticed a curiosity: some of the plates in Kenyon’s volume were photographed without the lens plane exactly parallel to the leaf. For example, the photograph of folio 39 verso (1 Cor 1.14–23) is slightly wider than the actual leaf on the bottom, while the top of the leaf is the same size in both the photograph and actual manuscript. And leaves 49 recto (1 Cor 10.1–10), 50 recto (1 Cor 10.21–30), and 51 verso (1 Cor 10.31–11.6) are also somewhat trapezoidal. This suggests that the camera got cocked, or the tape around the plates (for Kenyon’s photographs were of the papyrus under glass) got doubled up, creating the trapezoidal effect.

How much the incorrect sizes of the leaves in Kenyon’s volume impact the discussion of whether P46 originally contained the pastorals is yet to be seen. But in the least, this anomaly needs to be factored into the discussion. Much can still be learned from this renowned papyrus.


10 thoughts on “Some Notes on the Earliest Manuscript of Paul’s Letters

  1. Reblogged this on beliefspeak2 and commented:
    Papyri is ancient Egyptian paper that could take over a year to produce. It was cheaper than animal hides but would not generally last as long. It was exported from Egypt to lands in Europe and the MidEast. It served as media for the copied writings of Scripture in Egypt and elsewhere. The manuscripts that were stored in the desert regions of North Africa could survive thousands of years due to the dry climate.

    In 1976 I was able to examine the Bodmer Papyri in Switzerland traveling to its repository while in Europe as an apprentice church worker. Textual transmission and its study has fascinated me for 40 years. Dan Wallace, a leading textual scholar, has just examined the Chester Beatty p46 manuscript and reports his findings on his blog.


  2. Thank you for making the lessons from this manuscript understandable and accessible. I am often talking about how the transmission of the New Testament text gives us assurance that we have the original message of the New Testament firmly in hand. There is no chance of some “Satanic verses” being attributed to some lost copy of the New Testament. I was witnessing to an individual this last week who through out the objection that the message of the New Testament got distorted along the way and was able to quickly diffuse that argument. Non-scholarly people are really challenging our text on this basis and it is so helpful to be able to set the record straight. Thanks again for the work you are doing to preserve and even find more manuscripts.


  3. Chance

    How cool that you have had such great exposure to P46, Dan. One problem I had when I wrote my paper on P46 was the differing opinions on the scribe behind P46. Zuntz is on one end and Eldon Jay Epp is on the other end. What do you think about the scribe of P46?


    1. Actually, I think they’re compatible. The scribe of P46 was well trained in calligraphy. But that does not mean that he (or she) was especially careful in what he/she wrote as far as the text is concerned. Thus, both Epp and Zuntz can be right.


  4. Pingback: 104. Papyrus 46 | Bible differences

  5. Scott

    Dr. Wallace,

    Do you think it is possible that P46 preserves the original location of the doxology traditionally located at Romans 16:25-27?

    I find numerous verbal echoes of chapter 15 in the doxology:

    “Now we who are strong ought to bear the infirmities of the weak” (Romans 15:1)
    “Now to the one who is able to strengthen you . . .” (Romans 16:25)

    “For whatever was written aforetime was written for our instruction, so that through the encouragement of the scriptures we might have hope” (Romans 15:4)
    “. . . through the prophetic scriptures written at the command of the eternal God” (Romans 16:28)

    “I make it my ambition to preach the gospel where Christ has not already been named, lest I build on someone else’s foundation” (Romans 15:20)
    “. . . according to my gospel . . . which was kept secret for long ages” (Romans 16:25).

    “for the obedience of the Gentiles” (Romans 15:18)
    “to all the Gentiles for the obedience of faith (Romans 16:26)

    “glorify God the Father” (Romans 15:6), “the glory of God,” (Romans 15:7), “glorify God for his mercy” (Romans 15:9)
    “to the only wise God . . . be glory forever and ever” (Romans 16:27).

    I always thought that the doxology originally belonged at the end of chapter 15 on the basis of these verbal parallels. If the doxology is placed at the end of chapter 14 or chapter 16, it has nothing to do with the immediately preceding context, which would be unusual for Paul.


  6. Wendell R. Gideon

    On the page of P46 containing Second Corinthians 12:1-7 there is one line (comprising most of verse 7) which contains only twenty-seven (27) letters whereas the other lines average around thirty-one (31) letters) on this leaf. In my Nestle-Aland Greek New Testament (27th ed.) the word “dio” appears in this verse but there is a siglia indicating it should be deleted because P46 does not contain such a word. My examination reveals that indeed no such word (“dio”) is present in this line on this page of P46. The Church Father “D” agrees that “dio” should be deleted. So where is this Greek text containing “dio” if it is not P46? And can you explain if this means that there was a text older than P46 which contained such a word? Now, if “dio” is added then this makes this line more in line with the others as far as the number of letters is concerned. And if indeed something is missing but it is not “dio” what might these letters have been? I have come up with my own theory: if we add a couple of letters which are very similar to “dio” and combine them with “ina” (the word following “dio” we discover we have a word which changes the whole way these verses have been interpreted for over 1900 years. So, if “D” can delete “dio” centuries after Paul wrote what he did (assuming he wrote or dictated such a word), how come we cannot “add” to it in the 21st century? You see, the problem we have here is this: we still do not have the BEST EVIDENCE, THE ORIGINALS. We are stuck with what appears to be a copy of a copy of a copy, etc. So I ask: Is this the best that the Lord God Almighty can do or is this the best that we errant mortals can do? Your thoughts are appreciated.


Comments are closed.