Scraps of Papyrus: Earliest Gospel Texts?


These three scraps of papyrus right here contain text from Matthew 26. They were presented by the Reverend Charles Bousfield Huleatt to Magdalen College, Oxford in 1901, hence their name, the ‘Magdalen papyrus’. We don’t know exactly how Huleatt acquired these scraps or their ultimate origin, only that he acquired them while he was in Egypt (which was where quite a lot of papyrus manuscripts were found).

The Magdalen papyri (aka Papyrus 64) are complemented by another scrap of papyrus now currently in Barcelona, Spain (Papyrus 67), which contains text from Matthew 3 and 5. Most scholars now agree that Papyrus 64 and Papyrus 67 were written by the same scribe and come from the same codex (we know that it was a codex rather than a scroll since it was written on both sides.), so for all intents and purposes, we’ll consider both here together.

Now there is a third fragment, this time of Luke (Papyrus 4, currently in the Bibliothèque nationale de France), which shows a similar handwriting style as the scribe of the Matthew fragments, but scholars are divided as to whether they belong to the same codex as P64-67 (supposing that they were really written by the same scribe). Unlike the Magdalen papyri, we have a more concrete provenance for Papyrus 4: the fragment was used as stuffing for the binding of a late 3rd-century codex of Philo and found in a jar walled up in a house in Coptos (Qift), Egypt.

Papyrus 4.

An earlier dating of the Gospels?

When Huleatt donated P64 to Magdalen College, he tentatively assigned them a date of 3rd century. The papyrologist A. S. Hunt dated it even later, to the early 4th century. But Hunt’s conclusions were soon overturned by other papyrologists (including T.C. Skeat), who felt that he assigned too late a date to it. P64 was instead redated to the late 2nd-early 3rd century (ca. AD 200s), which has remained the general accepted date.

By itself, Papyrus 64-67 is really not that exceptional when compared to other NT manuscripts in terms of date. (The oldest fragment of Matthew is usually considered to be the late 2nd-century Papyrus 104 from Oxyrhynchus in Egypt. P64-67 ranks among some of the earliest copies of Matthew we have, but not the earliest.) Maybe the only feature that would set it apart is that it was possibly written in two columns (Papyrus 4 had two columns per page), which is a rarity among early NT manuscripts. (There was usually only one column per page - kind of like modern books.) If P64-67 and P4 actually come from the same manuscript as some have claimed, they could represent one of the earliest instances of a papyrus containing more than one gospel. (Our earliest manuscript for that is the late 2nd century Papyrus 75, which contains Luke and John).

Papyrus 104, the earliest Matthew fragment we have

What made P64-67 experience a surge in popularity though is the claims of the late scholar Carsten Peter Thiede. In 1994, Thiede claimed that P64-67 is much earlier than it is commonly thought to be: it was actually from the mid-to-late 1st century AD, which would make it the earliest NT manuscript we have. Thiede’s claims were picked up by the press, and soon he had a popular book written about it (Eyewitness to Jesus: Amazing New Manuscript Evidence about the Origin of the Gospels, republished as The Jesus Papyrus: The Most Sensational Evidence on the Origin of the Gospel Since the Discovery of the Dead Sea Scrolls).

What is interesting though is that while the press reports seem to imply that Thiede had dated the Magdalen papyrus to the ‘mid-1st century’ (AD 60s and whereabouts), what he actually says in his scholarly paper (Papyrus Magdalen Greek 17 (Gregory-Aland P64): A Reappraisal) is slightly different. Although Thiede is sympathetic to the possibility that the manuscript might have been written prior to AD 70, he repeatedly refers more generally to a more vague "first-century date’, or “prior to the turn of the century,” finally concluding:

The fragments of Matthew’s gospel in the Old Library of Magdalen College Oxford, henceforth to be listed as Magdalen Greek 17 rather than 18, remain the oldest extant papyrus of that gospel; but it may be argued that it could be redated from the late second to the late first century, some time after the destruction of the Temple in Jerusalem.

This is confirmed by a correspondence Thiede had with Peter M. Head of Tyndale House in Cambridge, who eventually wrote an article refuting his claims:

I do not give a precise date, but suggest a date in the last third of the first century: The ‘starting point’ is the middle of the century; I allow for a variation of c. 20 years + / - and then opt for the later end, ‘soon after A.D. 70’.


What made Thiede arrive at his mid-to-late 1st century date for Papyrus 64-67? (He didn’t believe Papyrus 4 is related to the Matthew fragments.) Essentially, he compared the handwriting of the manuscript with that in a few other Greek manuscripts which date from the 1st century BC or AD and which from the Judaean Desert - Qumran, Nahal Hever - and in Herculaneum (Villa of the Papyri). Thiede claimed to have seen similarities between the shapes of individual letters between P64 and these Greek manuscripts, which to him points to a possible 1st-century origin for P64-67. He argues that “the date commonly given to Magdalen Gr.17 (and P.Barc.1), ca. 200, may look like a safe “dumping ground”, but this might be too late.”

So, how do we actually determine the date manuscripts?

The easiest way, of course, is to check where these manuscripts were found. Take for example, the Villa of the Papyri in Herculaneum. We know that Herculaneum, like Pompeii, were destroyed when Mount Vesuvius erupted in A 79. So that would provide us with the terminus ante quem (‘limit before which’, the latest possible date) for the carbonized papyrus manuscripts found in Herculaneum: AD 79. Which would mean that most of the papyri would date before that period.

But, as is often the case with many NT manuscripts, either the place where they were found contains little or nothing which could provide us with a date, or (as was often the case) they were not found on-site in archaeological digs; instead they were purchased from dealers in antiquities (yes, shady, black-market-type people) or passed through many hands before they landed up in a museum or a library somewhere. (And, as was often the case in the 19th century, when archaeology was still relatively a young science, people who dug for artifacts could be sloppy by today’s standards.) In other words, we have no clue exactly where these manuscripts came from.

Very rarely, some scribes took the trouble to give some information which could furnish us a date (in the form of colophons), but this isn’t the case with most NT manuscripts, particularly the earlier ones. So what do we do?

There’s carbon-dating, of course, but that’s actually really rarely used, since it can be and is imprecise (a C-14 test might tell you the approximate age of the papyrus, but not necessarily when the text was written). Not to mention that it’s very impractical, since especially for older tests, it would usually involve destroying a part of the manuscript.

What’s really used for dating NT manuscripts is handwriting analysis - paleography (the study of the writing of manuscripts). It would involve, among other things, (1) comparing the handwriting of undated manuscripts with those which can at least be dated (using the latter as a reference point), and (2) studying the development and evolution of handwriting styles (aka scribal hands) and the way letters are formed over time. (Individual scribes of course had their personal flair, not to mention differences in level of skill, but generally speaking, different styles of handwriting go in and out of fashion as time goes on - which is a great help to paleographers.) Now paleography is not just about manuscript dating; paleographers concern themselves also where the manuscript was written, the scribe who wrote it, etc.

Of course, paleography is not an exact science; all its judgments are just approximate. ‘Book hands’ (the type of formal handwriting used for books and other formal/official documents) are more easier to date than casual hands - since as I just mentioned, different styles went in and out of fashion - but even then, most scribes will stick with a particular style as long as they live. Since the average working life period of a scribe would be around fifty years, even if we can accurately date a handwriting style to say, AD 125, it would not necessarily mean that a manuscript written in that form of handwriting literally dates from AD 125. For all we know, a scribe using that style could still have been working forty or fifty years later. Not to mention, some old handwriting styles could be preserved long after new ones have evolved or were invented. That’s why a good rule of thumb is, whenever one sees a date given for early NT manuscripts, remember that it’s not the exact date - it’s an approximation, give or take fifty years.

Now to get back on topic: were Thiede’s identifications really correct? That’s for the next installment.


In his paper, Thiede suggested that the late 2nd-3rd century dating assigned to P64-67 in the 1950s by papyrologist C.H. Roberts needed further revision in light of new manuscript finds: “Since the publication of Roberts’ paper An Early Papyrus of the First Gospel, 1953], new papyri have become available, and they appear to favour an even earlier date. This may not come as a surprise, since one tendency of the reevaluation of NT papyri at least since the [19]60s has been a redating with, occasionally, somewhat drastic and not undisputed consequences.”

All the manuscripts Thiede invokes come with archaeologically fixed final dates (termini ante quem). The first of these is a scroll containing the text of the Minor Prophets from Nahal Hever in the Judaean Desert (8HevXIIgr), dated by most scholars between 50 BC and AD 50.

Even at first glance and using, as a point of reference, the plate in Schmidt/Thiel/Hanhart, the identity and near-identity of several letters is striking: Alpha (A), Epsilon (a letter fluctuating in both scripts) (E), Iota (I), Omicron (O), Rho § and Ny (N) are particularly close. An equally obvious difference, on the other hand, may be seen in the Etas (H) and Mys (M); but the second scribe of the Nahal Hever scroll provides the comparable Eta and My more than once.

Now the question is: are the letters really as similar as Thiede claimed?

Letters from P64 compared to letters in the Nahal Hever scroll. The Nahal Hever scroll was written by two scribes; while both scribes generally wrote in the same style, the specific shapes and formation of letters are quite different.

Contrary to Thiede’s claim, even a non-expert could notice that despite the obvious general similarity of bilinear majuscule (a script where most letters fit between two imaginary lines and are of equal height - what we would call today ‘uppercase’) lettering, the letters between P64 and the Nahal Hever scroll are not as similar as he claimed they are - which is especially the case with Scribe B’s lettering.

You might notice that the Nahal Hever scroll (I’ll just call it NH) features a spindly, more decorated lettering than P64. Note that the letters feature blobs and hooks/serifs (actually a common feature of early formal hands - which is why they were more cumbersome to write). By contrast, P64 features a somewhat more sober, bold, undecorated (serif-less) lettering. While NH’s letters are quite narrow, P64’s could be placed within an imaginary square.

Thiede specifically named eight letters which he thought were similar between the two manuscripts: alpha (A), epsilon (E), eta (H), iota (I), mu (M), nu (N), omicron (O), and rho §. But close inspection shows:

  • The alphas of NH generally have a horizontal or a slightly-slanting cross-bar (like our modern As); P64’s have a distinct angle from lower left to upper right and is written more like this:
  • Most of the epsilons in NH - particularly Scribe B’s - have detached cross bars (it looks like a C with an unconnected line inside it)
  • NH’s iotas (I) have a sort of serif or curve at their bottom (Scribe A’s hooks turn to the viewer’s right; while Scribe B’s are in the opposite direction); P64’s Is are simple vertical strokes
  • Scribe B’s rhos § have a pronounced decorative blob on the end of their down-stroke; again, no such decoration in P64

Generally, the Scribes of NH (Scribe A in particular) have pronounced horizontal serifs to many of the verticals.


Thiede writes: “An equally obvious difference, on the other hand, may be seen in the Etas and Mys; but the second scribe of the Nahal Hever scroll provides the comparable Eta and My more than once.” One problem with the statement is, he never specified which scribe he was referring to. Does the “second scribe” refer to Scribe A (who incidentally, wrote most of the scroll) or to Scribe B (who wrote the last parts of Zechariah and maybe Malachi - which did not survive)? In any case, both Scribe A and B’s Hs and Ms are different from those found in NH. (Refer to the first pic in the last post.) The Hs generally have the horizontal serifs, while the mus have outward-pointing (slanting) down-strokes (kind of like an inverted W), unlike P64’s straighter M(s). So Thiede’s claim that either one of those scribes “provides the comparable Eta and My more than once” is questionable.

There’s actually a hole in Thiede’s methods here: he only focuses on (what he thinks are) similar-looking letters. Comparison of different hands requires the comparison of not just a few letters, but all the letters of the alphabet. Thiede did not even discuss the letters which are dissimilar in shape.

Thiede continues:

The Nahal Hever scroll of the Minor Prophets may be at the extreme end of the spectrum, but is not the only first century analogy. Further material is provided by papyri in the script of Herculaneum, for which AD 79 is the natural focal point.

It’s true that the Herculaneum papyri’s latest possible date is AD 79, but there’s a flaw here. Can you see what it is? It’s that Thiede fails to mention which of the Herculaneum papyri, if any, shows a similarity to P64. He does not offer any direct comparison, merely observing that it is another “first century analogy.” As Peter M. Head in his rebuttal to Thiede points out:

Thiede also appeals, in a general way, to parallels among the Herculaneum papyri and to 7Q61. …] Since numerous styles are exhibited among those Herculaneum papyri which have so far been unrolled, there are bound to be some general similarities to almost anything.

Next, Thiede says:

Interestingly, there is a small, unidentified Greek fragment from Qumran Cave 7, 7Q6, for which the archaeological terminus ante quem is AD 68, which has the characteristic Eta with the horizontal stroke above the median, evident in Magdalen Gr. 17.

Thiede’s argument essentially goes like: “P64’s eta (H) has the horizontal line slightly above the middle (median) of the letter = 7Q6’s H’s horizontal stroke is also above the median = since 7Q6 is from the 1st century, P64 must also be from the 1st century.” To be blunt, this is not a very good argument. Apparently, this ‘above the median’ eta occurs in all periods, in all but the squarest book hands: it’s not a 1st-century only element. Not to mention that there is no such thing as a ‘characteristic eta’ in P64, since the horizontal stroke occurs just about anywhere - above, in the middle, and below - across the different etas. Thiede was imputing too much meaning on the stroke’s placement - which doesn’t really mean anything.

7Q6. Note the eta (H) with its horizontal stroke in the second column.

I know many of the folks here are Americans, so I’ll use an ‘American’ analogy. Let’s say that I have a particular way of writing the letter C - it curves at the bottom, so that it might look like a G. A thousand years from now, someone discovers a small, undated sample of my handwriting and finds out that my C is somewhat similar to the way Carter Braxton wrote the C in his name in the Declaration of Independence. That person then claims that I must have lived during the late 1700s, due to this ‘characteristic C’. :stuck_out_tongue: I think that’s roughly what we have here with Thiede and P64.


If one was right-handed versus left-handed, the serifs would switch.

Make sense?

Good grist for someone’s PhD thesis.


They need to C14 date it in a non-destructive way like below.

It will at least provide bounds with its margin of error.


On a non-related note, the scholar Larry Hurtado actually talked about C-14 a year ago.

I returned yesterday from an invitational conference in Oklahoma City on dating papyri (sponsored by the Green Scholars Initiative, hereafter GSI). Out of respect for the presenters of papers, I won’t pre-empt publication by giving details. But I can say that I found the presentations on Carbon-dating especially informative and also of some significant import.

Essentially the GSI has access to the Green Collection of manuscripts & Bibles, and several papyri were chosen for rigorous Carbon-dating. The papyri in question had been dated first palaeographically, and then very small snippets were submitted to three respected laboratories in the USA for independent dating by Carbon-14 processes.

Many major libraries (e.g., the British Library) have a policy that does not permit any destruction of an item in any measure. So, since Carbon-dating requires that a tiny piece of an item be cut off and burned, hardly ever are we going to have Carbon-dating of items in these major collections. This is what makes the Green Collection policy so useful, not only for their own items, but also for the fields of papyrology and palaeography more widely.

To summarize results of the tests reported on in Oklahoma City, the results from the three labs were basically/broadly in agreement, which gives some assurance about the reliability of the process. But also, these results were broadly in agreement with the prior/independent palaeographical dating of these items. And this (as I see it) is the really larger import. It means (contrary to the reported comment by a distinguished papyrologist, who is not himself a palaeographer, that palaeographical dating is “bullst”), that palaeographical dating (using today’s standards and practices) by competent palaeographers can be treated as broadly reliable.**

And that means that collections that don’t allow Carbon-dating can take some further basis for confidence in the practice of palaeographical dating of their items as well.

Now, you must understand that Carbon-dating can, at best, offer a date-span of X plus/minus 50 years or so, e.g., X dated ca. 150-250 CE. That’s no more narrow than responsible palaeographers would date an item. But, as I say, the Green Collection’s tests do give us a second basis for some confidence in palaeographical dating practice.

Of course, one needs to take into account that C14 works on the writing material, not on the writing itself. It tells you when the papyrus ‘died’ (i.e. was made), not when the text was written on it. You need spectrometry tests on the ink (they’re non-destructive BTW).


Other than modern forgeries making use of old parchments, how common is it to have a huge disparity between the papyrus and the ink for documents. I assumed the procurement of the parchment would be on demand, not 50 years or more in advance (which would surpass the C14 margin of error).


Just to clarify: ‘papyrus’ is made from the papyrus plant. ‘Parchment’ is essentially treated and processed animal hide. (‘Vellum’, meanwhile is a specific type of parchment made from calf skin as opposed to that from other animals.) I’ll admit, I don’t know exactly just how much time elapsed between the making of a papyrus and writing on it. I suppose it would have varied, and I also think (like you) that there wouldn’t have been too great a time lapse - especially considering that papyrus (which while fresh can be as durable as paper) is more frail than parchment.

Oh yeah, I just remembered this. Remember ‘The Gospel of Jesus’ Wife’ (and a supposed Coptic fragment of John from the same group as GoJW) that made the rounds a couple years back? The writing was highly likelyto be a modern fake, ancient.was (Though the two C14 tests gave two conflicting dates: one earlier than the time of Jesus - 404 to 209 BC - the other around AD 741.) That’s a technique some forgers use: they would take real ancient (blank) papyrus, write stuff on them, and then present them as ancient texts. But in this case, the forger’s style (not to mention the text he wrote, and the dialect it was written in - which was already extinct by 741) gave him away.


Awesome post!

By the way, have you heard any info concerning the so-called oldest manuscript of the Gospel of Mark claimed by William Mounce? I read articles written a little after the discoveries but have not seen anything else about it.


Thiede continues:

There also is a Greek papyrus from Qumran Cave 4 which shows several letters resembling Papyrus Magdalen Gr.17, such as the Alpha, the Beta, etc.: pap4QLXXLeviticusb. As Parsons points out, the script is far from uniform, but this papyrus from Cave 4 could be dated to the mid-first century AD.

Unwittingly, he then procedes to offer an interesting case study: In his drawings of letters of the preceding fragment 4QLXXLeva (parts of a leather scroll which he dates to the first century BC), the Alpha, Beta, Delta, Epsilon, Eta, Iota, Kappa, Eta etc are identical or near-identical to what we find in Magdalen Gr.17. In fact, the letters he draws could have been taken straight out of Gr.17. Looking at the fragments themselves, there would seem to be at least two differences, however: the Qumran Leviticusa is sloping slightly to the right, and the letters are very close to each other, occasionally even connected (ligatures).

This is one of the parts of Thiede’s thesis that is savagely nitpicked the most by his critics. The reason? Thiede’s use of “etc.” Not only does Thiede not specify just what these two 'etc.'s are supposed to cover, he also fails to provide concrete examples of just how exactly the letters in these two manuscripts are “identical or near-identical” to P64. That’s sloppy scholarship. As textual critic David C. Parker (Was Matthew Written before 50 CE? The Magdalen Papyrus of Matthew) bluntly said:

Two other examples containing Leviticus (4QLXXLeva and b) are simply not used with enough precision for one to make a reasoned response. The former has eight letters ‘etc’ which are ‘identical or near-identical’; the other ‘shows several letters resembling [the Magdalen papyrus], such as the Alpha, the Beta, etc’. Unless we know the scope of ‘etc’, and the criteria according to which such words as ‘resembling’, ‘identical’ and ‘near-identical’ are used, we cannot comment.

Peter M. Head discusses the letters in more depth:

The second source to which Thiede appeals for comparison is pap4QLXXLevb (a papyrus manuscript of Leviticus in Greek from cave 4; see plate three). This manuscript, according to Thiede, ‘shows several letters resembling Papyrus Magdalen Gr. 17, such as the alpha, the beta, etc.’ It is not clear how many letters are covered by Thiede’s ‘etc.’! In fact, however, even those letters specified are not actually very similar: alpha has a horizontal half-serif at the lower end of its upright strokes and a horizontal cross-bar (contrast P64 as previously mentioned), while beta is not even fully attested on P64. In general, the style of pap4QLXXLevb is decorative with thin strokes and numerous hooks and (half-) serifs and no descenders below the bottom line (unlike P64). Some letters are very different from P64. For example, epsilon is very rounded, pi has a pronounced curve in its right upright, sigma is quite rounded, tau has an extended cross-bar, and upsilon has a lower hoop.

…] Another manuscript appealed to by Thiede is 4QLXXLeva, a fragment of the Greek Old Testament from Qumran (see plate four). Thiede argued that ‘the alpha, beta, epsilon, eta, iota, kappa, eta (sic) etc. are identical or near-identical to what we find in Magdalen Gr. 17’. The initial impression is that this script is quite different from those we have examined up to this point, generally lacking decoration and ornamentation. As will be obvious from the plate, the layout and general appearance is quite different from P64: more upright with narrower lettering and, as Thiede noted, a slight right-hand lean and a tendency to ligature.

As regards the letters mentioned by Thiede, we ought to note that the manuscript is not consistent in its letter-forms, so a number of different alpha-shapes occur (with very narrow horizontal bar, without any observable cross-bar, and with upward sloping cross-bar). The upward sloping cross-bar most closely approximates the alphas in P64 but could hardly be regarded as identical …] Since no beta is completely attested in P64, it is strange to find an appeal here, especially noting the proposed beta in P64 (the first letter of line 2 in Frag 3 verso) which has a very small lower circle, unlike those in 4QLXXLeva (see e.g. line 1, 9, 14 etc.). Only two deltas occur in 4QLXXLeva …], and they are narrower than those in P64. The epsilons of 4QLXXLeva are much more circular than those in P64. The etas and kappas are not particularly dissimilar, but the iotas do not extend below the line (as in P64). Other letters in 4QLXXLeva which are quite different from those of P64 include mu [M] (with outward sloping sides), pi Π] (with pronounced curve in right hand upright), rho [P] (more curved, single stroke), and upsilon [Y] (squarer upper section). Such a list of significant differences precludes any stylistic identification of 4QLXXLeva with P64.


Thiede claimed that he arrived at the conclusion that the letters of 4QLXXLeva are identical with that of P64 by looking at a drawing of the manuscript. We in 2014 have an advantage over Thiede in 1994-95: we now have a website where we can check high-resolution photos of the Dead Sea Scrolls ( These are three of the manuscripts Thiede appealed to - you check and judge for yourselves.

4QLXXLeva (4Q119)
Pap4QLXXLevb (4Q120)
Nahal Hever Minor Prophets (8HevXII)

I also took the trouble of making a couple of graphics yesterday (took up an hour or two of my time):

First, the letters of P64 compared with the Nahal Hever scroll (Scribe B’s hand - I forgot to make one for Scribe A) and the two Qumran Leviticus manuscripts. It would be better if you refer this graphic to both Thiede’s paper and to Peter Head’s response.

And yet another comparison with the three.


I did hear about a bit about it last year (or was it two years ago?), but the sources I’ve seen tend to associate Josh McDowell with it more than Mounce. The last things I heard about it are, well, these:

Oh, I forgot to add pics of P64 on the last post.


I forgot to add pics of P64 on the last post.



So I guess we can safely conclude that unlike Thiede’s assertion, the three or four manuscripts he invoked do not really show that much similarities with the lettering of P64. Peter Head, in his reply to Thiede, argues instead that P64 shows more similarities with 2nd and 3rd century manuscripts than it does with 1st century ones, thereby confirming the conventional date assigned to it (late 2nd-3rd century).

Although Thiede’s analysis fails to convince, a negative conclusion on that count hardly justifies the consensus date of around AD 200 without further discussion. In this section we shall assess the arguments, largely ignored by Thiede, of [C.H.] Roberts and [Ramón] Roca-Puig, with a particular focus on the palaeographical aspects of those discussions (analysis of handwriting style and appeal to comparable, ideally datable, manuscripts). These scholars identified the style of P64 as an early example of the ‘Biblical Uncial’, a style that is epitomised by the later biblical codices of Vaticanus and Sinaiticus, and appealed for comparison to other documents conventionally dated in the second or third centuries AD.

Most of the manuscripts Roberts and Roca-Puig - who were writing in the 60s-70s - gave as examples (and which Head cites in his article) are papyri found in Oxyrhynchus in Egypt, dated to within the same time range as the conventional dating of P64. (What makes Oxyrhynchus important - despite the fact that not much remains of the city, once the third-largest in Egypt - is the city dump. There archaeologists found a huge load of manuscripts, all dating from the period between the 1st to the 6th century AD.) These include:

A scroll of Plato’s Symposium
(P. Oxy. 843)
A fragment of Thucydides (P. Oxy. 1620)
A copy of St. Irenaeus’ Against Heresies (P. Oxy. 405)
Fragments of the poet and grammarian Antimachus (P. Oxy. 2516)
Fragments of a commentary on Aeschylus (P. Oxy. 2256)
An oracle book of Astrampsychus, with the papyrus reused for a 3rd century private letter (P. Oxy. 2832)
Fragment of Hesiod (P.Oxy. 2498)
A strip of papyrus containing lines from a comedy (P. Rylands 16)

This blog post here contains a good introduction to the various formal handwriting styles.

An example of ‘biblical majuscule’ (aka biblical uncial, 2nd century onwards). Note the more closer similarity of the letters with that of P64: as with P64, the letters are generally thick and square, with no serifs.


To sum up everything:

In 1994-95, Carsten Peter Thiede advanced a controversial theory concerning a papyrus fragment containing text from the gospel of Matthew (the Magdalen Papyrus, aka P64-67). While the fragment is conventionally dated to the late 2nd-3rd century, Thiede had argued that it should be redated earlier, well into the mid-to-late 1st century AD. (His paper, Papyrus Magdalen Greek 17 (Gregory-Aland P64): A Reappraisal, specifies “the late first century, some time after the destruction of the Temple in Jerusalem (AD 70),” which to be precise somewhat contradicts popular reports - and maybe even the book Thiede co-wrote with Matthew d’Ancona, Eyewitness to Jesus / The Jesus Papyrus, which insinuates that Thiede dated it to the AD 60s.)

Thiede claimed to have arrived at this late 1st century date by comparing P64 with three manuscripts that date from the 1st century BC-1st century AD: the Nahal Hever Minor Prophets scroll (8HevXII), and two Greek copies of Leviticus from Qumran (4QLXXLeva (4Q119) and Pap4QLXXLevb (4Q120)). Thiede had purportedly seen similarities in some letter forms between P64 and these manuscripts. In addition, Thiede also mentioned in passing a small Greek scrap, also from Qumran (7Q6) and the Herculaneum papyri.

However, critics of Thiede’s thesis - among whom was Peter M. Head of Tyndale House in Cambridge (The Date of the Magdalen Papyrus of Matthew) have pointed out that not only do the manuscripts Thiede invoked not show that much similarity with P64 as Thiede would have people believe, his methods which led him to arrive at his conclusion also, to put it frankly, not too scholarly. Thiede chose to focus only on a few select letters and ignored the other letters; he focused too much on similarities between letters (or what he thinks are similarities) and ignored their differences; his citations of similar letters between the two Leviticus manuscripts are too non-specific to be of any real worth (Leviticus b supposedly “shows several letters resembling Papyrus Magdalen Gr. 17, such as the alpha, the beta, etc.;” Leviticus a is purported to be similar with P64 in letters such as “the alpha, beta, epsilon, eta, iota, kappa, eta (sic) etc.”); his one-sentence reference to the Herculaneum papyri as a further “first century analogy” - “Further material is provided by papyri in the script of Herculaneum, for which AD 79 is the natural focal point” - is essentially, as Roger S. Bagnall (Early Christian Books in Egypt, p. 31) said, “just a drive-by shooting” (he offers no citation of specific texts or paleographical features as having similarities with P64); and his attempt to link the specific shape of the letter eta (H) in 7Q6 with that of P64 - a slightly raised horizontal bar - essentially proves nothing (since it was hardly ‘characteristic’ or a 1st-century only feature).

Critics instead point out that P64 actually shows a style of formal handwriting that only came into use during the 2nd century: the so-called biblical uncial (aka biblical majuscule). Head, in his reply to Thiede, compares P64 with some known 2nd-3rd century manuscripts written in this script (I listed some of these in the last post) and concludes that P64’s script shows more significant similarity with these than with the 1st-century manuscripts Thiede cited, which shows that, on basis of handwriting analysis, there’s really nothing that calls into question the conventional late 2nd-3rd century date assigned to the manuscript. He writes, “The very early manuscripts to which Thiede appealed for close parallels to P64 turned out to be not as close as the somewhat later ones which he had overlooked.”

I think that’s pretty much it for P64-67.


I just learned that Thiede, in his ‘for general audiences’ book (The Jesus Papyrus / Eyewitness to Jesus) invoked two more 1st-century manuscripts which he did not mention in his scholarly paper: one is a papyrus from Oxyrhynchus dated - yes, fortunately for us, the thing had a date written on it - around AD 66, while the other one is an ostracon - a pottery shard - from Masada (terminus ante quem: AD 73-74).

From The Oxyrhynchus Papyri, vol. 2 (Bernard P. Grenfell and Arthur S. Hunt, 1899)

This is P. Oxy. 0246, which contains a registration of seven newborn lambs by a guy named Harmiusis son of Petosiris, dated to “the present twelfth year of Nero Claudius Caesar Augustus Germanicus Imperator” (AD 66). (This was Harmiusis’ second registration in this year; he had earlier registered the birth of “twelve lambs which were born from sheep in my possession.”) You might notice that the main body of the text is written in a fine uncial hand; the cursive text bottom are the signatures of Harmiusis and three officials. Once again, Thiede claimed to have seen similarities between the letter-forms of the document and with P64.

Are they really similar? I’ll leave that for the reader to judge, but I’ll point out a few things. First: the two are really different scripts and documents: one (P64) is a book, written in a formal book hand, the other is a registration record, written in a less formal script (you might notice that the letters are thinner and a bit more semi-cursive than P64’s, not to mention that the Ks, Is, Ys, and Ts in particular have hooks/serifs - if you look closely). So you’re essentially comparing two different types of documents. (At least the three manuscripts Thiede named in his paper are copies of literature and are closer to what P64 is.) Second: just as with the other manuscripts Thiede had named, sure, some letters might look similar, but at the same time, other letters are different.

[D]ating by the style of handwriting is a matter of judgment and can only give an approximate answer, yet the assessment has to be made in the light of wide knowledge and constant acquaintance with the styles of centuries in question. By singling out certain letters for comparison and basing his conclusions on them, Thiede only undertook part of the exercise, for it is the whole range and style of the script that has to be compared with others, including the flow of the lines and the spacing as well as the shapes of individual letters and the ways they are formed.

The other is Masada 784 (aka Doc. Masada 784), a pottery shard found in Masada, a location in the Judaean desert famous for being the last bastion of the Jewish rebels in the first Jewish-Roman War (AD 66-73/74). To be honest, this piece doesn’t seem to be very promising either. I can’t find any picture of it on the Internet (there seems to be a photo of it on Roger Bagnall’s book Early Christian Books in Egypt though), but going by the available transcription of it, it’s very short - it apparently only contains nine legible letters (for a total of six different letters, if you group together the three instances of alpha and the two mus).

(line 2) vacant ? Λεα -ca.?- ]
καιαμμ̣ -ca.?- ]


I’m replying to your post again. Actually Thiede did present P64 in the carbon-dating lab of Oxford University for a feasibility test, but he was told that the fragments were too small and too light (the biggest fragment was only 45 milligrams, while the two smaller ones were 25 mg and 21 mg, respectively). If carbon tests were run on the fragments, the biggest fragment would be halved while the two smaller ones would be lost completely, since what was then the ‘newly-developed’ Accelerator Mass Spectrometry (AMS) method requires a minimum of 20 to 25 milligrams worth of material to be destroyed. (Even then, AMS was an improvement over the older Libby method, which required a larger amount of material to be destroyed.) And as mentioned, many major libraries don’t usually allow the destruction of their collection in any way.

I don’t think that C14 will be run on these fragments at the present, since it’ll take some time for this non-destructive method to be more commonplace.


So Patrick–what do you think is the date for the earliest manuscript fragment of scripture?

DISCLAIMER: The views and opinions expressed in these forums do not necessarily reflect those of Catholic Answers. For official apologetics resources please visit