The following series in an excerpt from my book, Can We Trust the Gospels? Investigating the Reliability of Matthew, Mark, Luke, and John.
Can We Know What the Original Gospel Manuscripts Really Said?
If you open a Bible and look for the gospels, you’ll find them in English translation, neatly collected at the beginning of the New Testament. You’ll see book names, chapter and verse numbers, punctuation, and paragraphs. None of these items were present in the original manuscripts of the writings we call Matthew, Mark, Luke, and John, however. Even spaces between words were rare. Aren’tyougladthingshavechanged? What you read in your Bible is the result of centuries of preservation, translation, and publication. Thus you might sensibly wonder if the gospels bear any resemblance to what the original writers actually penned almost 2,000 years ago.
It’s not uncommon these days for people to answer “No” to this question. Critics of Christianity often allege that the gospels as we know them don’t resemble the originals. This criticism appears, for example, on the lips of Sir Leigh Teabing, a fictional historian in Dan Brown’s wildly popular novel, The Da Vinci Code. Teabing “reveals” the true nature of the Bible in this way:
“The Bible is a product of man . . . . Not of God. The Bible did not fall magically from the clouds. Man created it as a historical record of tumultuous times, and it has evolved through countless translations, additions, and revisions. History has never had a definitive version of the book.” (p. 231)
There is a measure of truth here. The Bible is indeed a human product, though this in no way requires that it could not also be “of God.” For centuries, Christians have affirmed that the Bible was written by human authors who were inspired by God.
It’s true that the Bible “did not fall magically from the clouds.” It was in fact written by human beings who lived in “tumultuous times.” Yet the biblical documents were not created primarily as a “historical record” of these times. Though there is plenty of history in Scripture, the biblical writers focused mainly upon God’s action, salvation, and revelation within history.
Teabing exaggerates in saying that the Bible has been captured in “countless translations.” It has indeed been translated into more languages than any other book, by far. At last count, the New Testament has been translated into 1,541 languages.[i] But the Bible has not “evolved through countless translations,” as if our English versions stand at the end of a long chain of multi-lingual transformations. Every modern translation of Scripture is based upon manuscripts written in the same languages as those used by the original writers. The Old Testament in English comes directly from Hebrew and Aramaic manuscripts. Our New Testament is translated from Greek manuscripts.
The Relationship Between Existing Manuscripts and the Original Compositions
The documents we know as Matthew, Mark, Luke, and John were written sometime in the second half of the first century A.D. (I’ll say more about the dating of the gospels in Chapter 4.) They were written on scrolls of papyrus (a rough, paper-like substance). Papyrus was popular because it was readily available and relatively inexpensive. But, unfortunately for our purposes, it wasn’t especially durable. Thus it’s highly unlikely that any of the original gospel manuscripts, called by the technical term autographs, exists today. Probably, the biblical autographs were worn out through use, though perhaps they were misplaced by absentminded church leaders, destroyed by persecutors of the early Christians, or even eaten by critters.[ii]
Because ancient documents tended to have a relatively short shelf life, people who valued them had a way of preserving their contents: copying. Professional copyists, called scribes, would copy the words of one text into a fresh papyrus or parchment (a longer lasting material made from animal skins). Their training taught the scribes to minimize errors and maximize accuracy.
Yet copying manuscripts was not a slavish task, with scribal accuracy matching modern photocopy technology. At times scribes would make intentional changes as they copied. For example, they would correct what they believed to be a spelling error in their source text. Even the best of scribes, however, sometimes made unintended errors. Thus the best extant manuscripts of the gospels are likely to differ in some measure from the autographs.
Moreover, it’s probable that many of the first copies of the gospels were made, not by professional scribes, but by literate lay copyists. As the early church rapidly expanded throughout the Roman world in the first centuries A.D., there was a pressing need for multiple copies of authoritative Christian documents, including Matthew, Mark, Luke, and John. Non-professional copyists must have stepped in to meet this need.
The fact that the original gospel manuscripts have not survived to this day, combined with the fact that for centuries the text was passed on through a careful but imperfect process of copying, makes us wonder whether we can trust that the Greek text we have today looks anything like what the authors originally wrote down. Can we know what the original gospel manuscripts actually said?
Standards for Evaluating the Reliability of Gospel Manuscripts
Before we examine the data, let’s think for a moment about what might allow us to put confidence in the manuscripts of the gospels.
First, we’d look for antiquity. We’d want the manuscripts in existence to be old, the closer to the autographs the better. Less time between the original and an existing copy decreases the possibility of changes being introduced through many acts of copying.
Second, we’d prefer multiplicity. Clearly, it would be better to have many manuscripts at our disposal rather than just a few. More manuscripts would put us in a much better position to determine the original wording.
Third, we’d want trustworthy scholarly methodology. If the academics who study the biblical manuscripts, known as textual critics, utilize reliable methods, ones that maximize objectivity, then we’d have greater confidence in their conclusions.
Fourth, we’d look at the quantity and quality of textually ambiguous passages (made up of differences, called variants, among the manuscripts). If the existing copies of the gospels contain a high proportion of textual variants, then we’d have less confidence in our ability to access the content of the autographs. Moreover, if it turns out that the existing manuscripts are filled with all sorts of significant textual disagreements, then we’d question our knowledge of what was originally written. If, on the contrary, the differences among extant manuscripts are relatively insignificant, then we’d rightly place confidence in the critical Greek texts upon which our translations are based.
So how does reality measure up to these standards?
The Antiquity of the Gospel Manuscripts
The oldest manuscript of the gospels is a papyrus fragment of the Gospel of John. It is called P52 [capital ‘P’ with 52 superscript here and elsewhere], text critical shorthand for “Papyrus 52.” This fragment, which contains part of Jesus’s conversation with Pilate prior to the crucifixion (John 18:31-33, 37-38), has been dated to around 125 A.D. This means the copy of John of which P52 is a tiny part was made within a couple of generations of the original writing of the gospel.[iii] The next oldest manuscripts of the gospels come from the latter part of the second century and the early part of the third century. P4, P45, P64, P66, P67, P75 include significant portions of all four gospels.
As we move further into the third century and beyond, we find many more extant manuscripts, including one of the most important parchment copies of the entire Bible, known as Codex Sinaiticus. This book was found in the mid-19th century in a monastery near Mt. Sinai, from which it derives its name. It has been dated to the fourth century A.D., and is the oldest full Bible manuscript in existence.[iv]
How should we evaluate the antiquity of the gospel manuscripts? The smallest time gap, between P52 and the autograph of John, is two generations. The more complete manuscripts are about a century older than the original writings, with extant copies of the whole New Testament more than two centuries later than the time of composition. From our point of view, the period between the extant manuscripts of the gospels and the autographs may seem awfully long, and may raise doubts about the reliability of the gospel manuscripts.
But if we compare the antiquity of the gospel manuscripts with similar ancient writings, the case for the trusting the gospels gains considerable strength. Consider, for example, the writings of three historians more-or-less contemporaneous with the evangelists: the Jewish historian Josephus, and the Roman historians Tacitus and Suetonius. The oldest extant manuscripts of Tacitus and Suetonius come from the 9th century.[v] Those of Josephus are dated to the 11th century.[vi] We’re talking about a time gap of 800 to 1,000 years between the autographs and the extant manuscripts, yet historians accept the manuscripts as basically reliable representations of what was originally written. Lest it seem that I’ve chosen examples that are unusual, the oldest manuscripts of the classical historians Herodotus and Thucydides are separated from the autographs by about 500 years.[vii]
If someone were to claim that we can’t have confidence in the original content of the gospels because the existing manuscripts are too far removed from the autographs, then that person would also have to cast doubt upon our knowledge of almost all ancient history and literature. Such skepticism, which is not found among classical scholars and historians, would be extreme and unwarranted.
Therefore, on the antiquity scale, the New Testament gospels receive a top score.
The Multiplicity of the Gospel Manuscripts
Currently, scholars are aware of over 5,700 manuscripts that contain some portion of the New Testament, and the total is growing slowly as additional manuscripts are discovered. Among these manuscripts, a couple thousand contain all or portions of the biblical gospels.
Once again we should evaluate this total in light of comparable writings from the same period. What do we find if we look again at Tacitus, Suetonius, and Josephus? The histories of Tacitus exist today in three manuscripts, none of which contain all of his writings.[viii] We’re better off in the case of Suetonius, whose writings are found in more than 200 extant manuscripts. For Josephus we have 133 manuscripts.[ix] Once again, if it seems like I’m stacking the deck in my own favor, there are 75 manuscripts of Herodotus, and only 20 of Thucydides.
The number of gospel manuscripts in existence is about 20 times larger than the average number of extant manuscripts of comparable writings. I have not even considered the tens of thousands of manuscripts of gospel translations into languages such as Latin and Syriac, many of which were made in the earliest centuries A.D. I have also not taken into account the hundreds of thousands of quotations of the gospels found in the writings of early church leaders. Here’s what Bruce Metzger and Bart Ehrman have to say about these citations:
Besides textual evidence derived from the New Testament Greek manuscripts and from early versions, the textual critic has available the numerous scriptural quotations included in the commentaries, sermons, and other treatises written by early Church fathers. Indeed, so extensive are these citations that if all other sources for our knowledge of the text of the New Testament were destroyed, they would be sufficient alone for the reconstruction of practically the entire New Testament.[x]
After comparing the manuscripts of the New Testament with those for other ancient literature, Metzger and Ehrman conclude that “the textual critic of the New Testament is embarrassed by the wealth of material.”[xi]
The Reliability of Text Critical Methodology
Yet this “wealth of material” also complicates the work of textual criticism. What methods do text critics use to determine the earliest form of the gospel text?
First, they collect all of the known manuscripts, including ancient translations and writings of the early church fathers. The individual text critic doesn’t actually do this alone, of course, but relies upon the work of hundreds of other scholars, both present and past.
Second, text critics evaluate the manuscripts, looking for variants and seeking to determine which readings are the most likely to be original. They examine what is called external evidence and internal evidence. External evidence has to do with the number, antiquity, and relationships among the manuscripts. For example, if a variant is found in many, old manuscripts, then it is more reliable than one found in few, later manuscripts. Internal evidence concerns the actual content of the writing.
Though there is certainly a measure of subjectivity in text criticism, it is by far the most objective disciple in New Testament studies. If you were to take two different teams of text critics and ask them to work independently on a critical edition of the Greek New Testament, they would agree more than 99% of the time. In fact, for the vast majority of words in the gospels, text critics have come to an extremely high level of confidence concerning what was written in the autographs.
The Quantity and Quality of Textual Variants
Skeptics who try to cast doubt upon the reliability of the New Testament manuscripts point to the apparently large number of variants they contain. Bart Ehrman, for example, in Misquoting Jesus, suggests that there are 200,000 to 400,000 variants among the New Testament manuscripts. He adds, dramatically, “There are more variations among our manuscripts than there are words in the New Testament.”[xii] That sounds ominous, doesn’t it? But, in fact, the data give us no reason to doubt the reliability of the manuscripts. Let me explain why.
We have such a large number of variants because there are so many extant manuscripts. If you have over 2,000 manuscripts of the gospels, which comprise 64,000 words, that’s a lot of potential variants. But, as I’ve already shown, having many manuscripts actually increases the likelihood of our getting back to the original text. It also adds to the number of variants, however, which can sound negative to one who isn’t familiar with text critical issues.
Let’s me suggest a more hypothetical that might make clear what I’m saying. This book contains almost 50,000 words. Suppose I asked two people to make copies of this book by hand. Suppose, further, that they made one mistake every 1,000 words (99.9% accuracy). When they finished, each of their manuscripts would have 50 mistakes, for a total of 100. This doesn’t sound too bad, does it? But suppose I asked 2,000 people to make copies of my book. And suppose they also made a mistake every 1,000 words. When they finished, the total of mistakes in their manuscripts would be 100,000. This sounds like a lot of variants – more variants than words in my book, Bart Ehrman would say. But in fact the large number of variants is a simple product of the large number of manuscripts. Moreover, if text critics were going to try and determine what the autograph of this book said, they’d be in a much stronger position if they had 2,000 copies to work from, even though they’d be dealing with 100,000 variants. With 2,000 manuscripts, the text critics would be able to evaluate the variants more astutely and come up with something very close to what I originally wrote. If they only had two manuscripts, however, even though these included only 100 variants, they’d often be unable to determine what the original manuscript said.
So, the fact “there are more variations among our manuscripts than there are words in the New Testament” isn’t surprising. Nor it is bad news. It’s a reflection of the wealth of the manuscript evidence we have available to us. The actual number of variants represents a tiny percentage of possible variants among the manuscripts.
Moreover, the vast majority of variants in the New Testament manuscripts are insignificant, either because they appear so rarely that they are obviously not original, or because they don’t appear in the older manuscripts, or because they don’t impact the meaning of the text. In fact, the majority of variants that show up in enough older manuscripts to impact our reading of the text are spelling variations or errors.[xiii] Text critic Daniel Wallace concludes that “only about 1% of the textual variants” make any substantive difference.[xiv] And few, if any, of these have any bearing on theologically important matters. If you actually took out of the gospels every word that was text-critically uncertain, the impact on your understanding of Jesus would be negligible.
This, by the way, is exactly what most modern translations do with the two most obvious and significant textual variants in the gospels. One of these appears in John 7:53-8:11, the story of the woman caught in adultery. Virtually all translations put this story in brackets, adding a note that says something like: “The earliest manuscripts do not include this passage.” It’s likely that this story is true, but that it was added to John well after the evangelist finished his task. Similarly, the ending of Mark includes a bracketed passage because the old manuscripts do not include anything after Mark 16:8. These two disputed passages, though significant in some ways, do not substantially alter our understanding of Jesus.
Do the Gospel Manuscripts Misquote Jesus?
At this point I should say a few words about Bart Ehrman’s currently popular book Misquoting Jesus. Even when this book has fallen from the bestseller lists, its ideas will still be floating around in the cultural stream like bits of post-hurricane flotsam in the sea. (If you’re looking for a more extensive critique of Misquoting Jesus, check what I’ve written on my website,[xv] as well as several excellent scholarly reviews.[xvi])
Ehrman’s book is a popular introduction to textual criticism. When he sticks to objective descriptions, Ehrman’s insights are both helpful and readable. For a scholar, he’s an unusually effective popular communicator. But, unfortunately, this book was not written merely to introduce people to textual criticism, but also to undermine their confidence in the New Testament itself. I’m not reading between the lines here. Ehrman is very clear about his intentions from the beginning.[xvii]
One of the ironies of Ehrman’s book is the title, Misquoting Jesus. You’d expect to find a book full of instances in which the sayings of Jesus found in the gospels were corrupted by the scribes. In fact, however, very little of the book is actually about misquoting Jesus. As Craig L. Blomberg says in his trenchant review, “the title appears designed to attract attention and sell copies of the book rather than to represent its contents accurately.”[xviii]
Another irony comes when Ehrman talks about the number of variants among the New Testament manuscripts. Twice he says something like “there are more differences among our manuscripts than there are words in the New Testament.”[xix] This startling soundbite appears to undermine the reliability of the manuscripts. But Ehrman also qualifies this observation. He writes:
To be sure, of all the hundreds of thousands of textual changes found among our manuscripts, most of them are completely insignificant, immaterial, and of no real importance for anything other than showing that scribes could not spell or keep focused any better than the rest of us.[xx]
The changes [the scribes] made – at least the intentional ones – were no doubt seen as improvements of the text, possibly made because the scribes were convinced that the copyists before them had themselves mistakenly altered the words of the text. For the most part, their intention was to conserve the tradition, not to change it.[xxi]
One would expect to find these claims in a book touting the reliability of the New Testament manuscripts. Ehrman, in spite of his bias, is too good a scholar not to tell the truth here.
The greatest irony in Misquoting Jesus lies at the heart of Ehrman’s argument against the trustworthiness of the manuscripts. The main point of his book is to undermine confidence in the New Testament on the ground that copyists changed the manuscripts, both intentionally and accidentally. One would expect Ehrman to put forth dozens of examples where we simply don’t have any idea what the autographs actually said. Such repeated uncertainty would lead the conclusion that we can’t know with assurance what the New Testament writers, including the evangelists, actually wrote.
But, in fact, Ehrman’s book is filled with examples that prove the opposite point. He does indeed offer up many cases of textual variants. In virtually every case, Ehrman confidently explains what the change was, what the earlier manuscript actually said, and what motivated the copyist. In other words, Ehrman’s book, though intending to weaken our certainty about the New Testament text, actually demonstrates how the abundance of manuscripts and the antiquity of manuscripts, when run through the mill of text critical methodology, allows us to know with a very high level of probability what the evangelists and other New Testament authors wrote. This might explain why there are many textual critics who are committed Christians with an evangelical view of Scripture.[xxii]
Can we know what the original gospel manuscripts really said? Yes, we can. We can have confidence that the critical Greek texts of Matthew, Mark, Luke, and John represent, with a very high degree of probability, what the autographs of the gospels actually contained.
[ii] A fine discussion of how books were written in the time of early Christianity can be found in Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament: Its Transmission, Corruption, and Restoration, 4th ed. (Oxford, 2005). See pages 3-51, “The Making of Ancient Books.”
[iii] For more detail about P52 and the rest of the papyri, see Metzger and Ehrman, Text, 53-61.
[iv] For the fascinating story of the discovery of Codex Sinaiticus, see Metzger and Ehrman, Text, 62-67.
[v] Komoszewski, Sawyer, and Wallace, Reinventing Jesus, 71.
[vi] “Josephus” in Anchor Bible Dictionary, David Noel Freedman, ed. (New York: Doubleday, 1992).
[vii] Komoszewski, Sawyer, and Wallace, Reinventing Jesus, 71.
[viii] Metzger and Ehrman, Text, 50-51.
[ix] Komoszewski, Sawyer, and Wallace, Reinventing Jesus, 71.
[x] Metzger and Ehrman, Text, 126.
[xi] Metzger and Ehrman, Text, 51.
[xii] Ehrman, Misquoting Jesus, 90.
[xiii] Komoszewski, Sawyer, and Wallace, Reinventing Jesus, p. 56.
[xvi] Daniel Wallace, “The Gospel According to Bart,” http://www.bible.org/page.asp?page_id=4000#P174_80529; P. J. Williams, “Review of Misquoting Jesus,” http://evangelicaltextualcriticism.blogspot.com/2005/12/review-of-bart-ehrman-misquoting-jesus_31.html; Craig L. Blomberg, “Review of Misquoting Jesus,” http://www.denverseminary.edu/dj/articles2006/0200/0206.php.
[xvii] Ehrman, Misquoting Jesus, 10-15
[xviii] Blomberg, “Review of Misquoting Jesus.”
[xix] Ehrman, Misquoting Jesus, 11, 90.
[xx] Ehrman, Misquoting Jesus, 207.
[xxi] Ehrman, Misquoting Jesus, 215.
[xxii] See, for example, the text critics associated with the Evangelical Textual Criticism website: http://evangelicaltextualcriticism.blogspot.com/.