by Lee Strobel
Some people have likened textual criticism to the children’s game of telephone, in which a short message is communicated to an individual by whispering in that person’s ear. That person then whispers in the next person’s ear and so on for several people. Then the last individual says the message out loud, and inevitably it has become terribly garbled by the time it goes all the way down the chain. The implication is that because textual criticism is like this, people simply can’t trust what the New Testament says today. In short, we can’t have any confidence that it accurately represents the original.
Wallace, however, said that analogy breaks down at several key points.
“First of all,” he said, “rather than having one stream of transmission, we have multiple streams. Now suppose you were to interrogate the last person in, say, three lines. All of them repeat the message they heard in their own line, and that message ultimately goes back to one source. There would certainly be differences in the resultant message, but there also would be similarities. By a little detective work, you could figure out much of what the original message was by comparing the three different reports of it. Of course, you still would have a lot of doubt as to whether you got it right.
“A second difference with the telephone game,” he continued, “is that rather than dealing with an oral tradition, textual criticism deals with a written tradition. Now, if each person in the line wrote down what he heard from the person in front of him, the chances for garbling the message would be remote—and you’d have a pretty boring game!” he added with a smile.
“A third difference is that the textual critic—the person trying to reconstruct what the original message was—does not have to rely on that last person in the chain. He can interrogate several folks who are closer to the original source.”
His conclusion? “Putting all this together, the cross-checks among the various streams of transmission, the examination of early generations of copies—often exceedingly early—and the written records rather than oral tradition, make textual criticism quite a bit more exacting and precise than the game of telephone,” he said.
There is, however, another game that does demonstrate the effectiveness of textual criticism. Wallace himself has conducted seminars called “The Gospel According to Snoopy” for the past thirty years at universities and other settings. His goal is to demonstrate in a practical way how textual criticism can succeed in reconstructing a missing text.
“In the game, numerous people serve as ‘scribes,’ who copy out an ancient text on a Friday night,” he said. “There are six generations of copies. The scribes all make mistakes, intentionally or unintentionally. In fact, the resultant copies are actually significantly more corrupt than the manuscript copies of the New Testament.”
“How corrupt?” I asked.
“For a fifty-word document, they are able to produce hundreds of textual variants,” he said. “Then the next morning the rest of the folks at the seminar get to work as textual critics, with the scribes as silent onlookers. But they don’t have all the manuscripts to work with. The earliest copies were destroyed or lost. And there are many breaks in the chain. But the textual critics do the best they can with the materials they have.
“After about two hours of work, they come up with what they think the original text said. There are some doubts at almost every turn. But remarkably, even with the doubts, the core idea is hardly changed. Sometimes the doubts have to do with ‘too’ versus ‘also,’ or ‘shall’ versus ‘will.’ Then, I show the group the original text and we compare the two texts, line by line, word by word.”
“How successful are these amateur textual critics?” I asked.
“Altogether, I’ve conducted this seminar over fifty times in churches, colleges, and seminaries—and we have never missed reconstructing the original text by more than three words. In fact, we were off by three words only once. Often, the group has gotten the original wording exactly right—and the essential message of the original is always intact. Sometimes people break out into spontaneous applause at the end!”
“What’s the lesson, then?” I asked.
“It’s basically this,” he said. “If people who know nothing about textual criticism can reconstruct a text that has become terribly corrupted, then isn’t it likely that those who are trained in textual criticism can do the same with the New Testament?”
AN EMBARRASSMENT OF RICHES
As Wallace’s seminar demonstrates, having a handful of copies can help even amateur sleuths to determine the wording of the missing original text. Scholars trying to reconstruct the text of the New Testament, however, have thousands of manuscripts to work with. The more copies, the easier it is to discern the contents of the original. Given their centrality to textual criticism, I asked Wallace to talk about the quantity and quality of New Testament documents.
“Quite simply, we have more witnesses to the text of the New Testament than to any other ancient Greek or Latin literature. It’s really an embarrassment of riches!” he declared.
“Exactly how many copies are in existence?” I asked.
“We have more than 5,700 Greek copies of the New Testament. When I started seminary, there were 4,800, but more and more have been discovered. There are another 10,000 copies in Latin. Then there are versions in other languages—Coptic, Syriac, Armenian, Georgian, and so on. These are estimated to number between 10,000 and 15,000. So right there we’ve got 25,000 to 30,000 handwritten copies of the New Testament.”
“But aren’t many of these merely fragments?” I asked.
“A great majority of these manuscripts are complete for the purposes that the scribes intended. For example, some manuscripts were intended just to include the Gospels; others, just Paul’s letters. Only sixty Greek manuscripts have the entire New Testament, but that doesn’t mean that most manuscripts are fragmentary. Most are complete for the purposes intended,” Wallace said.
“Now, if we were to destroy all these manuscripts, would we be left without a witness?” he asked. Without waiting for a response, he said, “Not at all. The ancient church fathers quoted so often from the New Testament that it would be possible to reconstruct almost the entire New Testament from their writings alone. All told, there are more than one million quotations of the New Testament in their writings. They date as early as the first century and continue through the thirteenth century, so they’re extremely valuable for determining the wording of the New Testament text.”
I asked Wallace about the dates of the manuscripts. “About 10 percent of these manuscripts come from the first millennium,” he said. “Through the first three centuries, we have nearly fifty manuscripts in Greek alone. Yet remarkably, the additions to the text over fourteen centuries of copying amount to about 2 percent of the total. In other words, the New Testament grew over time, but at less than 2 percent growth per millennium—so banking on its expansion would be a poor investment!
“The quantity and quality of the New Testament manuscripts are unequalled in the ancient Greco-Roman world. The average Greek author has fewer than twenty copies of his works still in existence, and they come from no sooner than five hundred to a thousand years later. If you stacked the copies of his works on top of each other, they would be about four feet tall. Stack up copies of the New Testament and they would reach more than a mile high—and, again, that doesn’t include quotations from the church fathers.
“Even the great historians who give us much of our understandings of ancient Roman history are quite incomplete,” he added. “Livy, for example, wrote 142 volumes on the history of Rome, but only 35 survive. When you compare the New Testament to the second most copied Greek author, the differences are truly astounding. Homer’s Iliad and Odyssey combined have fewer than 2,400 copies—yet Homer has an eight-hundred-year head start on the New Testament! At bottom, textual criticism for virtually all other ancient literature relies on creative conjectures, or imaginative guesses, at reconstructing the wording of the original. Not so with the
New Testament.”
Another critical factor is how early manuscripts are dated. Obviously, those closest to the original are the most valued. When I asked Wallace about the dates of New Testament manuscripts, he smiled and started with a story.
“In 1844, F. C. Baur, the father of modern theological liberalism, argued that the Gospel of John was really a synthesis of Peter and Paul’s Christianity and it had to be dated after AD 160,” he said. “If this were true, then the historical credibility of that Gospel would be very questionable. Baur’s best guess was AD 170—but it was based on philosophical presuppositions. Well, as someone once said, ‘An ounce of evidence is worth a pound of presumption.’
“In 1934, a papyrologist named Colin H. Roberts was rummaging around in the basement of the John Rylands Library at Manchester University in England. He found a papyrus fragment that was no bigger than the palm of my hand. He read one side and—oh my gosh!—this was John 18:31–33. He flipped it over, and it was John 18:37–38.
“Now, you have to understand that finding a Greek New Testament fragment on papyrus is exceedingly rare. We’ve found somewhere in the neighborhood of 75,000 papyri, and only 117 are from the New Testament. So this finding was just remarkable. Then he sent the fragment to three leading papyrologists in Europe. Each wrote back independently and said, ‘This manuscript is not to be dated any later than AD 150 and is as early as AD 100—and I prefer the earlier date.’ A fourth expert, Adolph Deissman, said it should be dated to the 90s. So this one scrap of papyrus sent two tons of liberal German scholarship to the flames! An ounce of evidence really is worth a pound of presumption.”
“Is that the only fragment from the second century?” I asked.
“Not only isn’t it the only one, but in the last five years at least three or four others have also been found from the second century in a museum at Oxford. They were excavated from Oxyrinkchus, Egypt, in 1906, and have been sitting there for nearly a century. They didn’t have enough papyrologists to go through all the fragments! To this date we have between ten to fifteen papyri from the second century. That’s remarkable—from within a hundred years of when the New Testament was completed. It’s absolutely stunning to have that kind of data.
“And even though they’re fragmentary, they’re not always small. We have, for example, P66, which is from mid-to-late second century and has almost the entirety of John’s Gospel. P46, which dates to about AD 200, has got seven of Paul’s letters and Hebrews in it. P75, which is late second century to early third century, has John and Luke almost in their entirety. P45 is early too—and it has large portions of the four Gospels, so that’s a substantial amount of evidence. The earliest manuscripts were on papyrus, and all the papyri together equal about half of the New Testament.”
“So we have a really small gap, then, between the actual earliest papyrus and the New Testament documents,” I said in summary.
“Right. There’s just no comparison to others,” he said. “For other great historians, there’s a three-hundred-year gap before you get a sliver of a fragment, and then sometimes you have to wait another thousand years before you see something else.”
EXPLAINING THE VARIANTS
Among Ehrman’s disclosures that alarmed readers was that there are somewhere between 200,000 and maybe 400,000 variants between New Testament manuscripts—in fact, more variants than the 138,162 words in the published Greek New Testament. This was old news to textual critics, but it was shocking to the general public. Yet are these variants really significant—and do they jeopardize the message of the Gospels and their depiction of Jesus?
“Tell me about these variants—how are they counted, and how did they come about?” I asked Wallace.
“If there’s any manuscript or church father who has a different word in one place, that counts as a textual variant,” Wallace explained. “If you have a thousand manuscripts that have, for instance, ‘Lord’ in John 4:1, and all the rest of the manuscripts have ‘Jesus,’ that still counts as only one variant. If a single fourteenth-century manuscript misspells a word, that counts as a variant.”
“What are the most common variants?” I asked.
“Far and away, the most common are spelling variations, even when the misspelling in Greek makes absolutely no difference in the meaning of the word,” he said.
“For example, the most common textual variant involves what’s called a ‘moveable nu.’ The Greek letter nu—or ‘n’—is used at the end of a word when the next word starts with a vowel. It’s like in English, where you have an indefinite article—an apple or a book. It means the same thing. Whether a nu appears in these words or not has absolutely no effect on its meaning. Yet they still record all those as textual variants.
“Another example is that every time you see the name John, it’s either spelled with one or two n’s. They have to record that as a textual variant—but how it comes out in English is ‘John’ every time. It doesn’t make any difference. The point is, it’s not spelled Mary! Somewhere between 70 to 80 percent of all textual variants are spelling differences that can’t even be translated into English and have zero impact on meaning.”
I did some quick mental math: taking the high estimate of 400,000 New Testament variants, that would mean 280,000 to 320,000 of them would be inconsequential differences in spelling. “Please, continue,” I said to Wallace.
“Then you’ve got nonsense errors, where a scribe was inattentive and makes a mistake that’s an obvious no-brainer to spot,” he said. “For example, in a manuscript in the Smithsonian Institution, one scribe wrote the word ‘and’ when he meant to write ‘Lord.’ The words look somewhat similar in Greek—kai versus kurios. It was obvious that the word ‘and’ doesn’t fit the context. So in these cases, it’s easy to reconstruct the right word.
“There are also variants involving synonyms. Does John 4:3 say, ‘When Jesus knew’ or ‘When the Lord knew’? We’re not sure which one goes back to the original, but both words are true. A lot of variants involve the Greek practice of using a definite article with a proper name, which we don’t do in English. For example, a manuscript might refer to ‘the Mary’ or ‘the Joseph,’ but the scribe might have simply written ‘Mary’ or ‘Joseph.’ Again, there’s no impact on meaning, but they’re all counted as variants.
“On top of that, you’ve got variants that can’t even be translated into English. Greek is a highly inflected language. That means the order of words in Greek isn’t as important as it is in English. For example, there are sixteen different ways in Greek to say, ‘Jesus loves Paul,’ and they would be translated into English the very same way. Still, it counts as a textual variant if there’s a difference in the order of words, even if the meaning is unaffected.”
Wallace stopped for a moment to consider the situation. “So if we have approximately 200,000 to 400,000 variants among the Greek manuscripts, I’m just shocked that there are so few!” he declared. “What would the potential number be? Tens of millions! Part of the reason we have so many variants is because we have so many manuscripts. And we’re glad we’ve got so many manuscripts—it helps us immensely in getting back to the original.”
I asked, “How many textual variants really make a difference?”
“Only about one percent of variants are both meaningful, which means they affect the meaning of the text to some degree, and viable, which means they have a decent chance of going back to the original text.”
“Still, that’s a pretty big number,” I said.
“But most of these are not very significant at all,” he said.
“Give me an example.”
“Okay,” he replied. “I’ll describe two of the most notorious issues. One involves Romans 5:1. Did Paul say, ‘We have peace’ or ‘let us have peace’? The difference amounts to one letter in the Greek. Scholars are split on this, but the big point is that neither variant is a contradiction of the teachings of scripture.
“Another famous example is 1 John 1:4. The verse says eit
her, ‘Thus we are writing these things so that our joy may be complete,’ or, ‘Thus we are writing these things so that your joy may be complete.’ There’s ancient testimony for both readings. So, yes, the meaning is affected, but no foundational beliefs are in jeopardy. Either way, the obvious meaning of the verse is that the writing of this letter brings joy.”
It was simply amazing to me that two of the most notorious textual issues are, at bottom, so trivial in their implications.
INTENTIONAL CHANGES
There are a lot of reasons why textual errors occur, many of them involving scribes being inattentive. Ehrman puts a lot of emphasis, however, on scribes who intentionally altered the text as they reproduced it for the next generation of manuscripts. “That makes people very nervous,” I said to Wallace.
“Well, he’s absolutely correct,” Wallace replied. “Sometimes scribes did intentionally change the text.”
“What’s the most common reason?” I asked.
“They wanted to make the text more explicit. Through the centuries, for example, the church started using sections of scripture for daily readings. These are called lectionaries. About 2,200 of our Greek manuscripts are lectionaries, where they will set forth a year’s worth of daily or weekly scripture readings.
“Here’s what happened: In the Gospel of Mark, there are eighty-nine verses in a row where the name of Jesus isn’t mentioned once. Just pronouns are used, with ‘he,’ referring to Jesus. Well, if you excerpt a passage for a daily lectionary reading, you can’t start with: ‘When he was going someplace…’ The reader wouldn’t know whom you were referring to. So it was logical for the scribe to replace ‘he’ with ‘Jesus’ in order to be more specific in the lectionary. But it’s counted as a variant every single time.
“Or here’s another example: one lectionary reading says, ‘When Jesus was teaching his disciples.’ In the original, it doesn’t say ‘Jesus’ or ‘his disciples,’ but it’s clear from the context that this was meant. So the scribes were merely making things explicit in the lectionaries. No meaning is changed whatsoever—yet it’s counted as a variant.