Misquoting Jesus: The Story Behind Who Changed the Bible and Why

by Bart D. Ehrman

On a deeper level, however, the experience of learning Greek became a bit troubling for me and my view of scripture. I came to see early on that the full meaning and nuance of the Greek text of the New Testament could be grasped only when it is read and studied in the original language (the same thing applies to the Old Testament, as I later learned when I acquired Hebrew). All the more reason, I thought, for learning the language thoroughly. At the same time, this started making me question my understanding of scripture as the verbally inspired word of God. If the full meaning of the words of scripture can be grasped only by studying them in Greek (and Hebrew), doesn’t this mean that most Christians, who don’t read ancient languages, will never have complete access to what God wants them to know?

What good does it do to say that the words are inspired by God if most people have absolutely no access to these words, but only to more or less clumsy renderings of these words into a language, such as English, that has nothing to do with the original words?

Maybe when Mark says that Jesus was crucified the day after the Passover meal was eaten (Mark 14:12; 15:25) and John says he died the day before it was eaten (John 19:14)—maybe that is a genuine difference. Or when Luke indicates in his account of Jesus’s birth that Joseph and Mary returned to Nazareth just over a month after they had come to Bethlehem (and performed the rites of purification; Luke 2:39), whereas Matthew indicates they instead fled to Egypt (Matt. 2:19–22)—maybe that is a difference. Or when Paul says that after he converted on the way to Damascus he did not go to Jerusalem to see those who were apostles before him (Gal. 1:16–17), whereas the book of Acts says that that was the first thing he did after leaving Damascus (Acts 9:26)—maybe that is a difference.

Possibly it is easiest to put it in comparative terms: there are more differences among our manuscripts than there are words in the New Testament.

Mark did not say the same thing that Luke said because he didn’t mean the same thing as Luke. John is different from Matthew—not the same. Paul is different from Acts. And James is different from Paul. Each author is a human author and needs to be read for what he (assuming they were all men) has to say, not assuming that what he says is the same, or conformable to, or consistent with what every other author has to say. The Bible, at the end of the day, is a very human book.

On some occasions these authoritative interpretations of scripture appear, in effect, to countermand the laws of scripture themselves. For example, Jesus says, “You have heard it said, ‘Whoever divorces his wife should give her a certificate of divorce’ [a command found in Deut. 24:1], but I say to you that everyone who divorces his wife for reason other than sexual immorality, makes her commit adultery, and whoever marries a divorced woman commits adultery.” It is hard to see how one can follow Moses’ command to give a certificate of divorce, if in fact divorce is not an option.

To make sense of this initial attempt to establish the canon, we need to know a bit about Marcion’s distinctive teaching. Marcion was completely absorbed by the life and teachings of the apostle Paul, whom he considered to be the one “true” apostle from the early days of the church.

Marcion’s attempt to make his sacred texts conform more closely to his teaching by actually changing them was not unprecedented. Both before and after him, copyists of the early Christian literature occasionally changed their texts to make them say what they were already thought to mean.

Many Christians today may think that the canon of the New Testament simply appeared on the scene one day, soon after the death of Jesus, but nothing could be farther from the truth. As it turns out, we are able to pinpoint the first time that any Christian of record listed the twenty-seven books of our New Testament as the books of the New Testament—neither more nor fewer. Surprising as it may seem, this Christian was writing in the second half of the fourth century, nearly three hundred years after the books of the New Testament had themselves been written.

In the year 367 C.E., Athanasius wrote his annual pastoral letter to the Egyptian churches under his jurisdiction, and in it he included advice concerning which books should be read as scripture in the churches. He lists our twenty-seven books, excluding all others. This is the first surviving instance of anyone affirming our set of books as the New Testament. And even Athanasius did not settle the matter. Debates continued for decades, even centuries. The books we call the New Testament were not gathered together into one canon and considered scripture, finally and ultimately, until hundreds of years after the books themselves had first been produced.

We have reason to think that the earliest Christian leaders were among the wealthier members of the church, in that the churches typically met in the homes of their members (there were no church buildings, that we know of, during the first two centuries of the church) and only the homes of the wealthier members would have been sufficiently large to accommodate very many people, since most people in ancient urban settings lived in tiny apartments. It is not unreasonable to conclude that the person who provided the home also provided the leadership of the church,

These wealthier homeowners would probably have been more educated, and so it is no surprise that they are sometimes exhorted to “read” Christian literature to their congregations, as we have seen, for example, in 1 Tim. 4:13: “Until I come, pay special heed to [public] reading, to exhortation, and to teaching.” Is it possible, then, that church leaders were responsible, at least a good bit of the time, for the copying of the Christian literature being read to the congregation?

The third-century church father Origen, for example, once registered the following complaint about the copies of the Gospels at his disposal: The differences among the manuscripts have become great, either through the negligence of some copyists or through the perverse audacity of others; they either neglect to check over what they have transcribed, or, in the process of checking, they make additions or deletions as they please.

Origen was not the only one to notice the problem. His pagan opponent Celsus had, as well, some seventy years earlier. In his attack on Christianity and its literature, Celsus had maligned the Christian copyists for their transgressive copying practices: Some believers, as though from a drinking bout, go so far as to oppose themselves and alter the original text of the gospel three or four or several times over, and they change its character to enable them to deny difficulties in face of criticism. (Against Celsus 2.27)

Marcion’s “orthodox” opponent Irenaeus claimed that Marcion did the following: dismembered the epistles of Paul, removing all that is said by the apostle respecting that God who made the world, to the effect that He is the Father of our Lord Jesus Christ, and also those passages from the prophetical writings which the apostle quotes, in order to teach us that they announced beforehand the coming of the Lord. (Against Heresies 1.27.2)

Marcion was not the only culprit. Living roughly at the same time as Irenaeus was an orthodox bishop of Corinth named Dionysius who complained that false believers had unscrupulously modified his own writings, just as they had done with more sacred texts. When my fellow-Christians invited me to write letters to them I did so. These the devil’s apostles have filled with tares, taking away some things and adding others. For them the woe is reserved. Small wonder then if some have dared to tamper even with the word of the Lord himself, when they have conspired to mutilate my own humble efforts.

We find this kind of imprecation already in one early Christian writing that made it into the New Testament, the book of Revelation, whose author, near the end of his text, utters a dire warning: I testify to everyone who hears the words of the prophecy of this book: If anyone adds to them, God will add to him the plagues described in this book; and if anyone removes any of the words of the book of this prophecy, God will remove his share from the tree of life and from the holy city, as described in this book. (Rev. 22:18–19) This is not a threat that the reader has to accept or believe everything written in this book of prophecy, as it is sometimes interpreted; rather, it is a typical threat to copyists of the book, that they are not to add to or remove any of its words.

An interesting illustration of the intentional change of a text is found in one of our finest old manuscripts, Codex Vaticanus (so named because it was found in the Vatican library), made in the fourth century. In the opening of the book of Hebrews there is a passage in which, according to most manuscripts, we are told that “Christ bears [Greek: PHERŌN] all things by the word of his power” (Heb. 1:3). In Codex Vaticanus, however, the original scribe produced a slightly different text, with a verb that sounded similar in Greek; here the text instead reads: “Christ manifests [Greek: PHANERŌN] all things by the word of his power.” Some centuries later, a second scribe read this passage in the manuscript and decided to change the unusual word manifests to the more common reading bears—erasing the one word and writing in the other. Then, again some centuries later, a third scribe read the manuscript and noticed the alteration his predecessor had made; he, in turn, erased the word bears and rewrote the word manifests. He then added a scribal note in the margin to indicate what he thought of the earlier, second scribe. The note says: “Fool and knave! Leave the old reading, don’t change it!”

The only way mistakes get corrected is when a scribe recognizes that a predecessor has made an error and tries to resolve it. There is no guarantee, however, that a scribe who tries to correct a mistake corrects it correctly. That is, by changing what he thinks is an error, he may in fact change it incorrectly, so now there are three forms of the text: the original, the error, and the incorrect attempt to resolve the error. Mistakes multiply and get repeated; sometimes they get corrected and sometimes they get compounded. And so it goes. For centuries.

To conclude this chapter, I will discuss two such passages—passages from the Gospels, in this case, that we are now fairly certain did not originally belong in the New Testament, even though they became popular parts of the Bible for Christians down through the centuries and remain so today.

Of course, to a careful reader, the story raises numerous questions. If this woman was caught in the act of adultery, for example, where is the man she was caught with? Both of them are to be stoned, according to the Law of Moses (see Lev. 20:10). Moreover, when Jesus wrote on the ground, what exactly was he writing? (According to one ancient tradition, he was writing the sins of the accusers, who seeing that their own transgressions were known, left in embarrassment!) And even if Jesus did teach a message of love, did he really think that the Law of God given by Moses was no longer in force and should not be obeyed? Did he think sins should not be punished at all?

Notice what happens when these verses are taken away. The women are told to inform the disciples that Jesus will precede them to Galilee and meet them there; but they, the women, flee the tomb and say nothing to anyone, “for they were afraid.” And that’s where the Gospel ends. Obviously, scribes thought the ending was too abrupt. The women told no one? Then, did the disciples never learn of the resurrection? And didn’t Jesus himself ever appear to them? How could that be the ending! To resolve the problem, scribes added an ending.

Starting in the fourth century, then, copies of scripture began to be made by professionals; this naturally curtailed significantly the number of errors that crept into the text. Eventually, as the decades grew into centuries, the copying of the Greek scriptures became the charge of monks working out of monasteries, who spent their days copying the sacred texts carefully and conscientiously. This practice continued on down through the Middle Ages, right up to the time of the invention of printing with moveable type in the fifteenth century.

The great mass of our surviving Greek manuscripts come from the pens of these medieval Christian scribes who lived and worked in the East (for example, in areas that are now Turkey and Greece), known as the Byzantine Empire.

This form of the Bible in Latin—Jerome’s translation—came to be known as the Vulgate (= Common) Bible of Latin-speaking Christendom. This was the Bible for the Western church, itself copied and recopied many times over. It was the book that Christians read, scholars studied, and theologians used for centuries, down to the modern period. Today there are nearly twice as many copies of the Latin Vulgate as there are Greek manuscripts of the New Testament.

in the East (the Byzantine text) and in the West (the Latin Vulgate).

For the book of Revelation he had to borrow a manuscript from his friend the German humanist Johannes Reuchlin; unfortunately, this manuscript was almost impossible to read in places, and it had lost its last page, which contained the final six verses of the book. In his haste to have the job done, in those places Erasmus simply took the Latin Vulgate and translated its text back into Greek, thereby creating some textual readings found today in no surviving Greek manuscript. And this, as we will see, is the edition of the Greek New Testament that for all practical purposes was used by the translators of the King James Bible nearly a century later.

Stephanus’s third edition of 1550 is notable as the first edition ever to include notes documenting differences among some of the manuscripts consulted; his fourth edition (1551) is possibly even more significant, as it is the first edition of the Greek New Testament that divides the text into verses. Until then, the text had been printed all together, with no indication of verse division. There’s an amusing anecdote associated with how Stephanus did his work for this edition. His son later reported that Stephanus had decided on his verse divisions (most of which are retained for us in our English translations) while making a journey on horseback. Undoubtedly he meant that his father was “working on the road”—that is, that he entered verse numbers in the evenings at the inns in which he was staying. But since his son literally says that Stephanus made these changes “while on horseback,” some wry observers have suggested that he actually did his work in transit, so that whenever his horse hit an unexpected bump, Stephanus’s pen jumped, accounting for some of the rather odd verse placements that we still find in our English translations of the New Testament.

Now the status of the original text was thrown wide open to dispute. If one did not know which words were original to the Greek New Testament, how could one use these words in deciding correct Christian doctrine and teaching?

“the Papists”—would be all too happy to be able to show, on the basis of the insecure foundations of the Greek text of the New Testament, that scripture was not a sufficient authority for the faith—that is, that the authority of the church instead is paramount.

With this abundance of evidence, what can we say about the total number of variants known today? Scholars differ significantly in their estimates—some say there are 200,000 variants known, some say 300,000, some say 400,000 or more! We do not know for sure because, despite impressive developments in computer technology, no one has yet been able to count them all. Perhaps, as I indicated earlier, it is best simply to leave the matter in comparative terms. There are more variations among our manuscripts than there are words in the New Testament.

one can see how a scribe might inadvertently leave out a word when copying a text (an accidental change), but it is hard to see how the last twelve verses of Mark could have been added by a slip of the pen.

at the very beginning of Mark, where the author introduces his Gospel by saying, “Just as is written in Isaiah the prophet, ‘Behold I am sending a messenger before your face…. Make straight his paths.’” The problem is that the beginning of the quotation is not from Isaiah at all but represents a combination of a passage from Exod. 23:20 and one from Mal. 3:1. Scribes recognized that this was a difficulty and so changed the text, making it say, “Just as is written in the prophets….” Now there is no problem with a misattribution of the quotation. But there can be little doubt concerning what Mark originally wrote: the attribution to Isaiah is found in our earliest and best manuscripts.

Bentley’s method of proceeding was rather straightforward. He had decided to collate (i.e., to compare in detail) the text of the most important Greek manuscript of the New Testament in England, the early-fifth-century Codex Alexandrinus, with the oldest available copies of the Latin Vulgate. What he found was a wide range of remarkable coincidences of readings, in which these manuscripts agreed time and again with each other but against the bulk of Greek manuscripts transcribed in the Middle Ages. The agreements extended even to such matters as word order, where the various manuscripts differed.

The logic behind the method was simple: if, in fact, Jerome used the best Greek manuscripts available for editing his text, then by comparing the oldest manuscripts of the Vulgate (to ascertain Jerome’s original text) with the oldest manuscripts of the Greek New Testament (to ascertain which were the ones used by Jerome), one could determine what the best texts of Jerome’s day had looked like—and skip over more than a thousand years of textual transmission in which the text came to be repeatedly changed.

Moreover, since Jerome’s text would have been that of his predecessor Origen, one could rest assured that this was the very best text available in the earliest centuries of Christianity.

1 Tim. 3:16, had long been used by advocates of orthodox theology to support the view that the New Testament itself calls Jesus God. For the text, in most manuscripts, refers to Christ as “God made manifest in the flesh, and justified in the Spirit.” As I pointed out in chapter 3, most manuscripts abbreviate sacred names (the so-called nomina sacra), and that is the case here as well, where the Greek word God (ΘEOΣ) is abbreviated in two letters, theta and sigma (ΘΣ), with a line drawn over the top to indicate that it is an abbreviation.

Wettstein noticed in examining Codex Alexandrinus was that the line over the top had been drawn in a different ink from the surrounding words, and so appeared to be from a later hand (i.e., written by a later scribe). Moreover, the horizontal line in the middle of the first letter, Θ, was not actually a part of the letter but was a line that had bled through from the other side of the old vellum. In other words, rather than being the abbreviation (theta–sigma) for “God” (ΘΣ), the word was actually an omicron and a sigma (OΣ), a different word altogether, which simply means “who.” The original reading of the manuscript thus did not speak of Christ as “God made manifest in the flesh” but of Christ “who was made manifest in the flesh.” According to the ancient testimony of the Codex Alexandrinus, Christ is no longer explicitly called God in this passage.

in Acts 20:28, which in many manuscripts speaks of “the Church of God, which he obtained by his own blood.” Here again, Jesus appears to be spoken of as God. But in Codex Alexandrinus and some other manuscripts, the text instead speaks of “the Church of the Lord, which he obtained by his own blood.” Now Jesus is called the Lord, but he is not explicitly identified as God.

Wettstein does print the Textus Receptus, but he also amasses a mind-boggling array of Greek, Roman, and Jewish texts that parallel statements found in the New Testament and can help illuminate their meaning.

Lachmann therefore set for himself a simpler task. The Textus Receptus, he knew, was based on the manuscript tradition of the twelfth century. He could improve upon that—by eight hundred years—by producing an edition of the New Testament as it would have appeared at about the end of the fourth century. The surviving manuscripts in Greek, along with the manuscripts of Jerome’s Vulgate and the quotations of the text in such writers as Irenaeus, Origen, and Cyprian, would at the very least allow that. And so that is what he did. Relying on a handful of early majuscule manuscripts plus the oldest Latin manuscripts and the patristic quotations of the text, he chose not simply to edit the Textus Receptus wherever necessary (the tack followed by his predecessors who were dissatisfied with the T.R.), but to abandon the T.R. completely and to establish the text anew, on his own principles.

in 1831 he produced a new version of the text, not based on the T.R. This was the first time anyone had dared to do so. It had taken more than three hundred years, but finally the world was given an edition of the Greek New Testament that was based exclusively on ancient evidence.

Lachmann had broken through the unhelpful custom established among printers and scholars alike of giving favored status to the T.R., a status it surely did not deserve, since it was printed and reprinted not because anyone felt that it rested on a secure textual basis but only because its text was both customary and familiar.

Tischendorf ’s journeys to other lands in Europe and the Middle East to locate manuscripts. By all counts, his most famous discovery involves one of the truly great manuscripts of the Bible still available, Codex Sinaiticus. The tale of its discovery is the stuff of legend, although we have the account direct from Tischendorf’s own hand.

Westcott and Hort’s publication appeared in two volumes, one of which was an actual edition of the New Testament based on their twenty-eight years of joint labor in deciding which was the original text wherever variations appeared in the tradition; the other was an exposition of the critical principles they had followed in producing their work. The latter was written by Hort and represents an inordinately closely reasoned and compelling survey of the materials and methods available to scholars wanting to undertake the tasks of textual criticism. The writing is dense; not a word is wasted. The logic is compelling; not an angle has been overlooked. This is a great book, which in many ways is the classic in the field. I do not allow my graduate students to go through their studies without mastering it.

The area in which their work has perhaps proved most significant is in the grouping of manuscripts. Since Bengel had first recognized that manuscripts could be gathered together in “family” groupings (somewhat like drawing up genealogies of family members), scholars had attempted to isolate various groups of witnesses into families. Westcott and Hort were very much involved in this endeavor as well. Their view of the matter was based on the principle that manuscripts belong in the same family line whenever they agree with one another in their wording. That is, if two manuscripts have the same wording of a verse, it must be because the two manuscripts ultimately go back to the same source—either the original manuscript or a copy of it.

One can then establish family groups based on textual agreements among the various surviving manuscripts. For Westcott and Hort there were four major families of witnesses:

the Syrian text (what other scholars have called the Byzantine text), which comprises most of the late medieval manuscripts; these are numerous but not particularly close in wording to the original text;

the Western text, made up of manuscripts that could be dated very early—the archetypes must have been around sometime in the second century at the latest; these manuscripts, however, embody the wild copying practices of scribes in that period before t...

the Alexandrian text, which was derived from Alexandria, where the scribes were trained and careful but occasionally altered their texts to make them grammatically and stylistically more acceptabl...