Has the Text Been Changed?

Start with the anchor, because it is often used to say more than it says. Isaiah 40 opens the great section of the book addressed to Jews in exile in Babylon. 'Comfort, comfort My people' (40:1): the punishment is over and God is coming to lead his people home. Against that promise stand the exiles' doubts and the empires that seem so permanent. 'All flesh is like grass' (40:6). Babylon will wither; the word God has spoken about his people's future will not. Peter quotes the passage in 1 Peter 1:24-25 and identifies the enduring word as the gospel preached to his readers. The promise is that what God has said will come to pass and will not fail.

Notice what the verse does not say. It does not promise that no scribe will ever make a mistake in copying, that one particular manuscript will be kept from all error, or that every reader in every century will have the same wording in front of him. The inquiry's reading of Jeremiah 36 makes the point from inside the Bible: a king burned the scroll, the word was rewritten, and many similar words were added. The word survived the loss of its first written form. That is a fact a Christian can rest on, but it is a different claim from the claim that the copies never changed. A student who answers the textual question by quoting Isaiah 40:8 alone has answered a question the sceptic did not ask.

How God preserves the written form of his word is a question on which Christians genuinely disagree, and the disagreement should be stated fairly. Most evangelical and Catholic scholars hold that God has preserved the text in the whole body of manuscripts, translations and quotations, from which the original wording can be recovered with very high confidence by careful comparison; this is the view behind the BSB and most modern translations, and its defenders argue that it lets the evidence speak rather than deciding the result beforehand. A second view, the confessional or Received Text position, holds that God preserved the text in the form the church actually received and used, and that the Greek text printed by Erasmus, Stephanus and Beza in the sixteenth century, which stands behind the KJV, represents it; its defenders appeal to the Westminster Confession (1646), which says that the Scriptures in their original languages were 'by his singular care and providence, kept pure in all ages', and argue that a doctrine of preservation that leaves the text to the shifting judgments of scholars is no doctrine at all. A third view, Byzantine priority, defended by Maurice Robinson and others, holds that the large majority of later Greek manuscripts best preserves the original, because a text copied in the most places for the longest time is most likely to be the one the church received from the start. The Greek Orthodox Church, for its part, reads in its worship an ecclesiastical form of the Byzantine text, the Patriarchal Text of 1904. The Hermeneutics lesson Which Bible Are You Reading gives these views more room. For this lesson notice that all three answer the sceptic in the same way on the central point: the text has been preserved, and the disagreement among them concerns a small number of places.

The New Testament Survey lesson How the Text Reached Us describes the manuscripts and how textual critics compare them. Here is what an apologist needs exactly, because the popular figures are often out of date or overstated, and one wrong number will be used to dismiss everything else you say. About 5,800 Greek manuscripts of the New Testament are catalogued by the Institute for New Testament Textual Research in Münster. That figure counts every catalogued item, from a scrap with a few words to a complete Bible, and includes lectionaries prepared for reading in church. Most of them are medieval, copied between the ninth and the fifteenth centuries. Only a small number, most of them fragments, are dated before AD 200, and those dates are estimates. There are in addition thousands of manuscripts of early translations into Latin, Syriac, Coptic and other languages, and a vast body of quotations in early Christian writers.

Dating ancient manuscripts is mostly done by comparing handwriting, and handwriting gives a range, not a year. The case to know is Papyrus 52, a fragment the size of a palm in the John Rylands Library in Manchester, containing parts of John 18:31-33 on one side and 18:37-38 on the other. C. H. Roberts published it in 1935 and dated it to the first half of the second century, and for decades it was cited as from around AD 125. Brent Nongbri argued in 2005, in the Harvard Theological Review, that the handwriting has close parallels into the later second and even the early third century, so that no narrow date can be justified; Pasquale Orsini and Willy Clarysse, in a 2012 study of early New Testament papyri, placed it between 125 and 175. Say that P52 is one of the earliest surviving New Testament manuscripts and is probably from the second century. Do not say it is from AD 125, as though that were known.

A warning from recent history shows why. In a public debate with Bart Ehrman in 2012, the textual critic Daniel Wallace announced that a fragment of Mark had been dated to the first century. The claim spread through sermons and websites for years. When the fragment was finally published in 2018, as Oxyrhynchus Papyrus 5345, its editors dated it to the late second or early third century, and Wallace publicly apologised for having announced the earlier date. He was a careful scholar who trusted an unpublished report. The lesson for a student is simple: never repeat a manuscript claim that has not been published and examined, however exciting it is.

The comparison with classical literature is real, but state it carefully. Most works of Greek and Latin literature survive in far fewer manuscripts, and the earliest copies of many are separated from their authors by many centuries; for the first six books of Tacitus' Annals, which describe the reign of Tiberius, there is a single manuscript, from the ninth century. So the New Testament is better attested than any other ancient book in number and in nearness of its earliest copies. Two qualifications keep the comparison honest. The figures in popular books for Homer and other authors have often been out of date. And the number of manuscripts proves only that we have abundant evidence for reconstructing the text. It does not prove that what the text says is true. The first is a question for this lesson; the second is a question for the rest of the module.

Bart Ehrman is a leading textual critic, trained under Bruce Metzger at Princeton Theological Seminary, and his popular book Misquoting Jesus (2005) has shaped how many people think about this question. He tells there how his own evangelical faith in an inerrant Bible did not survive his study of the manuscripts. His case deserves to be stated at its strongest, because it is not merely the claim that there are many variants.

It runs like this. First, we do not have the original manuscripts of any book, and the copies we have were made by scribes who were often not professionals, especially in the first two centuries, when Christianity was a small and sometimes persecuted movement. Second, the number of differences among the manuscripts is very large; Ehrman wrote that there are more variations among the manuscripts than there are words in the New Testament, which has around 138,000 words in Greek, and Peter Gurry's careful estimate in 2016 put the number of variants, not counting spelling, at around half a million. Third, and most important, some of the changes were deliberate. Ehrman's earlier academic study, The Orthodox Corruption of Scripture (1993), argued that scribes sometimes altered the text to make it more clearly support the orthodox side in the controversies of their day. Luke 2:33, where the BSB reads that 'The Child's father and mother were amazed', appears in many later manuscripts, and so in the KJV, as Joseph and his mother, perhaps to protect the virgin birth. At Hebrews 2:9, where the BSB reads that 'by the grace of God He might taste death for everyone', a few witnesses read apart from God, and Ehrman argues that this harder reading is original and was changed. In Mark 1:41 the BSB's 'Moved with compassion' stands where one Greek manuscript, Codex Bezae, and a few Old Latin copies read becoming angry, and Ehrman argues that anger is original; the NIV, revised in 2011, now reads that Jesus was indignant. And 1 John 5:7 in the KJV contains a clear statement of the Trinity, the Comma Johanneum, that is found in the text of no Greek manuscript earlier than about the fourteenth century, and in only a handful after that, having come into the Greek from Latin copies; it was printed in the Complutensian Polyglot and, from his third edition of 1522, by Erasmus, and so passed into the Received Text behind the KJV.

Fourth, Ehrman draws a theological conclusion. If God inspired the very words of Scripture, why did he not preserve them? We cannot claim that the words are inspired if we do not know, at every point, what they were. Finally, the earliest period is the least documented. Between the writing of a Gospel and our first substantial copies of it lies a century or more, and we cannot see what happened to the text during that time. That is the case, and it is a serious one from a serious scholar.

The first part of the reply is about what the numbers measure. The number of variants grows with the number of manuscripts: a book that survived in two copies would have very few, and we would have no means of checking them. Most variants are differences of spelling, word order that does not change meaning in Greek, or obvious slips. A small proportion affect meaning, and a smaller proportion still are both meaningful and have a serious claim to be original. Daniel Wallace, Peter J. Williams in Can We Trust the Gospels? (2018) and many others make this case, and it is not only made by apologists: it is the ordinary view of the discipline. Metzger, Ehrman's teacher, held that no Christian doctrine depends on a disputed reading.

The second part concerns the examples. Each of the readings Ehrman discusses is known, printed in the notes of critical editions, and discussed openly in commentaries. The Comma Johanneum is absent from the BSB and every other modern translation. The NIV's choice at Mark 1:41 shows that evangelical translators follow the evidence where it leads, even to a reading a sceptic proposed. The deity of Christ does not rest on any variant: it is stated in passages whose text is not in doubt, such as John 1:1 and John 20:28, and in Paul's application of Old Testament texts about the LORD to Jesus. Some texts that speak of Christ as God do have variants, including John 1:18 and 1 Timothy 3:16, and a student should know this. But the doctrine would stand without them. In an interview printed in the paperback edition of Misquoting Jesus, Ehrman himself said that his position does not actually conflict with Metzger's view that the essential Christian beliefs are not affected by the variants, though he added that some particular beliefs and interpretations do turn on disputed readings. Students should know that he said this; it is one of the most important facts in this debate, and it is seldom mentioned by those who use his book.

The third part concerns the earliest period, which is Ehrman's strongest point and should not be waved away. It is true that we cannot see directly what happened in the first century of copying. But there are good reasons to think the text was not transformed in that period. The manuscripts come from many regions and many lines of copying, so a deliberate change in one place leaves the other reading preserved elsewhere, which is precisely how the changes Ehrman describes were detected. There is no trace in the evidence of a single moment when the text was rewritten. And where early papyri can be compared with later manuscripts, they often agree closely: Papyrus 75, usually dated around AD 200, and Codex Vaticanus, from the fourth century, are so similar in Luke and John that, on the usual dating, careful copying across more than a century is the best explanation. Brent Nongbri has argued that P75 may be later than usually thought, and if he is right the argument loses some of its force, so cite it with that caution.

One further honesty is needed. Many textual critics now speak not of recovering the original text but of the initial text, the earliest form of the text from which the surviving tradition can be traced. The editors of the major new critical edition, the Editio Critica Maior, use that language, and scholars such as Eldon Epp have argued that the very idea of an original text is more complicated than it seems, since some books may have circulated in more than one form from early on. This is not a confession that the text is lost. It is a careful way of saying what the evidence can show. A student should be able to say: we can recover the earliest form of the text with great confidence, and at a small number of places we are honestly unsure which of two readings came first, and our Bibles tell us where those places are.

For many students the charge that the Bible has been changed comes not from a sceptic but from a Muslim neighbour, colleague or relative, and it comes from a very different place. It is not a claim that no book can be trusted. It is a claim that God did send down a book through Jesus, and that it has not come down to us as he gave it. Describe it as an informed Muslim would recognise. The Qur'an speaks of the Tawrat given to Moses, the Zabur given to David and the Injil given to Jesus as genuine revelations from God (for example 3:3-4, 4:163 and 5:46), and it calls Jews and Christians the People of the Book. It also says that some of them distort words from their proper places (4:46, 5:13), conceal the truth (3:71), and write the scripture with their own hands and then say it is from God (2:79). The Arabic word for distortion is tahrif.

Muslim scholars have understood the charge in two main ways. One is tahrif al-ma'na, distortion of meaning: the words of the earlier scriptures remain, but Jews and Christians have misinterpreted them, for instance by reading Jesus' sonship literally. Several early Muslim writers held this. The other is tahrif al-lafz, distortion of the text itself: the words have been changed, added to or lost. The Andalusian scholar Ibn Hazm of Cordoba, who died in 1064, argued this at length, and it is the form most Christians meet today. Modern Muslim speakers add arguments of their own. The four Gospels are, by their own account, written by others about Jesus (Luke 1:1-4 says so plainly), so they cannot be the Injil, the book given to him, and contain at most traces of it. Paul, who never met Jesus in his lifetime, is held to have changed his message into a religion about him. And Western textual critics, Ehrman above all, are cited as confirming from the Christian side that the text has been altered. Ahmed Deedat of South Africa, whose booklets and debates, including Is the Bible God's Word?, circulated through Africa and Asia from the 1980s, made these arguments familiar to millions. A student who has never heard them stated this way will be unprepared for the conversation he is most likely to have.

The Christian reply should be given with the same care. First, it should begin with the real difference between the two faiths' ideas of scripture, rather than talking past it. Christians do not believe that God sent down a book to Jesus. The gospel is the good news about Jesus, 'the beginning of the gospel of Jesus Christ, the Son of God' as Mark 1:1 puts it, and the four Gospels are testimony to it by those who knew him and those who knew them. So the observation that Luke was not Jesus does not trouble Christians; it is what they themselves say. Second, the manuscript evidence matters here more than in the sceptic's case. Codex Sinaiticus and Codex Vaticanus, from the fourth century, and papyri from the third, with smaller fragments probably from the second, contain the Gospels substantially as we read them, centuries before the Qur'an. If the text was corrupted, it was corrupted before that, and across the whole Roman world at once, leaving no trace in any manuscript. Third, several passages of the Qur'an appear to assume that the Christians and Jews of Muhammad's time possessed their scriptures in a form worth consulting: 5:43 speaks of the Torah the Jews have, which contains God's judgment; 5:47 tells the People of the Gospel to judge by what God revealed in it; 10:94 tells the one in doubt to ask those who read the scripture before him. Muslim interpreters read these verses in more than one way, for instance as referring to the parts that remain uncorrupted or as an argument directed at the people's own standards, and a Christian should acknowledge that rather than claim the verses settle the matter.

Fourth, and this is a matter of character as much as argument, do not answer the charge by attacking the Qur'an's own textual history. There are real scholarly questions about the early transmission of the Qur'an, and Muslim scholarship has its own long tradition on the variant readings, but a Christian who reaches for those questions in order to score a point has usually ended the conversation and lost the friend. The Module 1 lesson Gentleness and Respect applies here with full force. The Module 5 lesson Talking With Muslim Friends goes further into the conversation itself, including the question of the crucifixion, and it says plainly what this lesson says now: in some places and families, leaving Islam is dangerous. Never press a friend toward a step that could cost him his safety or his life, and read the Missions lesson The Persecuted Church before you advise anyone who is considering it.

Add the third paragraph to your page headed My defence: Jesus. It answers the question: hasn't the Bible been changed?

First, decide who is most likely to ask you this question where you live: a Muslim neighbour, a sceptical student, a member of another religious group, or someone in your own church who has seen a video. Do not write any person's name; write only the kind of person. Then go back to what you wrote in the inquiry and ask whether it would make sense to that person, given what he actually believes.

Second, do the footnote exercise from the inquiry if you have not: look up Mark 16:8, John 7:52 and 1 John 5:7 with the BSB footnotes on a free site such as Bible Hub. Then write your paragraph, no more than 150 words. It must say what Isaiah 40:8 promises and what it does not; give one exact fact about the manuscripts with its proper caution; name one real variant and say honestly what it does and does not affect; and, if your questioner is Muslim, say plainly how Christians understand the Gospels differently from the Injil. End with an invitation rather than a victory: something like, would you like to read one of the Gospels with me and see for yourself what it says?