This series has followed the Bible from its first writing, through generations of scribes, to the manuscripts and printed editions behind modern translations. This last article asks the question that lies behind all the others: how confident can we be that the Bible we read today says what its authors wrote?

There are two easy answers, and both are wrong. One says that the text has been so corrupted by copyists that we cannot know what the originals said. The other says that the text has come down without any change at all. The evidence supports neither. The honest answer is more nuanced, and also more reassuring than many people expect.

The size of the evidence

Start with the New Testament. The Institute for New Testament Textual Research in Münster has catalogued more than 5,800 Greek manuscripts: about 140 papyri, some 320 majuscules (parchment manuscripts in capital letters), nearly 2,900 minuscules (in a small cursive hand) and around 2,500 lectionaries. Many are fragments; some are complete New Testaments. To these we add thousands of manuscripts of early translations, especially Latin, and an enormous number of quotations in early Christian writers.

That is far more than for any other work of ancient Greek or Latin literature. For many classical authors, the earliest complete manuscripts were copied hundreds of years after the author’s death, and the number of manuscripts is in the dozens or hundreds. Comparisons are sometimes exaggerated in popular books, with outdated figures for other works, and recent scholarship has corrected some of them; the volume Myths and Mistakes in New Testament Textual Criticism (2019) is a good guide. But the basic point stands: no other ancient book is attested by so many manuscripts, so early.

For the Old Testament the numbers are smaller and the evidence is different. We have thousands of medieval Hebrew manuscripts, the great Masoretic codices of around 1000 CE, more than two hundred biblical manuscripts among the Dead Sea Scrolls from a thousand years earlier, and ancient translations, above all the Greek Septuagint.

Half a million variants?

With so many manuscripts, it is no surprise that they differ. A variant is any place where manuscripts do not agree. How many are there in the New Testament?

Peter Gurry, in a careful study published in 2016, estimated that the Greek manuscripts contain around 500,000 variants, not counting differences in spelling. That is more than the number of words in the New Testament, which has about 138,000.

That figure sounds alarming, and it is sometimes presented that way. But it needs to be understood. The number of variants is large because the number of manuscripts is large. Every manuscript that is copied adds a few new slips of its own. If we had only ten manuscripts, we would have far fewer variants, and far less certainty. The more copies we have, the more variants we count, and the easier it is to see which reading is original.

What kinds of variants?

Scholars such as Daniel Wallace sort variants into groups.

  • Spelling and nonsense errors. By far the largest group. A word spelled differently, a letter dropped, a slip that produces no real word. These are easy to recognise and do not affect meaning.
  • Changes that do not affect translation. Greek can vary word order without changing the meaning, so “Christ Jesus” and “Jesus Christ,” or a verb before or after its subject, often make no difference in English. Greek also sometimes uses the article before a name (“the Jesus”), which is not translated.
  • Meaningful but not viable. A difference that would change the meaning, but is found only in one late manuscript or a small group, with no real chance of being original.
  • Meaningful and viable. A difference that affects the meaning and has a real claim to be original. This group is a small fraction of the total, often estimated at well under one percent.

It is this last group that matters, and that good translations mark in their footnotes.

The passages that really are uncertain

Some of the most important variants concern whole passages that appear in the King James Version but are absent from the earliest manuscripts.

  • Mark 16:9–20. The earliest complete manuscripts of Mark, Codex Sinaiticus and Codex Vaticanus, end at 16:8. Some early writers say that the most accurate copies ended there. Other manuscripts have a short ending, a long ending (16:9–20), or both. Most scholars think the long ending was added in the second century. Most modern Bibles print it with a note or in brackets.
  • John 7:53–8:11, the woman caught in adultery. Absent from the earliest Greek manuscripts and early commentators on John; in some later manuscripts it appears in other places, including after Luke 21:38. Many scholars think it is an authentic early story about Jesus that circulated independently before being added to John.
  • 1 John 5:7–8, the Comma Johanneum. The words about the Father, the Word and the Spirit bearing witness in heaven are found in no Greek manuscript before the late medieval period, and entered printed Greek texts through Erasmus’s third edition (see From manuscripts to printed Bibles). Almost all scholars agree they are not original.

Other well-known examples include Luke 22:43–44 (the angel and the sweat like drops of blood), Luke 23:34 (“Father, forgive them”), Acts 8:37 (the eunuch’s confession) and John 5:4 (the angel stirring the water). Within verses, there are variants like “the one and only God” or “the one and only Son” in John 1:18 (discussed in One and only Son), and “we have peace” or “let us have peace” in Romans 5:1.

These are real questions, and textual critics debate them seriously. But it is important to notice how we know about them: they are not hidden. They are printed in the footnotes of almost every modern Bible.

Does any teaching depend on them?

Bart Ehrman, whose Misquoting Jesus (2005) brought textual criticism to a wide audience and emphasised the scale of the variation, agreed in an appendix to its paperback edition with his teacher Bruce Metzger that essential Christian beliefs are not affected by textual variants in the manuscript tradition. Ehrman argues that some variants affect how particular passages should be interpreted, which is true. But the major teachings of the Christian faith, such as the deity of Christ, his death and resurrection, salvation by grace, and the Trinity, are taught in many passages whose text is not in doubt. Removing the Comma Johanneum, for example, removes one verse that was used in debates about the Trinity, but the doctrine was formulated in the fourth century by theologians who did not have that verse in their Bibles.

Daniel Wallace and others who disagree with Ehrman on other matters reach the same conclusion on this one. Scholars of very different convictions agree that the text of the New Testament is known with a high degree of confidence, and that the places of real uncertainty are known and limited.

The Old Testament

The Old Testament presents a different picture. Its transmission within Judaism, especially from the first century CE onwards, was extraordinarily stable, as the comparison of the Dead Sea Scrolls with the medieval Masoretic manuscripts shows (see The Great Isaiah Scroll). At the same time, the scrolls and the Septuagint show that before that period, some books existed in more than one edition, and that the Masoretic text, for all its care, preserves some early errors (see What the Dead Sea Scrolls show).

The result is that Old Testament translators more often face difficult textual decisions than New Testament translators do, especially in books like Samuel, and in places where a word seems to have been damaged early. But the overwhelming majority of the text is not in doubt, and the alternatives that are, are known and can be weighed.

Confidence, not certainty

What, then, should we say?

We do not have the original manuscripts of any biblical book, and we will not find them. We cannot claim that every letter in a printed Bible is exactly what the author wrote. In a small number of places, scholars are genuinely uncertain, and in a few of those places the decision affects the meaning of a sentence.

But we can say this. The Bible’s text is attested by more manuscripts, closer in time to the originals, than any other ancient writing. The work of scribes, from the Masoretes to the copyists of the early church, was in the main careful and devoted. Their mistakes follow patterns that can be recognised. Over three centuries of study, scholars of every background have reconstructed a text that, in the great majority of its words, is not in dispute. And where it is uncertain, we know where the uncertainty lies.

That is not blind trust. It is confidence based on evidence. A translation should honour that by being open about the evidence, and that is what Open Word Bible tries to do: every sentence carries its reasoning, including any textual decision behind it, so that readers can see not only what the text says, but how we know.

Sources

  • Ehrman, Bart D. Misquoting Jesus: The Story Behind Who Changed the Bible and Why. San Francisco: HarperSanFrancisco, 2005. Paperback edition with appendix, 2007.
  • Gurry, Peter J. “The Number of Variants in the Greek New Testament: A Proposed Estimate.” New Testament Studies 62 (2016): 97–121.
  • Hixson, Elijah, and Peter J. Gurry, eds. Myths and Mistakes in New Testament Textual Criticism. Downers Grove, IL: IVP Academic, 2019.
  • Metzger, Bruce M. A Textual Commentary on the Greek New Testament. 2nd ed. Stuttgart: Deutsche Bibelgesellschaft, 1994.
  • Stewart, Robert B., ed. The Reliability of the New Testament: Bart D. Ehrman and Daniel B. Wallace in Dialogue. Minneapolis: Fortress, 2011.
  • Tov, Emanuel. Textual Criticism of the Hebrew Bible. 3rd ed. Minneapolis: Fortress, 2012.
  • Wallace, Daniel B., ed. Revisiting the Corruption of the New Testament: Manuscript, Patristic, and Apocryphal Evidence. Grand Rapids: Kregel, 2011.
  • Institute for New Testament Textual Research, Kurzgefasste Liste: ntvmr.uni-muenster.de/liste.
  • Bible quotations are from the World English Bible (public domain), except where the article gives its own literal rendering.