The Text of Scripture

A Little History

In AD 1453 the fall of Constantinople to the twenty-one-year-old Sultan Mehmed of the Ottoman Empire brought an end to the Byzantine (i.e., Eastern Roman) Empire which had stood for 1,000 years. Greek Orthodox monks fled westward into Europe bringing with them Greek New Testament manuscripts. At the very same time in Germany, the invention of the printing press by Gutenberg made the mass production of books possible. In 1516, the first printed Greek New Testament, published by Erasmus, called into question the Latin Vulgate (i.e., ‘common version’), which had stood for 1,000 years as the Bible of the Roman Catholic church. Just one year later, in 1517, came the Reformation.

Erasmus’ edition of the Greek New Testament was a runaway bestseller. Indeed, it was said that ‘Erasmus laid the egg that Luther hatched’. It was used as the basis for his German Bible, and later by King James’ translators in 1611. Erasmus’ text, reprinted with variations over the next century, was advertised in 1633 as ‘the text received by all’, the Textus Receptus (TR).

Many more Greek manuscripts became available in the following 400 years, including the fifth century Codex Alexandrinus, offered to King James by Cyril Lucar, the Reformation-sympathetic Orthodox Patriarch of Constantinople. In the nineteenth century, two fourth-century manuscripts, Sinaiticus and Vaticanus, were discovered. In the twentieth century, even earlier fragmentary Greek New Testament papyri were unearthed in Egypt, some dating from the second century AD.

These early manuscripts showed many differences from Erasmus’ TR, and after two centuries of scholarly attempts at revision, in 1881 two Cambridge professors, B. F. Westcott and F. J. A. Hort, produced an edition of the Greek New Testament relying heavily on early manuscripts, with about 6,000 differences from the TR. This became the basis of the English Revised Version of 1885.

In 1947, the Dead Sea Scrolls were discovered, dating from the second century BC to the first century AD. Although 1,000 years older than previously available Hebrew Old Testament manuscripts, they were very similar to, and confirmed the reliability of, the standard Hebrew Masoretic text.

While differing little textually in the Old Testament, modern Bibles fall largely into two camps: those whose New Testament is translated from the TR (KJV, NKJV), and those based on the modern NU (Nestle-Aland/United Bible Society) texts, which are similar to Westcott and Hort’s ???? text (RV, NASB, NRSV, NIV, ESV, NLT).

Two Extreme Reactions

Differences between Greek New Testament manuscripts have provoked two extreme reactions. First, atheists argue that we cannot trust the text of scripture because (a) we do not possess the original documents, (b) the manuscript copies contain too many variant readings, and (c) scribes intentionally changed it. However, while we do not possess the original autographs (i.e., written by the apostles themselves), and no two manuscript copies we possess are identical, the atheist conclusion is baseless. On the contrary, we possess an ‘embarrassment of riches’, over 5,500 Greek New Testament manuscripts, some very early (from the second century, AD), in addition to over 15,000 copies of ancient ‘versions’ (i.e., translations into Latin, Syriac, Coptic, Armenian, etc.) and scripture quotations in early Church Fathers’ writings. We can be certain that the original readings have been preserved among this evidence. Additionally, most variants are inconsequential, e.g., spelling differences, and the vast majority of scribes were faithful copyists. No doctrines of scripture are endangered by manuscript difference because doctrines are repeated in many verses where all manuscripts agree.

At the other extreme, some Christians insist that Erasmus’ Textus Receptus (TR), or the KJV which was based on it, is the original New Testament. However, Erasmus’ TR had many problems. In a rush to beat another edition into print, Erasmus himself described it as ‘precipitated rather than edited’. Its hundreds of typographical errors were corrected over five editions. It was also based on a very small number of late Greek manuscripts that Erasmus had on hand in Basle, dating from the twelfth century onwards. In Revelation, he had only one manuscript, and Erasmus resorted to back-translating the missing last six verses from the Latin into Greek! As a result, KJV has ‘book of life’ instead of ‘tree of life’, Rev. 22. 19, an error found in no Greek manuscript.

Erasmus was a Roman Catholic priest (and the illegitimate son of a priest)1 who never left the Church during the reformation. He produced the TR, in part, for fame and money, which was why the TR was such a hurried job. Erasmus included some verses from the Latin, found in very few Greek manuscripts,2 which have no real claim to be part of the original Greek. The inspired apostolic authors wrote in Greek, not in King James Version English (or Latin), and we must base our Bible on the best Greek manuscript evidence, not vice versa.

Problems with the Modern NU Greek Text

Most textual scholars today favour the NU text, underlying modern Bible versions, for two reasons, first, because it is based on the earliest available manuscripts, and second, because of scribal habits. That is, scholars have believed that scribes tended to add to, and relieve difficulties in, the text as they copied. Preferring ‘shorter readings’, the NU text is about 2,500 words shorter than the TR, and omits many words or verses found in the KJV.3

Preferring ‘more difficult readings’, the NU text also adopts certain readings precisely because they make less sense. As a result, modern Bibles contain readings that are jarring for believers: Christ being angry (or indignant) with a leper, instead of compassionate, Mark ?. ??, NIV ???? update; Matthew being mistaken about Christ’s ancestors’ names, Matt. ?. ?, ??, ESV; the Lord saying He was not going to Jerusalem, John 7. 8, NASB, which He did in verse 10; or all anger being sinful, Matt. 5. 22 RV instead of anger ‘without a cause’ – although the RV margin does recognize that many ancient authorities insert ‘without a cause’.

The NU is a shorter and more difficult text than the TR. However, recent academic research, including the author’s at Cambridge, shows that scribes tended to omit rather than add at a rate of sixty to forty.4 This is only common sense – omission is the easiest mistake to make. The author’s study also showed that scribes created more difficulties rather than improving the text. Scribes are the (unintentional) corrupters, not improvers(!), of God’s word.

Some commentaries which follow the NU text speak of the so-called ‘best’ manuscripts, but this is now admitted by textual scholars to involve circular reasoning – the ‘best’ manuscripts are ones which contain the ‘best’ readings, but the ‘best’ readings are those found in the ‘best’ manuscripts. Avoiding this circularity, studies in scribal habits show that these ‘best manuscripts’ contain many errors. Another problem with so-called ‘earliest and best’ manuscripts, Sinaiticus and Vaticanus, is that they disagree with each other over 3,000 times in the Gospels alone, ignoring minor spelling variants.5 Simple maths means that one or the other is wrong, at least, 3,000 times – thirty times per chapter. Few people copying out a chapter of an English Bible today would make as many mistakes. The signature errors of these manuscripts are excessive omissions, and more difficult readings.

The NU text is largely Alexandrian, that is, based on manuscripts related to the early Coptic (Egyptian) versions and quotes from Alexandrian Church Fathers. It preserves an early, but likely localized, form of text. The NU text is also often based on very few manuscripts, and in a few places, none at all – the editors abandon all manuscripts and conjecture, i.e., guess, the original reading, Acts 16. 12; 2 Pet. 3. 10.

The Majority (M) Text

The text of the majority of over 5,000 Greek manuscripts was dismissed by Westcott and Hort as corrupt, the result of a third-century official revision. This theory is now abandoned by textual scholars, and some champion the Majority Text as the original text. It is similar to the TR,6 and comes from the heartland of the Greek-speaking church (from Antioch through to Greece, where most New Testament Epistles were originally located), with manuscripts dating from the fifth to the fifteenth centuries.

However, for all its numerical superiority, the Majority Text is not perfect. It has some bizarre features: it moves the Roman doxology from chapter 16 verses 25 to 27 to chapter 14 verse 23; in Revelation chapter ? verse ? the angels say ‘Holy’ nine times; and some other verses, e.g., James chapter 2 verse 18, and 1 Peter chapter 1 verse 8, are virtually nonsensical. Occasionally, all the early manuscripts and all geographically separate ancient versions oppose the Majority Text reading; we should reject the majority here and follow the early manuscripts.

The Old Testament also shows us that the majority of later manuscripts are not always right; the ‘majority (Masoretic) text’ is sometimes not used in New Testament quotations. In Hebrews chapter 1 verse 6, we read, ‘let all the angels of God worship him’. We search in vain for the source of this quote in the KJV, which is based on the Hebrew Masoretic Text, where this phrase (Deut. 32. 43) is missing. It is present in the Dead Sea Scrolls and the Greek Septuagint. The inspired New Testament thus proves that the Masoretic majority text is sometimes wrong.

A balanced approach

Textual scholarship has swung from one extreme to another, from one eccentric, late and very long text based on few manuscripts (Erasmus’ TR) to another early, eccentric, localized, very short text based on few manuscripts [the NU text].

Avoiding extreme positions involves adopting balanced principles: (1) Antiquity: preferring earlier readings; (2) Quantity: avoiding readings based on few manuscripts; (3) Propinquity: preferring readings with broader geographical attestation among manuscripts, versions and ‘fathers’; (4) Explanatory: avoiding readings better explained as scribal errors, i.e., omissions and harder readings; (5) Theology: prayerfully considering what the inspired authors were more likely to have written, depending upon divine illumination.

What Bible should we read? I suggest a TR-based Bible, like the KJV or NKJV (which have fuller and more sensible texts). Textual criticism is a complex task and for those not able to evaluate the textual information found in a modern Greek New Testament, the author recommends the NKJV marginal notes. These show the readings of the three main texts, the TR, the Majority Text (M) and the NU text. While the Majority Text is generally preferable, we should be wary of readings based on only one of the three texts but take readings from a number of these sources for a balanced view of scripture.

In conclusion, let us not forget that scripture is not an end in itself irrespective of the text we use. It is a means of enjoying fellowship with God and our Saviour Jesus Christ.



See Wikipedia or


For example, Luke 17. 36; Acts 8. 37; 9. 5b-6a; 15. 34, and, most famously, 1 John 5. 7, 8, which is only found in four late Greek MSS from the 14th century onwards, as well as in six other manuscripts, in marginal notes written after the 16th century.


For example, Matt. 17. 21; 18. 11; 23. 14; Mark 11. 26; 15. 28; Luke 23. 17. Two longer passages are marked as doubtful: Mark 16. 9-20 and John 7. 53 – 8. 11.


Andrew Wilson, ‘Scribal Habits in Greek New Testament Manuscripts’, Filologia Neotestamentaria, 24 [2011], pp. 95-126.


H. C. Hoskier, Codex B and its Allies, Bernard Quaritch, 1914, Vol 2, pg. 1.


The Majority Text differs from the TR in 1,838 places. Professor Daniel Wallace, The Majority Text and the Original Text: Are They Identical?


Your Basket

Your Basket Is Empty