The Lord’s Prayer is one of the most treasured passages in all of Scripture and has been throughout all the ages since the New Testament was written. It is a text so familiar that even hearing it recited in slightly different words can seem almost “wrong.” So, when we find not only different wording but, in fact, variants in the text of this passage between manuscripts and translations, it can seem a bit unsettling, even if those variants do not really affect the meaning. One of the clearest examples of this is found in the last verse of Matthew’s version of the prayer, a doxology which reads:

“And lead us not into temptation, but deliver us from evil: For thine is the kingdom, and the power, and the glory, for ever. Amen,” (Matthew 6:13, KJV).

The final clause, however, “For thine is the kingdom, and the power, and the glory, for ever. Amen,” is actually absent in many modern translations, such as the ESV, NIV, CSB, NET, and others. It is present in the NASB and HCSB, but is marked off in brackets and noted to be lacking in the earliest manuscripts. Why would this beautiful and familiar ending to the Lord’s model prayer be lacking? If we can lay our emotions aside and examine the evidence reasonably, there is actually a good reason for it.

The Lord’s Prayer and the manuscript evidence

The clause in question here is lacking in Codex ‭א and Codex B (both early fourth century), Codex D (fifth century), Uncial 0170 (late fifth or early sixth century), Codex Z (sixth century), and other Greek manuscripts and lectionaries on through the middle ages, even on up to the 16th century. Thus, the earliest Greek witnesses do not have the clause, and there is a continued witness to its absence throughout church history. The vast majority of the Old Latin manuscripts also lack the clause, including all the earliest Latin copies such as it^a (fourth century), it^b and it^h (fifth century), and others. The Latin Vulgate lacks it as well. There are also Coptic manuscripts without the clause.

What is also important to note is that this closing doxology is absent in the earliest commentaries on the Lord’s prayer, most of which not only lack the clause but explicitly identify “but deliver us from evil,” as the conclusion of the prayer. These commentators include Tertullian (On Prayer, Chapter 8)¹Ante Nicene Fathers, Volume 3, pg 684, Origen (Prayer, Chapter 18, section 2)²John J. O’meara, Ancient Christian Writers: Volume 19 (Paulist Press, 1954) 65, Cyprian (Treatise on the Lord’s Prayer, Section 27)³Ante Nicene Fathers, Volume 5, pg 454-455, Cyril of Jerusalem (On the Mysteries, Lecture 5, Section 18)⁴Nicene and Post Nicene Fathers, Volume 7, pg 156, Ambrose of Milan (The Sacraments, Book 5, Chapter 4)⁵Roy J. Deferrari, The Fathers of the Church: Volume 44 (The Catholic University of America Press, 1963) 314, Augustine of Hippo (Our Lord’s Sermon on the Mount, Book 2, Chapter 4)⁶Nicene and Post Nicene Fathers, Volume 6, pg 38, John Cassian, (The First Conference of Abbot Isaac, Chapter 23)⁷Nicene and Post Nicene Fathers, Volume 11, pg 395 Peter Chysologus (Sermon to the Catechumens on the Lord’s Prayer),⁸George E. Canss, Fathers of the Church: Volume 17 (The Catholic University of America Press, 1953) 115-119 and others.

Thus, when English translations of the Bible began, the John Wycliffe Bible (the first complete Bible in the English language) did not have the clause. Neither, in fact, did William Tyndale’s New Testament, even though it was a translation from earlier editions of the TR, the same Greek tradition from which the KJV came!

As to evidence supporting the reading, the full clause is present in Codex W (late fourth/early fifth century), Codex Σ (sixth century), and in John Chrysostom’s Homily on the prayer⁹Nicene and Post Nicene Fathers, Volume 10, pg 136 (fourth century), but is otherwise absent from all Greek sources until about the 8th or 9th century, though it does dominate the majority of later Greek manuscripts in the medieval era.

Other forms of the clause also occur in various older manuscripts and translations. One fifth-century Latin copy, it^k, contains a much shorter form that simply reads “thine is the glory forever.” Several Coptic manuscripts contain a version that lacks the words “the kingdom, and.” The earliest Syriac manuscript to contain this passage (the fifth-century Curetonian gospels) contains a form that lacks “and the power,” as does at least one later Greek minuscule. A handful of Greek copies add a Trinitarian formula in the middle, stating that the glory is “of the Father and the Son and Holy Spirit.” Conversely, Lectionary l¹⁰¹⁶ lacks “and the glory,” entirely, while minuscule 1252 lacks “and the power and the glory.” Minuscule 2148, on the other hand, expands “forever” out into “forever and ever.” Having one or two of these kinds of variants would be normal and even expected, but this large cluster of different versions of the clause raises some red flags.

If not original, from where did it come?

Of course, for all the evidence against it, the question still remains: if this phrase wasn’t originally part of the gospel, where did it come from? Why would a scribe add something like this into the text? There are a number of possibilities.

The Didache

A shorter but similar clause is also found in a very ancient Christian document known as the Didache (Chapter 8), which concludes the Lord’s Prayer with “for yours is the power and the glory forever.”¹⁰Michael W. Holmes, The Apostolic Fathers in English (Baker Publishing Group, 2006) 167 This section of the Didache is preserved only in one manuscript dated to 1056 AD,¹¹ibid, 160 however, since the form of the clause here notably differs from that found in manuscripts of Matthew, it seems less likely that the clause was a later addition here. Is the Didache, then, evidence that the clause is original? Not exactly.

Not only does the version of the doxology in the Didache differ from that in Matthew, but it is also not unique. The Didache is largely a book of liturgical instruction on regular ritual practice for early churches, and this is only one of the prayers it prescribes. It offers other model prayers as well which have no parallel in the gospels. All of these prescribed prayers have similar clauses appended to the end of them, such as “to you be the glory forever,¹²Ibid, 167 and “for yours is the glory and the power through Jesus Christ forever.”¹³Ibid, 168 Thus, it is actually quite likely that the author(s) of the Didache added this common concluding doxology to the Lord’s model prayer just as he/they did to the other prayers offered. The Didache was well-known and highly regarded by many in the ancient churches. It may well be that, through the Didache, this doxology became widely used in corporate church worship and in private devotional life, thus becoming more and more familiar, making it more and more likely that a scribe might habitually conclude the prayer that way in his copy without even thinking about it. Since regional variation on the doxology would be likely, it is unsurprising that the manuscript tradition displays various versions of the doxology creeping in at different times and in different places. Thus, the Didache provides a quite plausible origin for this reading.

The Diatessaron

A less likely but still possible contributor is Tatian’s Diatessaron. Late in the second century. a man named Tatian produced a harmony of the four gospels into one single narrative called the “Diatessaron.” The doxology may have been present in Tatian’s Diatessaron (Chapter 9). It is, at least, found in the much later Arabic translation of Tatian’s work,¹⁴Ante Nicene Fathers, Volume 9, pg 58 though this does not make it certain that it was part of Tatian’s original, especially since copies in other ancient translations lack the clause. Even if it is original to the Diatessaron, Tatian sometimes added extrabiblical details, such as his testimony of a great heavenly light shining forth at Jesus’ baptism. Thus, even if one favors the idea that Tatian originally included the clause (a very tentative conclusion at best), that would not prove it original. He could well have inserted it as one of his interpretive expansions, adopted it from the Didache, or brought it in habitually from the same common liturgical practice from which the Didache derived it. The Diatessaron is known to have influenced some scribes in their copying, and thus this could help further explain why some later scribes may have added in the wording. At any rate, the Arabic copies of the Diatessaron are not considered especially reliable, thus the claim that the reading was originally in the Diatessaron at all rests on very shaky ground.

Elsewhere in Scripture

Let’s say the Didache did record an early Christian practice of ending prayers with such doxologies, as it certainly seems to have done. Still, where did this practice itself come from? Bruce Metzger speculates that Christians may have derived such lofty conclusions to their prayers from King David’s prayer in 1 Chronicles 29 and that this is thus the origin of the doxology added to the Lord’s prayer:¹⁵Bruce Metzger, A Textual Commentary on the Greek New Testament: Second Edition (United Bible Societies, 1994) 14

“Thine, O Lord is the greatness, and the power, and the glory, and the victory, and the majesty: for all that is in the heaven and in the earth is thine; thine is the kingdom, O Lord, and thou art exalted as head above all,” (1 Chronicles 29:11).

Likewise, some scholars have speculated that the words were drawn from Paul’s words in 2 Timothy:

“And the Lord shall deliver me from every evil work, and will preserve me unto his heavenly kingdom: to whom be glory for ever and ever. Amen,” (2 Timothy 4:18).

This conclusion is drawn, not only due to the verbal parallels, but also the connection in themes between Paul’s words here and the words of the Lord’s prayer.¹⁶Roy Hammerling, The Lord’s Prayer in the Early Church (Palgrace Macmillian, 2010) 20 Perhaps none of these hypotheses are correct. What they show us, though, is that there is a very reasonable explanation of how this passage might have been expanded through liturgical tradition. Based on the overwhelming evidence of the passage’s absence in the early manuscripts, translations, and commentaries, something like this seems the most likely explanation for the KJV’s form of the text and its many other variations in the manuscript tradition.

Luke 11 and the Lord’s Prayer

Luke 11 also contains a form of Lord’s prayer. Unlike Matthew’s version, which is largely the same in all translations besides the issue of the final clause, Luke’s version is a bit more complicated. A much shorter form is found in most modern translations and a longer form in translations like the KJV which reads pretty much just like Matthew:

“And he said unto them, When ye pray, say, Our Father which art in heaven, Hallowed be thy name. Thy kingdom come. Thy will be done, as in heaven, so in earth. Give us day by day our daily bread. And forgive us our sins; for we also forgive every one that is indebted to us. And lead us not into temptation; but deliver us from evil,” (Luke 11:2-4, KJV).

“And He said to them, “When you pray, say: ‘Father, hallowed be Your name. Your kingdom come. Give us each day our daily bread. And forgive us our sins, For we ourselves also forgive everyone who is indebted to us. And lead us not into temptation,’” (Luke:11:2-4, NASB).

Unlike Matthew 6:13, modern translators’ position here is not quite so solid. Not only are the vast majority of manuscripts and ancient translations on the side of the long version found in the KJV, but a number of copies that are quite old also affirm that reading, going back as far as the fourth and fifth centuries. The case for the longer reading in Luke 11 is much stronger than that for the doxology of Matthew 6:13, and should not be written off lightly.

The short form is found in only a very small number of sources, but that’s not the full picture. It is found in P⁷⁵, an extremely early manuscript from the late second or early third century. It is likewise what we find in Origen’s third-century commentary on the Lord’s Prayer.¹⁷John J. O’meara, Ancient Christian Writers: Volume 19 (Paulist Press, 1954) 65 Additionally, it is the reading in Codex א and Codex B (early fourth century). Thus, while it is by far the minority reading, it is the universal reading of all the very earliest sources, and they go back quite far! It’s also found in a handful of later Greek manuscripts and is the reading in much of the Latin manuscript tradition.

It is also noteworthy that, again, there are a number of other “intermediate” versions in various manuscripts that are longer than the short version but still lack material found in the long version. Conversely, there are also expanded versions that are longer than the KJV’s version. A KJV Only advocate, or even a majority text scholar, might argue that the language of Luke 11 is such that it lends itself to these kinds of scribal deletions, but the problem is that we don’t see parallel trouble in Matthew 6, where the language is so similar to the longer version of Luke 11. Why would scribes struggle so much copying the one and not the other? Thus, modern scholars observe at least three major factors:

1. We have extremely early sources that unanimously agree on the shorter reading
2. There are numerous alternate “longer” forms of Luke’s version which are unparalleled in Matthew’s version
3.  It is far more likely that scribes were trying to harmonize Luke with Matthew (or with liturgical tradition) than that they would delete all these words.

Fully considered, the modern case is much stronger than it first appears. But there is still another factor to consider. If the longer reading of Luke is original, and thus Luke’s version was originally pretty much identical to Matthew’s, then Luke’s version of the prayer is actually further evidence that the doxology was not originally part of Matthew’s gospel! Even Luke would be in agreement with all our other early sources that Jesus’ model prayer ended with “lead us not into temptation; but deliver us from evil.” Thus, if modern scholars are wrong on Luke 11 (and they may well be), it only further demonstrates them to be right about Matthew 6:13. Either way one falls on Luke’s version, it is not an argument in favor of King James Onlyism.