In Lecture 15 of my ongoing series of video-lectures on YouTube, I discuss Greek numerals, and describe several textual variants in which numerals are involved, at John 19:14, Luke 24:13, Mark 6:41, Acts 27:37, Luke 10:1 and 10:17, Acts 13:33, and Revelation 13:18.
I also describe the Eusebian Canons, and mention a few points in the text where they indicate what kind of text Eusebius was using when he made them.
(27 minutes 10 seconds)
Before we focus the next two lectures on two major textual variants, there is one small concern that still needs to be covered: textual variants that involve Greek numerals. In Greek, numerals were not always written out in full; in some manuscripts, they were represented by letters of the Greek alphabet that represented specific quantities.
This was not some sort of secret code; this was the ordinary way of writing numerals in Greek. This chart shows the 24 usual letters of the Greek alphabet, expanded by the inclusion of three extra letters stau (or, digamma), koppa, and sampi. Arranged in three rows of 9 letters, you can see the numerical value that was assigned to each letter:
Α = 1 Ι = 10 Ρ = 100
Β = 2 Κ = 20 Σ = 200
Γ = 3 Λ = 30 Τ = 300
Δ = 4 Μ = 40 Υ = 400
Ε = 5 Ν = 50 Φ = 500
Ϝ, ϛ = 6 Ξ = 60 Χ = 600
Ζ = 7 Ο = 70 Ψ = 700
Η = 8 Π = 80 Ω, ω = 800
Θ = 9 Ϙ, Ϟ = 90 ϡ = 900
A horizontal line was added above these letters to show that they were being used as numerals. Using these letters to represent quantities, any sum from 1 to 999 could be written using no more than three letters.
With a mark to the lower left of a letter, it signified an amount of thousands. In manuscripts, large numbers sometimes appear in colophons, or notes, at the end of a book, where they refer to the year in which the manuscript was made. The standard dating-method in colophons was not a calculation of the number of years from the birth of Christ, but a calculation of the number of years from the creation of the world, which was believed to have happened in 5,508 B.C. So if we were to encounter a Greek manuscript with a colophon stating that the manuscript was made in the 6,508th year of the world, we would probably feel justified if we gave it a production-date around the year 1000.
One of the earliest textual variants mentioned by a patristic writer involves the numerals in Mark 15:25 and John 19:14. Mark 15:25 says, “Now it was the third hour when they crucified Him,” that is, about 9:00 in the morning. But in John 19:14, John states that it was “about the sixth hour” when Pilate was yet to deliver a sentence regarding Jesus’ case, before he finally handed Jesus over to be crucified in verse 16.
Some interpreters have reckoned that John used a different method of hour-counting, starting at midnight, whereas for Mark, the day have 12 hours and began at the beginning of hour #1. Thus Pilate could be making his decision at around the sixth hour – 6:00 a.m. – and after he handed Jesus over to be crucified, some time elapsed, during which Jesus was whipped, given a crown of thorns, beaten, and mocked, and was led through the streets of Jerusalem, until, at about the third hour – 9:00 a.m. – He was crucified.
But a different solution was proposed by the early writer Ammonius, whose proposal was later echoed by Eusebius of Caesarea in the early 300s, and by Epiphanius of Salamis in the late 300s, and by Jerome. Ammonius explained that whereas the letter gamma ought to be written, representing the number “three,” so as to refer to the third hour, a copyist wrote the similar-looking letter “gabex,” or digamma, so as to refer instead to the sixth hour.
Epiphanius indicates that Clement of Alexandria and Origen endorsed this solution to the harmonization-problem. It is also attributed to Peter of Alexandria, who was martyred in 311. In Peter of Alexandria’s testimony, preserved in very late manuscripts as part of the Chronicon Paschale, it is stated that in the text that was written by the hand of the evangelist, which is still preserved at Ephesus, and is adored there by the faithful, the reading in John 19:14 is “about the third hour,” and this is the reading in the correct books.
We don’t know if Peter of Alexandria was making an informed statement or not, but this is interesting evidence no matter how you slice it. In a small number of manuscripts, the text in John 19:14 supports the reading “the third hour,” including Codex L, Codex Delta, and minuscule 72. It is significant that in the vast majority of manuscripts, copyists did not give in to the temptation to alter a single letter, or numeral, and thus remove the apparent difficulty.
Another interesting textual variant involving a numeral
occurs in Luke 24:13: how far was the
But some manuscripts, including Codex Sinaiticus and Codex Π, support the reading “160 stadia.” This was also initially the reading in Codex N. It is supported by the Armenian version, by the Palestinian Aramaic version, by the Latin Codex Fuldensis, from the mid-500s, by a small number of Greek minuscules, and by a significant cluster of Arabic manuscripts, and it is endorsed by a margin-note in minuscule 34.
probably reflects a belief that the city of
this restoration of the city, Eusebius of Caesarea advocated the view that
Nicopolis and Emmaus are the same place in his composition Onomasticon. Jerome, who
made a Latin translation of Eusebius’ Onomasticon,
expressed the same view in his Epistle
108 and in his composition Lives of
Illustrious Men. However, 160 stadia
is more than 18 miles. That is a long
distance for two people to cover in an afternoon, walking from
The view that
Nicopolis and Emmaus were synonymous was not exclusive to
Another interesting textual variant involving numerals appears in Mark 6:41. The original text, with very broad support, refers specifically to the five loaves and the two fish. But in Papyrus 45, for some reason, the copyist did not write “five” and he did not write “two” in this verse, even though his text does include the same numbers in verse 38. Possibly in his exemplar, these words were written as numbers, and the lines above the numerals were very short, and he misunderstood them as if they were dots, that is, as if they were marks that meant, “do not write this.”
Codex Vaticanus is another important manuscript with an unusual reading involving a numeral. In Acts 27:37, where most manuscripts state that there were 276 souls on board the ship that was about to be shipwrecked. Codex Vaticanus, however, has “about seventy-six” written out in full. This is also supported by the Sahidic version. What has happened here?
John Burgon perceived the answer: basically, after a transposition of the words in this part of the verse, the number 276, written as a numeral, that is, as Sigma, Omicron, Stau, followed the phrase “in the ship.” The letter omega, and the end of the word for “ship,” ploíw, was misread as if it was part of a word, hōs, meaning, “about,” and this left the overlined letters omicron, 70, and stau, six, creating the reading “about 76.”
Burgon also noted, from a common-sense perspective, “Although one might say, ‘about seventy,’ or ‘about eighty,’ is it not obvious to everyone that ‘about 76’ is an impossible expression?” Fortunately, although Westcott and Hort adopted Vaticanus’ reading, against all other Greek evidence, Burgon’s cogent case against Vaticanus’ reading was favored by later writers, including F. F. Bruce and Bruce Metzger.
Another textual variant that involves numerals is in Luke chapter 10, in verses 1 and 17: does the Lord sent out 70 individuals, or 72? The Byzantine reading, 70, is supported by Codex Alexandrinus, Codex W, and almost all other Greek manuscripts. The Western reading, 72, is supported by Codex Bezae, most of the Old Latin copies, and probably by the Sinaitic Syriac. The Alexandrian witnesses are divided: Sinaiticus and Codex C and Codex L support “70,” but Vaticanus and Papyrus 75 support “72.”
Papyrus 45 is not extant in Luke 10:1, but it is extant in verse 17. Unfortunately, its testimony was misrepresented when the manuscript was first published, and the first printings of the 28th edition of the Nestle-Aland text continued to misrepresent it, as if it supports “72.” This mistake was only recently corrected. In real life, Papyrus 45 supports “70,” as Bruce Metzger has observed: the letter Omicron, representing “70,” is not followed by another numeral, but by an ordinary space-filling mark.
The early writer Tertullian also supports “70,” and he drew a parallel to the numbers in Exodus 15:27, in his composition Against Marcion, at the beginning of chapter 24 of Book 4. Tertullian says that the 12 springs of water at Elim correspond to the 12 apostles, and the 70 palm trees at Elim correspond to the 70 disciples.
The scope of the support for the reading “70” is sufficient to decide the question. “72” probably originated as an allegorical representation of the Gentile nations, as listed in the Septuagint in Genesis chapter 10.
Another textual variation: In Acts 13:33, where almost all Greek manuscripts say that Paul is quoting from the second Psalm, Codex Bezae says that Paul quoted from the first Psalm. Somehow this reading survived to the early 1500s in the early editions of Erasmus’ Greek New Testament, and was featured in William Tyndale’s English translation in 1526. It is indirectly supported by some patristic writers. When Tertullian quoted Psalm 2:7 in Against Marcion, Book 4, chapter 22, he does not describe it as part of the second Psalm; he says that he is quoting from the first Psalm.
The minority-reading in Codex D may echo the influence of an early tradition that what we know as Psalm 1 and Psalm 2 were considered a single Psalm, with what we know as Psalm 1 being a sort of Preface. This tradition seems to have had an effect on how the Latin text of Psalms was arranged in the text used by Tertullian in the late 100s, and by Cyprian in the mid-200s, and even as late as the Venerable Bede in the late 600s and early 700s.
The best-supported Greek reading is clearly “the second Psalm.” Some early translators of Acts into Latin probably were comfortable paraphrasing Paul’s reference so as to adopt the arrangement that he expected his readers to recognize.
Seventh, the most famous textual variant in the New Testament that involves a numeral is without doubt the one that occurs in Revelation 13:18, where the number of the beast is given as 666 in most manuscripts, including Papyrus 47. But in a few important copies, including Codex C and Papyrus 115, the number in Revelation 13:18 is “616,” written as chi-iota-stau.
The early patristic writer Irenaeus made a detailed comment on this passage, in what may be the first mention of a textual variant, in Against Heresies, Book Four, chapters 29-30.
Irenaeus made several guesses about the name that is represented with the numerical value of 666: Euanthas was one guess, Lateinos was another one, and Teitan was another one. This last possibility, Teitan, was the option preferred by Irenaeus, but he emphasized that it was only a guess, stating that if it were necessary for people in his time to know the name, it would have been revealed in John’s vision, instead of just the number of the name.
Irenaeus thus shows that he used the text with “666,” because in each of these names, the value of the letters adds up to a total of six hundred and sixty and six. And as if more evidence were needed, he also compared this number to Noah’s age before the floor (600 years) and the dimensions of Nebuchadnezzar’s idol in Daniel 3:1: 60 cubits high and 6 cubits wide.
In chapter 30 of Book 5 of Against Heresies, Irenaeus’ statements get even more detailed: he affirms that 666 is the number that is found “in all the most approved and ancient copies,” and that it is endorsed by “those men who saw John face to face.” When we consider that when Irenaeus wrote, the book of Revelation was less than 100 years old, this is extremely weighty testimony.
Irenaeus also stated, “I do not know how it is that some have erred following the ordinary mode of speech, and have reduced the middle number in the name, deducting the amount of fifty from it, so that instead of six tens-units, they will have it that there is but one.” To put it another way, Irenaeus refers to approved and ancient copies that support “666,” but he also is aware of copies that have the reading “616.”
He says, “I am inclined to think that this occurred through the fault of the copyists, as it tends to happen, since numbers also are expressed by letters; so that the Greek letter which expresses the number sixty was easily expanded into the letter iota of the Greeks.”
Irenaeus does not say precisely how the letter chi (Xi, Ξ) be accidentally changed by copyists into iota. Iota is a straight vertical line, like the letter “I,” but chi is very different.
seems unlikely that a copyist could accidentally
make 616 out of 666. It may be more
likely that someone believed that John was referring to the concept of “Nero redivivus,” a sort of urban legend
that the Emperor Nero, who died in the year 68, was actually still alive and
would one day return, leading an army from the east. There are some references to this belief in
the composition that is known as the Sibylline
Oracles, and in about the year 420,
Commenting on Second Thessalonians 2:7, Augustine stated that some individuals believe that this verse refers to Nero, whose deeds already seemed to be as the deeds of Antichrist. And thus, he continues, “some suppose that he shall rise again and be Antichrist. Others, again, suppose that he is not even dead, but that he was concealed that he might be supposed to have been killed, and that he now lives in concealment, in the vigor of the same age which he had reached when he was believed to have perished, and will live until he is revealed in his own time and restored to him kingdom.”
It is unlikely that anyone encountering the text of Revelation in Greek would stray from the reading “666,” which fits a pattern in which the Antichrist mimics the true Christ; the name “Jesus” in Greek has a numerical value of 888. But someone encountering the text in some other language might look for an alternative explanation. If one writes “Neron Caesar” in Hebrew consonants, their value adds up to 666. If one drops the Hebrew letter nun, so as to correspond to a Latin form of Nero’s name, the name’s value thus decreases by 50, yielding the value of 616.
This is a somewhat complicated theory. But it might be how the reading “616” was created – via an interpretation that the Antichrist – either literally or thematically or typologically or some other way – was expected to be the Emperor Nero.
Finally, a consideration of numerals in Greek New Testament manuscripts would be incomplete without a description of the Eusebian Canons and Sections. Technically, the Eusebian Canons and Sections are part of the para-text, or meta-text – not part of the text itself. They are a guide to cross-references in the Gospels.
At the beginning of many manuscripts of the Gospels, instead of jumping right into the text, and even before a chapter-list appears, there is a composition called “Ad Carpianus,” which is Eusebius’ brief explanation of how to use his cross-reference system for the Gospels. In a few manuscripts this material is presented within a fairly unusual frame, shaped like a quatrefoil, or a symetrical rounded cross. The Eusebian Canons and Ad Carpianus are included in the 27th edition of the Nestle-Aland compilation.
Eusebius began by mentioning that he got the idea for a cross-reference system for the Gospels from Ammonius the Alexandrian, who had arranged the text of the Gospel of Matthew with the parallel-passages from the other Gospels alongside it. Eusebius wanted to keep each Gospel-account intact, and so instead of dividing up the texts of Mark, Luke, and John, he gave each pericopé its own number, and then made a ten-part chart, in which the parallel-passage were listed, by their numbers, side by side.
are ten parts to this list:
● The first one contains the list of passages for which there are parallels in Matthew Mark, Luke, and John.
● The second one lists passages, or sections, for which there are parallels in Matthew, Mark, and Luke.
● The third one lists passages for which there are parallels in Matthew, Luke, and John.
● The fourth lists passages for which there are parallels in Matthew, Mark, and John.
● The fifth lists passages for which there are parallels in Matthew and Luke.
● The sixth lists passages for which there are parallels in Matthew and Mark.
● The seventh lists passages for which there are parallels in Matthew and John.
● The eighth lists passages for which there are parallels in Mark and Luke.
● The ninth lists passages for which there are parallels in Luke and John.
● The tenth lists passages that do not have parallels, but which are unique in each Gospel.
The same numbers are written in the margin alongside each passage. Accompanying these numbers, called the Section-numbers, is a Canon-number, written in red, which identifies the list, one through ten, in which the passage is found. If you see a number from 1-10 in the margin written in red below the Section-number, you will know which list to consult to find the number of the passage.
So, if you open a Gospels-book to any passage, and want to see what the other Gospel-writers wrote about the same event, then after you find the Canon-number, written in red, you can consult that list, and see the numbers of the parallel-passages in the other Gospels. Then by finding those numbers in the margins in those Gospels, you can read the parallel-passages themselves.
After this introductory guide, the Canon-Tables themselves occupy several pages. These can be very plain, or in some cases spectacularly ornate, with complex colorful golden designs, and paintings of animals, birds, and other decorations in the margins. In some cases the artistic effort that was given to the Eusebian Canons resulted in the theft of these pages, as works of ark. The tradition of decorating the Eusebian Canons is abundantly shown not only in Greek manuscripts but also in Latin, Ethiopic, and, especially, Armenian manuscripts.
Although Eusebius got the idea for a cross-reference system for the Gospels from the earlier writer Ammonius of Alexandria, he clearly did not closely follow Ammonius’ Matthew-centered system. As John Burgon pointed out in 1871, in a detailed Appendix to his book about the last 12 verses of Mark, Canon 8 and Canon 9 cannot have been part of a Matthew-centered cross-reference system. In addition, when it is noticed that Mark has 21 unique sections, Luke has 72 unique sections, John has 97 unique sections, and 24 sections are shared by Mark and Luke, and 21 sections are shared by Luke and John, this makes a total of 225 sections which have no parallel in Matthew and thus could not be part of a Matthew-centered cross-reference system.
The Eusebian Canons also have an impact on the testimony of Eusebius regarding the last 12 verses of Mark. Eusebius is often quoted as if he said, in the composition Ad Marinum, that Mark 16:9-20 was absent from almost all manuscripts, but in real life, his statement is much more nuanced: he wrote that that was one of several things that something that someone might say about the passage. Eusebius himself instructed Marinus to retain the passage, and gave instructions about how Mark 16:9 was to be read, with a pause between “Rising” and “early on the first day of the week.” And further along in the same composition, Eusebius quoted from Mark 16:9. So, when he wrote Ad Marinum, Eusebius appears to favor the inclusion of Mark 16:9-20.
But according to a note that appears in some members of the textual cluster known as family-1, Mark 16:9-20 is not included in the Eusebian Canons. Specifically, in codices 1 and 1582, although Eusebian Section-numbers appear in the margin alongside verses 9-20, these two manuscripts, along with the manuscripts 205, 2886, and 209, have a prominent note before Mark 16:9, which states, “In some copies, the Gospel comes to a close here, and so do the Canons of Eusebius of Pamphilus. But in many, this also appears.”
So it is possible that at some point after advising Marinus to keep Mark 16:9-20, Eusebius might have changed his mind, and decided not to include these verses in the text upon which he based the Eusebian Canons.
The Eusebian Canons occasionally have text-critical significance where they testify to the presence or absence of other specific passages. For example, Luke 22:43-44 is not in Papyrus 75 or Codex Vaticanus or Codex Alexandrinus or Codex W, but in the 100s, Justin and Irenaeus both refer to the passage. In the Eusebian Canons Luke 22:43-44 is included as Section #283, implying that it was in the text that was used by Eusebius.
Mark 15:28 is not in Codex Vaticanus, or Sinaiticus, or Codex D, and is also missing in over 100 minuscules – but it is listed as Section #216 in the Eusebian Canons.
And, by not featuring an entry for Matthew 27:49 and John 19:34 in Canon Nine, Eusebius shows that his text did not contain a parallel-passage between those two passages. In Codex Vaticanus and in Codex Sinaiticus, Matthew 27:49 is expanded so as to create a parallel between those two verses; the Alexandrian Text of Matthew 27:49 says that before Jesus died, someone came and pierced Him in the side with a spear, and blood and water flowed from the wound.
By not including a reference to this reading in his cross-reference system, where it would have belonged in Canon 7, Eusebius shows that his manuscripts did not have this reading. This is a very strong indication that Eusebius did not supervise the production of Codex Vaticanus and Codex Sinaiticus.
Finally, although a little anecdote about a detail in the early Christian composition The Epistle of Barnabas is not directly related to the New Testament, it might illustrate the kind of figurative interpretations some early Christians could give to some numerals. The Epistle of Barnabas was probably written in the early second century. It appears in Codex Sinaiticus, after the book of Revelation.
its ninth chapter, the author refers to Genesis 14:14, emphasizing the exact
number of the men under Abraham’s command who went to rescue