Luke 24:47 and the ESV

          In Luke 24:46-47, according to the ESV, as Jesus commissioned His disciples to preach the gospel, He gave the church its agenda for evangelism:  He commanded them to regard ethnic differences as no barrier to brotherhood, and to invite all people to surrender to God in light of the reconciliation brought about by Jesus Christ:  “Thus it is written, that the Christ should suffer and on the third day rise from the dead, and that repentance and forgiveness of sins should be proclaimed in His name to all nations, beginning from Jerusalem.”

          At least, that’s what the ESV said in 2001, and in several editions that were released since then.  Now in 2016, Crossway is releasing a “Permanent Text” edition of the ESV – and it has introduced a few changes in the text.  One of them is in Luke 24:47:  instead of referring to “repentance and forgiveness of sins,” Jesus now refers to “repentance for forgiveness of sins.”  This is not a matter of translation-methods; the change in English reflects a decision to favor the Alexandrian reading εις (for) instead of the reading that is found in all other transmission-streams, and in almost all Greek manuscripts:  και (and). 

          A recent announcement at Crossway’s website listed the changes to the text, and announced that this is how all copies of the ESV are going to be from now on.  This is the “Permanent Text of the ESV Bible, unchanged forever, in perpetuity.”  The problem, as far as Luke 24:47 is concerned, is that the ESV’s Oversight Committee’s decision to prefer εις (for) instead of και (and) is incorrect. 

          Three ancient Greek manuscripts support εις:  Papyrus 75, Sinaiticus, and Vaticanus.  But, as Daniel Mace (a researcher and Bible translator in the 1700’s) noted, no manuscript is as old as common sense.  The scholars who produced the ESV did not rigidly follow these witnesses.  In Luke 23:34, P75 and Vaticanus do not have Jesus’ statement, “Father, forgive them, for they know not what they do,” but the ESV has these words.  In Luke 24:32, P75 and B do not have the words εν ημιν (“within us”) but the ESV does.  In Luke 23:29 and 24:49, P75 does not have the word ιδου (“behold”), but the ESV does.   

          Meanwhile, support for και is found in the writings of Cyprian (in the mid-200’s) and Eusebius (early 300’s), and in the ancient codices Alexandrinus, Ephraemi Rescriptus, Bezae, and Washingtonensis, plus over a thousand later manuscripts.  Και is also supported by the Old Latin version which predates the 380’s, as well as by the Vulgate, the Sinaitic Syriac manuscript, the Armenian version (made in the early 400’s), the Ethiopic version, and the Palestinian Aramaic version, and in Augustine’s Sermon 2291.  The diversity of these witnesses is impressive. 

          Secondary support for εις is found in the Peshitta version, and in five Byzantine manuscripts, which Wieland Willker identifies as minuscules 1253, 1519, 2445, 2796, and 2808.  To a great extent, where the Byzantine Text disagrees with the Alexandrian flagship-manuscripts Vaticanus and Sinaiticus, these five witnesses favor the Byzantine reading.  In genetic terms, these five manuscripts are not close relatives of P75, Vaticanus, and Sinaiticus. 

          In a manuscript that contains line after line of Byzantine text, if the text suddenly and momentarily deviates from the usual Byzantine text, we are probably not seeing the sudden intrusion of a secondary exemplar, as if a copyist possessed a second exemplar but only used it once.  Instead, we are observing a scribal error.  When the same rare reading that appears sporadically in Byzantine witnesses also appears in genetically unrelated witnesses, it suggests that the same scribal error has independently recurred.  To express this as a text-critical canon:  Agreement among witnesses which lack a genealogical connection is likely the result of coincidental independent error.

From 2001-2015, the ESV said "and" in Luke 24:47,
as shown on page 236 of the ESV Reader's Gospels.

          That is what we are looking at in these five Byzantine manuscripts.  The mechanism that caused their copyists to deviate from the Byzantine text also caused an early copyist to create the reading found in P75, Vaticanus, and Sinaiticus.  The mechanism is simple:  harmonization.  

          If it were possible to introduce the compilers of the Nestle-Aland text to the variant-unit in Luke 24:47 and describe only the variants, without saying which witnesses supported which reading, the verdict would be clear:  “repentance for forgiveness of sins” is a harmonization to the phrase in Mark 1:4 and Luke 3:3, where the authors refer to “baptism of repentance for [that is, εις] forgiveness of sins.”  The word “for” (εις) also precedes “forgiveness of sins” in Matthew 26:28 and Acts 2:38.  These passages are far more likely to have been recalled by copyists than Acts 5:31, where “and” (και) precedes “forgiveness of sins” but the words “to Israel” also appear between the reference to “repentance” and the reference to “forgiveness of sins.”  Compared to the passages in the Gospels, Acts 5:31 is not much of a parallel.  (Bruce Terry misquoted Acts 5:31 in his attempt to defend the decision of the UBS4 Committee.)  

          So, while I am glad that the ESV Permanent Text edition did not make any drastic changes to the text, this particular last-minute alteration in Luke 24:47 is incorrect; the reading και should have been retained.  The Gideons Edition of the ESV remains the best available edition of the ESV.

(I note, in passing, that the UBS Greek New Testament in 1966 had και in the text, the 1984 NIV had “and” in the text, and Michael Holmes also adopted και in the 2010 SBL-GNT.)

Scripture quotations are from The ESV Bible® (The Holy Bible, English Standard Version®), copyright © 2001 by Crossway, a publishing ministry of Good News Publishers.  Used by permission.  All rights reserved. 

Hand to Hand Combat: Codex Sinaiticus vs. Minuscule 4

          Codex Sinaiticus returns to the combat-ring today!  Facing this heavyweight is a relatively obscure challenger:  minuscule 4.  This Gospels-manuscript from the 1200’s is so little-known that it seems appropriate to offer some information about its background (or possible background) before the contest begins.

          In the 1430’s, a clergyman named John of Ragusa (a place known today as the coastal region around Dubrovnik, Croatia) took part in church councils at Basel (in Switzerland) and Florence (in Italy) as a representative of the Dominican monastic order.  The top item on the agenda of these councils was reconciliation with the Orthodox Church based in Constantinople.  In the course of his work, John of Ragusa served as an emissary to Constantinople in 1435-1437.  The leaders of the Orthodox Church sent representatives – including Basilios Bessarion, who united with the Roman Catholic church and was made a cardinal in 1439 – to the Council of Florence, where, after much debate, a formal statement of unity was drawn up – only to be rejected later by Orthodox clerics in Constantinople.

          On May 29, 1453, Constantinople fell to Islamic invaders.  Something else had happened ten years earlier which, as it turned out, was also significant:  when John of Ragusa died in 1443, he bequeathed a small collection of manuscripts – possibly souvenirs from his visit to Constantinople, or perhaps gifts from Bessarion – to the Dominican convent at Basel.  This collection included not only some classical works but also a few manuscripts of parts of the New Testament.

          Seventy-two years later, in 1515-1516, Desiderius Erasmus visited Basel in order to use the manuscripts known to be in the possession of the Dominican monks there.  With the help of those manuscripts, along with his vast knowledge of patristic material and other data amassed in his previous research (plus the research-notes of Lorenzo Valla), Erasmus produced the first published printed edition of the Greek New Testament.

          Minuscule 4 – presently kept at the National Library of France as Greek MS 84 (formerly Regius MS 2867) – is either one of those manuscripts which John of Ragusa gave to the monks at Basel, or else it came from some other source.  There seems to be some confusion about whether or not 4 was ever at Basel, and whether or not it was used by Erasmus.  But in any event, readings from this manuscript were known to Robert Stephanus; this manuscript was denoted by the Greek letter γʹ (gamma) in the notations in Stephanus’ 1550 edition of the Greek New Testament.  Minuscule 4 helped define the Textus Receptus.     

          Now let’s get down to business.  Two pages of minuscule 4 conveniently contain, more or less, John 6:65-71 and 7:1-16.  These 23 verses constitute the textual battleground of today’s contest.  Here are the contest’s rules, applied to the text in both manuscripts:

          ● The text of the first hand of each manuscript will be compared to the text of NA27.  Words in brackets in the text of NA27 are considered part of the text.

          ● Transpositions that do not involve any gain or loss of words are mentioned, but are not included in the final totals.

          ● The introduction of one non-original letter = 1 point.

          ● The loss of an original letter = 1 point.

          ● Abbreviations of sacred names, abbreviations of και, and easily deciphered contractions are not counted as variants.

          ● If a sacred name is absent, all its letters are considered absent, even if it is very probable that it would have been abbreviated in the manuscript.

             

Here are the deviations from the text of NA27 in 4.  I have marked each reading in 4 that agrees with the Byzantine Text (RP2005) with a triangle, except for transpositions.

6:65 – 4 has ηρηκα instead of ειρηκα.  (+1, -2)

6:65 – 4 has ει instead of η (+2, -1)

6:65 – 4 has μου (+3) ▲

6:66 – 4 does not have the second εκ (-2) ▲

6:66 – 4 has a transposition

6:67 – 4 has θελεται instead of θελετε (+2, -1)

6:68 – 4 has ουν (+3) ▲

6:69 – 4 has (in abbreviated form) Χριστος ο υιος instead of Αγιος (before του Θυ) (+12, -7)  [Note:  one could plausibly reduce this to +9, -4, since both readings, when uncontracted, share the letters ιος at the end.  But I did not.] ▲

6:69 – 4 has (after του Θυ) του ζωητος (+9) ▲

6:70 – 4 does not have ο Ιησους (-7)   

6:71 – 4 has ελεγε instead of ελεγεν (-1)

6:71 – 4 has Ισκαριωτην instead of Ισκαριωτου (+2, -2) ▲

6:71 – 4 has a transposition

6:71 – 4 has ων (after εις) (+2) ▲

7:1 – 4 has a transposition

7:2 – 4 has no differences

7:3 – 4 has has θεωρησουσι instead of θεωρησουσιν (-1)

7:4 – 4 has a transposition

7:4 – 4 has another transposition

7:5 – 4 has επιστευων instead of επιστευον (+1, -1)

7:6 – 4 has no differences

7:7 – 4 has no differences

7:8 – 4 is missing εγω ουκ αναβαινω εις την εορτην (-26) 

7:8 – 4 has a transposition

7:8 – 4 adds ο before transposed εμος (+1) ▲

7:9 – 4 has αυτοις instead of αυτος (+1) ▲

7:10 – 4 has a transposition

7:10 – 4 has αλλ instead of αλλα (-1) ▲

7:11 – 4 has ελεγων instead of ελεγον (+1, -1)

7:12 – 4 has a transposition

7:12 – 4 does not have δε (-2) ▲

7:12 – 4 has ελεγων instead of ελεγον (+1, -1)

7:13 – 4 has μεντι instead of μεντοι (-1)

7:14 – 4 has ο before Ις (+1) ▲

7:15 – 4 has Και before εθαυμαζον (+3) ▲

7:15 – 4 does not have ουν (-3) ▲

7:15 – 4 has ουτως instead of ουτος (+1, -1)

7:16 – 4 has no differences

Thus in John 6:65-7:16, compared to NA27, minuscule 4 contains 45 non-original letters, and has lost 61 original letters, for a total of 106 letters’ worth of corruption.  (In addition, there are six transpositions in this passage in 4.)

Now let’s look at the deviations from NA27 in the same passage in Codex Sinaiticus (in the “Western” portion of the manuscript):

6:65 – א has ουδις instead of ουδεις (-1)

6:65 – א has εμε instead of με (+1)

6:65 – א does not have αυτω (-4)

6:66 – א has ουν after τουτου (+3)

6:67 – א does not have εκ after πολλοι (-2)

6:67 – א does not have αυτου after μαθητων (-5)

6:67 – א has υμις instead of υμεις (-1)

6:67 – א has θελεται instead of θελετε (+2, -1)

6:68 – א has no differences

6:69 – א has ημις instead of ημεις (-1)

6:70 – א does not has αυτοις ο (-7)

6:70 – א has και ειπεν (+8)

6:70 – א has ουχι instead of ουκ (+2, -1)

6:70 – א has a transposition

6:70 – א does not have τους (-4)

6:70 – א does not have εις (-3)

6:71 – א does not have τον (-3)

6:71 – א has απο (+3)

6:71 – א has Καρυωτου instead of Ισκαριωτου (+1, -3)

6:71 – א has και after γαρ (+3)

6:71 – א has εμελλον instead of εμελλεν (+1, -1)

6:71 – א has a transposition

6:71 – א has ων after εις (+2)

7:1 – א has αποκτιναι instead of αποκτειναι (-1)

7:2 – א has no differences

7:3 – א has a transposition

7:3 – א has θεωρουσιν instead of θεωρησουσιν (-2)

7:3 – א has a transposition

7:4 – א has ουδις instead of ουδεις (-1)

7:4 – א has ποιων instead of ποιει (+2, -2)

7:4 – א does not have και (-3)

7:4 – א has ζητι instead of ζητει (-1)

7:5 – א has no differences

7:6 – א does not have ουν (-3)

7:6 – א does not have ο before Ις (-1)

7:7 – א has a transposition

7:7 – א does not have εγω (-3)

7:7 – א does not have περι αυτου (-9)

7:7 – א has πονιρα instead of πονηρα (+1, -1)

7:8 – א has αναβηται instead of αναβητε (+2, -1)

7:8 – א has ταυτην before εγω (+6)

7:9 – א does not have δε (-2)

7:9 – א has εμινεν instead of εμεινεν (-1)

7:10 – א has αλλ instead of αλλα (-1)

7:10 – א does not have ως (-2)

7:11 – א has εκινος instead of εκεινος (-1)

7:12 – א has a transposition

7:12 – א has τω οχλω instead of τοις οχλοις (+2, -6)

7:12 – א does not have δε (-2) 

7:13 – א has ουδις instead of ουδεις (-1)

7:13 – א has a transposition

7:14 – א has ερτης instead of εορτης (-1)

7:15 – א has no differences

7:16 – א does not have ο before Ις (-1)

Thus in John 6:65-7:16, compared to NA27, Codex Sinaiticus contains 39 non-original letters, and has lost 83 original letters, for a total of 122 letters’ worth of corruption.  (In addition, there are seven transpositions in this passage in Codex Sinaiticus.)

          Now we all know what we have been told about the manuscripts upon which the Textus Receptus was based:   they were “The feeblest of manuscript resources” and “Late medieval manuscripts of inferior quality” and so forth.  But this collides with what we see in John 6:65-7:16, where minuscule 4 has less corruption than Codex Sinaiticus.  (This calculation is made, remember, using the NA27 compilation as the standard for comparison.  If the Byzantine Text were used as the standard of comparison instead, then 4 would only have 52 letters’ worth of corruption in this passage (half of which would be in 7:8), and Codex Sinaiticus would still have over twice that amount.)

          If the rate of corruption were even in all transmission-streams, then we could expect late manuscripts to be more corrupt than early ones.  But direct evaluations of sample-passages show that the rates of corruption in different transmission-streams were not the same.  About 260 years after the Gospel of John was written, Codex Sinaiticus had more corruptions in John 6:65-7:16 than minuscule 4 had at the end of a 1,100-year-long transmission-stream.

          To sum up:  even though the copyist of 4 made a major mistake in John 7:8, the text of 4 still has less corruption in John 6:65-7:16 than Codex Sinaiticus.  To be precise:  4 has 106 letters’ worth of corruption (45 non-original letters gained; 61 original letters lost – and six transpositions) and Codex Sinaiticus has 122 letters’ worth of corruption (39 non-original letters gained; 83 original letters lost – and seven transpositions).  Once again the famous heavyweight has lost to a younger opponent.

          (A tangential note:  Papyrus 66, Papyrus 75, and Codex Vaticanus read ουπω in John 7:8, thus agreeing with the usual Byzantine reading, and disagreeing with Codex Sinaiticus.)

          [Readers are invited to double-check the data and the math in this post.]

Why Codex Sinaiticus Doesn't Say What Its Website Says It Says

The website of Codex Sinaiticus
has been online since 2009.

How important is Codex Sinaiticus as a witness to the text of the New Testament?  So important that it has its own website, which features an English translation of portions of the manuscript’s contents, including the entire New Testament.

Or does it?  The home page of the website gives visitors no reason to reach any other conclusion about the English translation – but if one pokes around a bit, searching for information about the translation, this note can be found:

  

“English translation

The Codex Sinaiticus Project was primarily a conservation, digitisation, transcription and publication project.  It did not aim to undertake a new English translation of the writings preserved in the manuscript.”

Henry Tompkins Anderson
1812-1872

The note continues:  “The English translation of the New Testament part of Codex Sinaiticus used on the website was taken from the translation by Henry Tompkins Anderson in The New Testament: translated from the Sinaitic manuscript discovered by Constantine Tischendorf at Mt Sinai (Cincinnati, 1918). This out-of-copyright translation does not provide a literal translation of the text in Codex Sinaiticus, but was included in the website to serve as a navigational aid.”

In other words:  when you read the English translation of the New Testament books at the Codex Sinaiticus website, you are not really reading an English translation of the contents of Codex Sinaiticus!  You are reading a translation made by Henry Tompkins Anderson, who was a Christian preacher in Harrodsburg, Kentucky in the 1860’s – a translation which at many points does not represent the text of Codex Sinaiticus.  The only way to know whether or not the English translation really conveys what Codex Sinaiticus says is to compare the translation to the Greek text itself.   

          Here are fifteen of the many disagreements between the translation at the Codex Sinaiticus website and the text of the manuscript.  

● Matthew 5:19:  Whoever therefore shall make void one of the least of these commandments, and shall teach men so, shall be called least in the kingdom of the heavens; but whoever shall do and teach, he shall be called great in the kingdom of the heavens.

          In Codex Sinaiticus, the copyist skipped the Greek words represented by “but whoever shall do and teach, he shall be called great in the kingdom of the heavens.”  This blunder was caused when a copyist’s line of sight drifted ahead to similar letters; the technical term for the recurrence of similar letters or words, eliciting such mistakes, is homoeoteleuton (“same ending”).  

● Matthew 7:27:  “And the rain descended, and the floods came, and the winds blew, and they beat upon that house, and it fell; and great was its fall.”

          The copyist of Codex Sinaiticus skipped the Greek words represented by “and the winds blew,” another mistake elicited by homoeoteleuton.

● Matthew 8:13:  “And Jesus said to the centurion:  Go; as thou hast believed, be it done for thee. And the servant was restored to health in that hour.”

          The text written by the copyist of Codex Sinaiticus has a bit more text which says, “And the centurion, returning to his house in that hour, found his servant in good health.”   This is a very bad attempt at harmonization, loosely based on the parallel in Luke 7:10.  (Also, considering that this verse ends the standard Gospels-lection to be read annually on the fourth Sunday after Pentecost, the expansion may be an early attempt to round off the anecdote with a flourish.)

● Matthew 12:13:  “Then he says to the man: Stretch forth thy hand. And he stretched it forth; and it was restored to soundness as the other.”

          Codex Sinaiticus’ copyist did not write the Greek words that mean “as the other.”

● Matthew 12:47:  “And some one said to him: Behold, thy mother and thy brothers stand with out, seeking to speak to thee.”

          In Codex Sinaiticus, this verse is missing.  This appears to have been elicited by another case of homoeoteleuton; verses 46 and 47 end with the same word.  

● Matthew 23:35:  “that there may come on you all the righteous blood shed on the earth, from the blood of Abel the righteous to the blood of Zachariah son of Barachiah, whom you murdered between the temple and the altar.”

          The text of Codex Sinaiticus does not have the Greek equivalent of the words “son of Barachiah.”  This could be another example of parablepsis (the technical term for the “sight detours” that sometimes occurred when a copyist lost his line of sight, especially due to homoeoteleuton or to homoeoarcton (same beginning, or similar beginnings, of words).  But it may also be a copyist’s attempt to remove a perceived difficulty from the text, if the copyist understood this passage to refer to the events in Second Chronicles 24:20-22, and believed that that this passage in Matthew referred to Zechariah the priest – who was the son of Jehoida, rather than the son of Barachiah – instead of the prophet Zechariah.

● Matthew 27:48-50:  “And immediately one of them ran and took a sponge, and having filled it with vinegar and put it on a reed, gave it to him to drink.  But the rest said: Wait, let us see if Elijah is coming to save him.  But Jesus again cried out with a loud voice, and gave up the spirit.”

          In Codex Sinaiticus (and in Codex Vaticanus) there is additional material after verse 49.  Codex Sinaiticus says, “But another took a spear and pierced his side, and water and blood flowed forth.”  Now I invite those apologists who commonly claim that textual variants do not have an impact on significant Christian doctrine to explain how, if this variant were to be adopted into the text, the doctrine of inerrancy would survive.  It is obvious that this reading in Codex Sinaiticus says that Jesus was pierced before He died, and it is equally obvious that John 19:30-37 says that Jesus was pierced after He died.  Let it be noted that there is more Greek manuscript support for the inclusion of this sentence than there is for the non-inclusion of Mark 16:9-20.  (Codex Sinaiticus also lacks the Greek equivalent of the words “of them” in verse 48.)   

● Mark 7:4a:  “And when they come from market, unless they immerse themselves, they eat not.”

          In Codex Sinaiticus (and in Codex Vaticanus), the Greek word βαπτισωνται, which refers to washing by immersion in water, has been replaced by ραντισωνται, which refers to washing by the pouring of water.    

● Luke 8:47:  “And the woman, seeing that she had not escaped notice, came trembling, and having fallen before him told before all the people for what cause she had touched him, and how she had been immediately restored to health.”

          The Greek text in Codex Sinaiticus is much shorter, omitting all the words in bold print, so that the text simply says, “And falling before Him, she told before all the people how she had been immediately restored to health.”

● John 4:19-20:  “The woman says to him:  Sir, I perceive that thou art a prophet.  Our fathers worshipped in this mountain; and you say that in Jerusalem is the place where we ought to worship.”

          The copyist of Codex Sinaiticus did not write Κύριε (the Greek word for “Lord,” rendered as “Sir” in this context) in verse 19, and also did not write ὁ τόπος (“the place”) in verse 20.  

● John 9:37-39:  “Jesus said to him:  Thou hast both seen him, and he it is that talks with thee.  He said, Lord, I believe; and he worshiped him.  And Jesus said:  For judgment have I come into this world, that those that see not may see, and that those that see may become blind.”

          The copyist of Codex Sinaiticus did not write the bold-print words in verses 38 and 39.  These words are also missing in Papyrus 75 and Codex W – apparently the result of an early Alexandrian copyist’s misunderstanding of an exemplar in which the passage was marked for liturgical use.

● John 10:40:  “And he went away again beyond the Jordan into the place where John was first baptizing, and abode there.”

          Codex Sinaiticus’ copyist did not write the Greek words εις τον τόπον.  There is nothing to elicit an accidental mistake here.  This reading is probably a deliberate omission by someone who considered these words superfluous.

● First Corinthians 13:3:  “And though I give all my goods away in food, and though I deliver up my body that I may be burned, but have not love, I am profited nothing.”

          Instead of the usual reading καυθήσωμαι, Codex Sinaiticus (and Papyrus 46 and Codex Vaticanus) reads καυχήσωμαι, which means, “that I may boast.” 

● Second Peter 1:1:  “Simon Peter, a servant and apostle of Jesus Christ, to those that have obtained equally precious faith with us in the righteousness of our God and Saviour Jesus Christ.”

          This verse is sometimes used as proof that the author affirmed the deity of Christ, particularly when the Granville Sharp rule is applied to it.  However, with or without the Granville Sharp rule, Codex Sinaiticus does not support the deity of Christ in this verse:  instead of the name-abbreviation for the word “God” (Θυ, that is, Θεου), Codex Sinaiticus’ copyist wrote the name-abbreviation for the word “Lord” (Κυ, that is, Κυριου).  I doubt that a modern translation with this reading could easily escape the charge that it was produced by Arians.

● Jude verse 3a:  Beloved, giving all diligence to write to you concerning this common salvation, I found it necessary to write to you.”   

          The actual reading of Codex Sinaiticus is different:  instead of referring only to salvation, the text of Codex Sinaiticus refers to “our common salvation and life.”  This is a remarkable reading, because it is a conflation, or combination, of two other readings:  σωτηρίας (salvation, the reading found in most manuscripts) and ζωης (life, the reading found in a group of medieval manuscripts known as family 2138, also called the Harklean Group because the text in these manuscripts frequently agrees with the Harklean Syriac version).  Although the Greek manuscripts in this group are not particularly ancient, the reading of Codex Sinaiticus in this verse suggests that ancestor-manuscripts with a text of the General Epistles similar to what is attested by the members of family 2138 – minuscule manuscripts such as 1505, 1611, 2138, and 2412 – existed prior to the production of Codex Sinaiticus.

          As you can see, as a representation of the contents of Codex Sinaiticus, the translation made by H. T. Anderson is highly inaccurate.  If the Codex Sinaiticus website is going to continue using Anderson’s translation, its creators should take whatever steps are necessary to ensure that visitors cannot use Anderson’s translation without being informed that it is provided for navigation-purposes only and that it deviates frequently from the actual text of the manuscript.  There is no good reason to hide this important explanation on a secondary webpage.  (Another option, of course, is to adjust the translation so as to accurately conform to the contents of the manuscript.)

Trusting the Bible - A Response to John Piper

John Piper

          Earlier this week, the following question was addressed at the Desiring God website/podcast:  “How can I trust the Bible if there have been so many add-ins, such as Mark 16:9–20 and John 7:53–8:11 and First John 5:7–8?”

          John Piper, in his response, not only took for granted that all three of those passages are spurious, but he also linked the validity of his response to the validity of his view that all three passages should be rejected:  “If there is a science that can spot these three texts that he mentioned as not part of the original biblical manuscripts, then that same science, in the same way, can perform the same function for all the other passages. There is the answer.”

          It is not that simple.  When I look at the methods used by some textual critics to present their cases against Mark 16:9-20 and John 7:53-8:11, I don’t see a lot of detached, disinterested balance in how the evidence is handled.  Instead, authors often mold and manipulate the evidence so as to produce one-sided propaganda, and important evidence for the opposite position never gets mentioned.  As an example, just look in the NET Bible’s notes about John 7:53-8:11 and try to find any mention of the testimony of Jerome and Augustine about that passage.  Look in the ESV Global Study Bible’s note about Mark 16:9-20 and try to find any mention of Irenaeus’ specific quotation of Mark 16:19.  (Irenaeus wrote in the 170’s/180’s, over a century before the production-date of the earliest existing manuscript of Mark 16, so his testimony is extremely important.)   

          But Piper’s answer would still be problematic even if commentators and Bible annotators were not seasoning their evidence-presentations with gallons of bias in favor of the Nestle-Aland compilation.  It is problematic, first, because although one of the basic axioms of textual criticism is that manuscripts should be weighed, rather than merely counted, the first thing that Piper did, in his reply, was to count manuscripts: 

          “Here is the reason we may have strong confidence that the science of textual criticism is successful in discerning the original wording of the manuscripts:  There are over 5,800 Greek manuscripts.”

          If it were merely a matter of favoring textual variants that are supported by over 80% of the manuscripts, then Mark 16:9-20 would be universally accepted (with support from over 99.8% of the Greek manuscripts), and so would John 7:53-8:11 (with support from 85% of the Greek manuscripts).  Clearly, the quantity of manuscript-support is not very important to the textual critics who reject those two passages.  In the Nestle-Aland compilation, two readings are in the text even though they have zero support among Greek manuscripts.

          When the textual critics who made the Nestle-Aland compilation sat down to do their work, they did indeed have access to the readings of thousands of manuscripts to consult and compare.  But they ignored most of them.  Most Greek manuscripts closely agree with one another, and this large collection of manuscripts, displaying the Byzantine Text, has been treated as a very large family that descended from a single ancestor-manuscript, and for that reason, their collective weight is regarded as the weight of that ancestor-manuscript.  This is why, in the apparatus of the UBS Greek New Testament (the text of which is the same as the text in the Nestle-Aland compilation), these hundreds of manuscripts are not listed individually; they are given a single emblem, “Byz.” 

          Meanwhile, the flagship-manuscripts of the Alexandrian Text – Vaticanus and Sinaiticus – are consistently listed individually, along with whatever allies for them can be found.  And with astounding consistency, when a textual contest occurs between a reading attested by a smattering of Alexandrian manuscripts, and a reading attested by over a thousand Byzantine manuscripts, the compilers of the Nestle-Aland text preferred the Alexandrian reading.    

          In other words, the text-critical method that is used by the editors of the compilation that John Piper uses (i.e., the NT base-text of the ESV) does not support his position:  the textual critics do indeed have thousands of manuscripts to consider, but their method routinely operates on the premise that thousands of manuscripts can contain a spurious reading where only a few contain the original text. 

          Daniel Wallace, as John Piper mentioned, has claimed that “New Testament scholars face an embarrassment of riches compared to the data of classical Greek and Latin scholars have to contend with.”  However, when one examines Wallace’s own approach to the text, it becomes obvious that he believes that in some passages – at the end of Mark, and at Mark 1:41, for example – almost every coin in the royal treasury is counterfeit. 

          When the medieval Byzantine manuscripts are treated as descendants of a single early edition of the text, numbers mean nothing, and thus the claim that the New Testament has “1,000 times the manuscript data” as the average Greco-Roman author, while true, is pointless.  The Nestle-Aland compilers favor earlier manuscripts, especially early Alexandrian manuscripts.  Maurice Robinson provides a sobering observation:  “Even if the text-critical evidence is extended through the eighth century, there would be only 424 documents, mostly fragmentary.  In contrast to this meager total, the oft-repeated apologetic appeal to the value and restorative significance of the 5000+ remaining Greek NT MSS becomes an idle boast when those manuscripts are not utilized to restore the original text.” 

          If John Piper used a text that he considered to be historically authenticated by virtue of the massive support of its contents in Greek manuscripts, then he could confidently assure his listeners that there is no likelihood of significant changes in its contents:  the readings that are supported by the overwhelming majority of Greek manuscripts today will be supported by the overwhelming majority of Greek manuscripts until Judgment Day.  However, as Bill Mounce has confirmed, John Piper uses a text that he considers to be critically authenticated through the analyses of the Nestle-Aland compilation-committee (which presently includes David Trobisch, a member of the secularist Center for Inquiry). 

          John Piper should not convey that the Nestle-Aland compilers will never change their minds about significant, translation-impacting textual variants.  Some of the thirty-two changes introduced in the 28th edition of the Nestle-Aland compilation have an impact on translation.  (Textual changes in the 28th edition were limited to the General Epistles – the rest of the text is the same as it was in the 27th edition.)  

          The introduction of a conjectural emendation into the text of Second Peter 3:10 reverses the meaning of the sentence.  And consider the contest between ὀλίγως and ὄντως in Second Peter 2:18.  In 1966, ὀλίγως was adopted and the UBS committee gave it a “C” ranking, meaning, according to the UBS Introduction, that “There is considerable degree of doubt whether the text or the apparatus contains the superior reading.”  Metzger’s Textual Commentary, however, stated, “ὄντως is far more likely to be secondary than ὀλίγως.”   By the time the fourth edition of the UBS Greek New Testament was printed, the editors were more confident; the reading ὀλίγως was ranked as an “A” reading – meaning, according to the Introduction of UBS4, “the text is certain.”  But in NA28, that was thrown out the window:  ὄντως appears in the text.   

          The Nestle-Aland compilation (and English translations that depend on it, such as the ESV) is not as stable as John Piper wants it to be.  He wants to assure his listener that the text of the New Testament is not going to drastically change.  Yet he says that 7% of the New Testament’s text is in question – subject to revision in the event of new discoveries or a change in the compilers’ views. 

          John Piper also wants the passages that constitute that unstable 7% to be inconsequential.  This seems about as realistic as saying that if a person has 100 dollars, and knows that 93 dollars are genuine, but he is not sure about the remaining seven dollars, he has nothing to worry about; he is free to claim that he has 100 dollars, and may use every dollar as if it is genuine without hesitation.  

          There are some important textual contests within that 7%.  (By the way, I do not accept the claim that this proportion accurately represents the amount of instability in the New Testament text – but I overlook this detail in the interest of brevity.)  Their outcomes are not likely to erase sentences from the Apostles’ Creed, but they definitely have an impact on how one interprets the passage in which they occur.  Consider:

          ● The first half of Luke 23:34 says, “And Jesus said, ‘Father, forgive them, for they know not what they do.’”  John Piper used these words as Scripture in a sermon he preached on January 27, 2002.  (Perhaps he was convinced of their genuineness by “peculiar glory” emanating from them.  More about that in a moment.)  But in NA-27, these words are double-bracketed, just like Mark 16:9-20 and John 7:53-8:11.  The UBS4 Introduction says that these double-bracketed passages “are known not to be a part of the original text.”  That is not trivial.    

         ●  It is difficult to see how the doctrine of inerrancy would survive if the compilers of the Nestle-Aland/UBS text decided to adopt (as Eberhard Nestle did) the reading of Codex Sinaiticus in Matthew 13:35, since it says that Isaiah is the author of Psalm 78.  It is also difficult to see how the doctrine of inerrancy would survive if the NA/UBS compilers decided to adopt the reading shared by Sinaiticus and Vaticanus (and some other witnesses) in Matthew 27:49. 

         ● What would happen to the doctrine of the Virgin Birth if compilers decided (as Von Soden did) to adopt the reading in Matthew 1:16 in the Sinaitic Syriac:  “Joseph, to whom was betrothed Mary the virgin, begot Jesus who is called the Christ”?  

         ● In Matthew 17:21 and Mark 9:29, does it have no impact whether or not Jesus said that a particular kind of demon will not be exorcised except by prayer and fasting? 

          ● In Mark 7:19, does Jesus refer to a bodily function, or does Mark comment that Jesus thus declared all meats kosher? 

          ● Does it make no difference to interpreters whether Mark 1:1 does (as in the ESV, and in almost all manuscripts) or does not (as in the TNIV, and a smattering of manuscripts) include the words, “the Son of God”? 

          ● Does it have no theological impact whether Acts 20:28 refers to “The church of God, which he bought with his own blood,” or to “The church of the Lord, which he bought with his own blood,” or to “The church of the Lord and God, which he bought with his own blood”? 

          ● Is it likely that the same message will be preached from the opening verses of the Epistle of Jude regardless of whether the fifth verse refers to “The Lord” (Κυριος) or “Jesus” (Ιησους) or “God Christ” (Θεος Χριστός)?   

          ● In the angelic proclamation in Luke 2:14, did the angels say, “And on earth peace; goodwill toward men,” as the vast majority of manuscripts (and the KJV, NKJV, and MEV) say, or “And on earth peace among those with whom he is pleased” (as the ESV says, allied with the Alexandrian text)?  On December 24, 2011, John Piper preached from this passage and stated that the KJV’s rendering “was not an accurate translation.”  However, this difference between the KJV and the ESV is not a matter of translation; it is a textual difference – the difference between εὐδοκία and εὐδοκίας.  The theological nuance of the sentence pivots upon this small difference.

          It is not my intention here to tour all the textual contests that affect interpretation.  I merely mention these few as evidence that any promise that the scientific evidence demands confidence that the text of the New Testament will not ever materially change is poorly grounded as long as the text under discussion allows a few manuscripts to outweigh all the rest.  The product of the text-critical methodology that has yielded 28 editions of Novum Testamentum Graece is, and will always be, inherently non-definitive.  

          That approach does not preclude that in the event of the discovery of several very early manuscripts, the text of the New Testament could change, as we saw the text of Novum Testamentum Graece change significantly in Luke 24 after Papyrus 75 was discovered.  The same methodology that presently assigns overwhelming weight to Codex Vaticanus and its Alexandrian allies is likely to assign overwhelming weight to newly discovered early manuscripts.  Thus there is no scientific reason to absolutely preclude significant textual shifts.

          How, then, can one confidently maintain that the compilation that one can hold in one’s hands contains nothing but the original text?  Piper’s solution:  ask if you see “the peculiar glory of God shining through those words and confirming to your own mind and heart that these are the very words of God.”  That is a spectacularly bad idea.  

          Such a misapplication of Second Corinthians 4:6 renders textual criticism superfluous and replaces it with subjective contemplation.  Many Bible-readers see the glory of God shining through the words, “Go ye into all the world, and preach the gospel to all creation.”  Many Bible-readers see the glory of God shining through the words, “Neither do I condemn you; go and sin no more.”  Advocates of the KJV claim to see the glory of God shining through the Comma Johanneum.  Mormons claim to see the glory of God in their Book of Mormon, and Roman Catholics claim to see the glory of God in the book of Tobit, and so forth.  I don’t think we really want to make subjectivism the crucial factor in textual contests.

          But how else can one answer the listener’s question, as long as one agrees with the methodology of the compilers of the NA/UBS text?  How can one believe that the NA/UBS text, in its present condition, is the Word of God, considering that in the future, its compilers might reject many of its present readings?  It’s a question which I leave to advocates of the NA/UBS compilation to answer. 

          In the meantime, I close with an observation and a suggestion.  First:  scientifically speaking, the Byzantine Text does not have the instability problem that the Nestle-Aland text has.  Even at points where the Byzantine Text is unstable – where there is no clear-cut majority reading – the options are limited to known variants with significant manuscript support.  Meanwhile the Nestle-Aland compilation is susceptible to the introduction of readings which are presently unknown in Greek manuscripts.  (Of course one could argue that this aspect of the Nestle-Aland text is not a bad thing – if a large second-century papyrus manuscript of the Gospels were discovered tomorrow, wouldn’t you want the compilers to pay attention to it?)      

The Byzantine Text

          Second:  theologically speaking, I suggest that most Christians, instead of insisting on confidently possessing the exact form of the text of the autographs, should be content to possess the message that was conveyed by the original text.  (This is how almost everyone, except Greek-readers, encounters the New Testament’s message.  It was even the attitude of copyists of New Testament manuscripts; they routinely used abbreviations and contractions, thus altering the text’s form but not its message.) 

          And while the Byzantine Text does not constitute the exact form of the original text, the advocates of the Nestle-Aland compilation relentlessly insist that the textual differences between the Byzantine Text and the Nestle-Aland text do not affect the general message.  Everyone thus seems to agree that the message conveyed by the Byzantine Text is the definitive message of the original text, even if the form of the words conveying that message is not always the original form.  So, in response to the listener’s initial question, I would say that it does not require very much faith to confidently believe that no future discoveries, and no valid methodology, will ever introduce material into compilations of the New Testament text that conveys a significantly different message than what is conveyed by the Byzantine Text. 

Who's Making Your Bible?

David Trobisch

          When the 28th edition of the Nestle-Aland Novum Testamentum Graece was published in 2012, it was soon followed by A User’s Guide to the Nestle-Aland 28 Greek New Testament, released in September of 2013.  The author of this introduction to NA28 is David Trobisch, who in 2011 became a member of the editorial committee entrusted with the preparation of future editions of the Nestle-Aland Novum Testamentum Graece.  

          Trobisch’s User’s Guide to NA28 has been met with some concerns among evangelicals; Dan Wallace, for example, noted that Trobisch “got some facts wrong,” and recommended the removal of an entire chapter.  I too have some concerns.  

          One might expect all of the compilers of Novum Testamentum Graece to be Christians, since future compilations of this text will likely be the basis for future translations of the New Testament used in Christian congregations.  However, Trobisch is a fellow of The Jesus Project, an undertaking of a group called the Center for Inquiry.  His fellow-members include Frank Zindler (an atheist who is also a Jesus Mythicist, that is, he denies that Jesus ever existed), Paul Kurtz (President of the International Academy of Humanism), James Crossley (an atheist), James Tabor (perhaps best-known for his theory that the Talpiot Tomb is the tomb of Jesus), Robert M. Price (Jesus Seminar member, and also a Jesus Mythicist), and Richard Carrier (another Jesus Mythicist).    

          At the website of the Center for Inquiry, the organization is defined:  “A world-wide movement of humanists, skeptics, freethinkers, and atheists.”  And its members’ mission is plainly stated:  “To foster a secular society based on science, reason, freedom of inquiry, and humanist values.”  The website also states that it is a priority of the Center for Inquiry “to oppose and supplant the mythological narratives of the past, and the dogmas of the present.”  

          Somehow I suspect that the phrase “dogmas of the present” encompasses the historical doctrines of the Christian church.  One of the research-programs of the Center for Inquiry mentioned at the website is the Council for Secular Humanism.  It is rather surprising to learn that a member of that organization, which is clearly dedicated to erode and marginalize the cultural influence of Christianity, is also an advisor for the American Bible Society, and the curator of the Museum of the Bible which is scheduled to open in Washington, D.C. in late 2017.

          An article by David Trobisch appeared in Volume 28 of the secular humanist magazine Free Inquiry (from Dec 2007/Jan 2008).  His article is mentioned on the cover and it is accessible online (via this link).  In that article, Trobisch expresses some unusual interpretations of some parts of the New Testament.  He proposes, for example, that John 21:24 was written with all four Gospels in mind:  “This sentence does not refer to only one author and one manuscript; instead, it talks about “books” in the plural. The reader of John will have just finished reading the fourth account of “things that Jesus did.”  A modern rendition of this sentence may sound like:  “If everything Jesus did was written down, I suppose that the world could not contain all the books that would have to be published.  Four books are plenty!” The last sentence of John does not refer only to the Gospel according to John; it refers to the Gospel collection as a whole.”

           Trobisch also states, “The New Testament was published by Polycarp of Smyrna between 166 and 168 C. E.”  As corroborating evidence, he points to Second Timothy 4:9-20 and proposes that this passage “may contain the names of the publisher and forger of this letter.”  He focuses on the two names in these verses (without mentioning the four names in verse 21) that do not appear elsewhere in the New Testament:  Crescens and Carpus.  The name “Carpus,” Trobisch proposes, “could easily be interpreted as referring to Bishop Polycarp.”  

          He then goes on to propose that the reference to Crescens in Second Timothy 4:9 was added as an acknowledgement of the role of Polycarp’s secretary (mentioned in Polycarp’s letter to the Philippians which introduced the Letters of Ignatius).  “Although this argument cannot carry the burden of proof,” Trobisch concludes, “it is a nice example of corroborating evidence.”

          That theory is not merely weak.  It’s quackery.  And it is not the only highly dubious theory of the origins of New Testament books that Trobisch has promoted.  In a speech delivered in 2015, he referred to the text about Jesus promoted by the second-century heretic Marcion as “the oldest Gospel,” and began his speech with the claim that “Scholars now know of a Gospel-book that is probably older than the Gospels that are part of the New Testament.”  Trobisch also claimed that the author of the Gospel of Luke used Josephus as a source.        

          One can harbor all kinds of unusual beliefs and still be a competent textual critic.  However, Trobisch apparently believes that the Gospel of Luke post-dates the works of Josephus, and that the earliest text of the four canonical Gospels descends from the 150’s-160’s.  That position, it seems to me, is very likely to have an impact on some text-critical decisions, just as different solutions to the Synoptic Problem yield different implications about some textual variants in the Gospels. 

          Trobisch has also written that the opening sentences of Acts refer, not to the closing verses of Luke, but instead to the closing verses of John – implying that the composition of Acts post-dates the collection of the four canonical Gospels.  He has also written, “Historically speaking Paul probably did not heal.”  Trobisch’s doubts about Paul’s healing-miracles might not affect Trobisch’s text-critical work.  But does anyone think that if a textual critic believes, as Trobisch seems to, that Acts was written in the middle of the second century, this will have no impact on his text-critical decisions pertaining to the text of Acts?

          And does anyone think that it does not matter that Trobisch believes (as he has recently written) that “scribes and editors felt free to revise the Greek text during the fourteen centuries of its manuscript transmission,” rather than the normal view that a scribes’ primary ambition was to make an accurate copy of the text of his exemplar?  Do any specialists besides Trobisch believe that a typical copyist “felt free” to revise the text of the Gospels?  There were indeed some reckless copyists, but to present them as if they were typical is like saying that human beings have six digits on each hand.

          In addition to the objection that Trobisch brings some strange ideas to the compilation-committee’s table, there is a pastoral concern here.  I have never met David Trobisch but from what I have read and watched, the religion to which he subscribes is very different from the Christianity which is taught in the New Testament.  It seems to be a baptized “social gospel” philosophy which does not remotely affirm – and which directly opposes – the statement of faith of the National Association of Evangelicals, which, among other things, affirms the infallibility of the Bible, Christ’s virgin birth, His bodily resurrection, His deity, His future return, the final judgment, and salvation through the work of the Holy Spirit in personal spiritual rebirth.  

          Yet very many evangelical leaders who consider those things to be essentials of the faith – people such as D. A. Carson, James White, Craig Evans, Bill Mounce, and Steve Green – seem perfectly fine when the task of compiling the text of the Greek New Testament is entrusted to someone who denies every one of those tenets of Christianity.  At least, I have not heard much protest from them so far.  Most evangelical preachers probably would not share their pulpits with hyper-liberals and atheists.  Why, then, do they seem perfectly content to have a hyper-liberal edit the book on the pulpit?  

          It may be that our wise evangelical leaders have reckoned that just because a fox is a fox, that is no reason why a fox cannot be a skillful guardian of the chicken coop.  Nothing but bias, they might insist, would elicit a suspicion that an unbeliever might – whether purposefully or unconsciously – render the base-text of the New Testament unstable, or introduce readings into the text which have very little manuscript-support (or even none).  “It would be a gross employment of the genetic fallacy,” someone might insist, “if Christian translators deliberately avoided using a base-text compiled by someone ideologically opposed to Bible-believing Christianity.” 

          Against such politically correct wisdom I protest in the name of common sense.  The gold of the king of Sodom was as solid as the next man’s; yet Abraham (in Genesis 14:21-24) refused to receive any of it.  There is a principle being illustrated there that should not be ignored. 

          Second Corinthians 6:14 says, “Do not be yoked together with unbelievers.  For what partnership does righteousness have with lawlessness?  And what fellowship does light have with darkness?”  Paul stresses this theme emphatically for several verses:  “What agreement has the temple of God with idols?” and so forth.  He utilizes two stirring passages from the Old Testament in his call to the church:  “Come out from among them.”  And what co-operation can there be between Christ-centered churches, and members of the Center for Inquiry?  No one can serve two masters.  Paul’s warning against being yoked together with unbelievers is often unheeded in today’s society.  Still, one might think that in the enterprise of compiling the text of the Greek New Testament, this principle should not be ignored when alternatives are readily available.