Wednesday, August 16, 2017

Codex Macedonianus

The first page of John in Codex Y.
            In 1901, twenty years had passed since the publication of Westcott and Hort’s groundbreaking compilation of the Greek New Testament.  The Textus Receptus and its primarily Byzantine readings had, for the most part, been pushed aside.  The compilation that stood in its place, and which formed the basis for the American Standard Version which was released in 1901, was essentially Alexandrian.
            Also in 1901, Codex Macedonianus was discovered – an important manuscript of the Gospels, also known as Codex Y or 034 – but hardly anyone seemed to notice.  Before the release of Codex Y, the Sinaitic Syriac and its numerous rare (but wrong) readings captured the imagination of scholars, and a few years after the existence of Codex Y was announced, Charles Freer obtained Codex Washingtoniensis, which is still famous for the interpolation between Mark 16:14 and 16:15 that bears the name of the manuscript’s purchaser, the Freer Logion
            In between the discoveries of those two manuscripts, it is not surprising that the discovery of Codex Y by J. Bevan Braithwaite, and its subsequent analysis by his brother, W. C. Braitwaite, did not capture the spotlight.  Codex Y is younger (its production-date is assigned to the 800’s) and its text is mainly Byzantine, which, in the early 1900’s, was understood by leading textual critics to mean that it was far less important than Alexandrian and Western texts.  Even though Bruce Metzger drew attention to Codex Y in 1963, stating that it deserved more attention than it had received up to that time, not much attention seems to have been given to it. 
             More recently, however, the stewards of Codex Y at Cambridge University have digitized the entire manuscript, indexed its entire text, and produced a detailed description of its physical features.  So this might be a good time to become acquainted with this impressive Gospels-manuscript.
            Codex Y measures approximately 18 centimeters tall and 13 centimeters wide.  Its uncial letters are neatly written.  Chapter-titles in large red uncial lettering appear at the top of the page on which chapters begins.  The text is divided into Eusebian sections, and the section-numbers are written in red (except in Luke 1:1-11:26) in the margins (365 for Matthew, 233 for Mark, 342 for Luke, and 232 for John).  Sections frequently begin with a red initial that protrudes into the left margin; where such an initial does not appear at the beginning of a section, an obelus (two dots, arranged like a colon, and separated by a small wavy horizontal line) appears in the text at the start of the section, and another obelus accompanies the section-number in the margin. 
            When the manuscript was in pristine condition, each Gospel (as far as can be discerned) was preceded by a brief introduction and a chapter-list.  In addition, artistically executed headpieces for Mark, Luke, and John are extant. 
            Codex Y possesses an interesting lectionary-apparatus.  Symbols for arche (start) and telos (stop) are written in red in the text; blank space was left so that they could be inserted without harming the aesthetics of the pages.  Lectionary-related notes and incipits (that is, the opening phrases with which the lector was to begin reading the assigned passage for the day) frequently appear at the foot of the page.   
            The copyist of Codex Y was relatively accurate; a few small omissions (at Matthew 24:6, Luke 2:25, Luke 10:38, Luke 11:7, and at John 6:43) are corrected in the margins, and each is accompanied by an asterisk; an asterisk also appears in the text where the omission occurred.  Margin-notes also supplement Matthew 22:14, Mark 15:28, and the last part of John 8:14 – although whether the initial non-inclusion of these passages was the fault of the copyist, or a reflection of his exemplar, may be an open question.  The phrase καὶ οἱ μαθηταὶ αὐτοῦ is also missing from the end of Matthew 9:19 – a parableptic error that went undetected by the proofreader.
At the end of Matthew 25:13, the words
"in which the Son of Man comes"
are added, in red, in a marginal note,
linked to the end of the verse (in line 2
of the text) by comet-like symbols
            The text of Codex Y is written in one column per page, and the number of lines on each page is strangely inconsistent:  at first, there are 16 lines per page; then the last page of the Gospel of Matthew has 17 lines (very probably so that the following page would not be occupied by a single line of text); the number of lines per page then returns to 16 when Mark’s text begins, until Mark 3:13, at which point the number of lines per page jumps to 19, and stays there until Mark 16:20.  The text of the Gospel of Luke is written with 16 lines per page, at first, but suddenly changes to 21 lines per page in chapter 11, and then back to 19 lines per page in chapter 13.  These shifts may indicate that the copyist worked on each Gospel (or two sections of a Gospel in the case of Luke) separately, using different batches of differently prepared parchment.

            Some pages are missing:  all of the pages that contained Matthew 1:1-9:11 and Matthew 10:35-11:4, Luke 1:26-36, Luke 15:25-16:5, Luke 23:22-34, and John 20:37-21:17.      

At the online presentation of Codex Y (034, catalogued at Cambridge as MS Add. 6594) at the Digital Library of Cambridge University, a chapter-by-chapter guide (using the ancient kephalaion-list as the chapters) is accessible in the Contents menu at the site.  Here is a basic index:
            Matthew 10:11 (first extant page of Matthew) 
            Mark 1:1 
            Luke 1:1 (Lk. 22:43-44 included
            John 1:1

            Page-view 528 shows that the pericope adulterae is not included in the text of Codex Y; John 7:52 ends on the same line in which 8:12 begins.  There is, however, more to the story:  in the outer margin, and in between the end of 7:52 and the beginning of 8:12, the lectionary-apparatus instructs the lector to jump ahead (υπ, that is, υπερβαλε).  Within the text, immediately following the υπ-symbol, the lector is instructed to resume (αρξαι).  Also, in the margin alongside the line in which 7:52 ends and 8:12 begins, there is an asterisk, and the letters λιθ.  This did not go unnoticed by W. C. Braithwaite, who wrote the following in the course of a brief article that appeared in the Journal of Theological Studies in 1905:
John 8:12 follows 7:52 in the text of Codex Y,
but the lectionary-apparatus implies
the existence of the PA in an earlier copy
            “The rubrics for the Pentecost lesson John 7:37-52,8:12, include rubrics at the end of v. 52 and at beginning of v. 12, although the text of a omits the intervening verses (Pericope adulterae) and the rubrics accordingly come together on the same line.  The rubricator must have known of the verses and indeed puts λιθ. in the margin, that is, perhaps, περὶ τοῦ λιθάζειν or some similar phrase.  Dr. C. R. Gregory, however, suggests to me that the marginal note stands for λήθη ‘an omission’, the rubricator noting in this way the discrepancy between the text which he was rubricating and the copy of the Gospels out of which the rubrics were taken, which must have contained the Pericope.”  
            Braithwaite’s description of Codex Macedoniensis in the 1901 Expository Times (beginning on page 114) includes a list of some of its interesting readings, which include the following:
            ● Mt. 16:2-3 – Y does not include most of the passage, agreeing with the Alexandrian Text.
            Mt. 22:14 – Y does not include this verse; it is added in the margin.
            Mt. 24:18 – Y has το ιματιον (garment) rather than τὰ ιμάτια (garments).  
            Mt. 24:36 – Y does not have μου.
            Mt. 25:13 – Y does not have the final phrase “in which the Son of Man comes.”  The phrase has been added in the margin, apparently by the rubricator (in red ink). 
            ● Mk. 4:30 – Y reads υπο την λυχνιαν επιτεθη, which, with υπο instead of επι, means, “set in place under the lampstand,” rather than “set in place upon the lampstand.”  It is tempting to suppose that some copyist pictured lampstands as something like simple chandeliers underneath which lamps were suspended.
            ● Mk. 10:20-21 – Y adds τι ετι ὑστερω (“What am I missing?”) at the end of the man’s question, and adds ει θελης τέλειος ειναι (“If you want to be perfect”) at the beginning of Jesus’ answer.  Both harmonizations are supported by Codices K, M, N, W (which transposes the first part), and Π.
            ● Mk. 14:65 – Y reads ελαβον instead of εβαλλον.  (This makes a difference in translation; with εβαλλον or εβαλον the soldiers strike Jesus, whereas with ελαβον – a reversal of letters – the soldiers receive, or welcome, Jesus.)           
            ● Luke 14:5 – Y reads ονος (“donkey”), not υιος (“son”).  Codex Y thus adds to the array of witnesses which favor this reading, which is neither Alexandrian nor Byzantine (both support υιος, and so do P45 and P75) but which was adopted in the Textus Receptus, and has strong intrinsic appeal (as well as a diverse array of external support which includes ﬡ, K, L, Ψ, 33, family-1, the Palestinian Aramaic version, various Old Latin copies, and the Vulgate).
            ● Luke 18:24 – Y reads των ουρανων (“of heaven”) instead of του Θεου (“of God”).  Again Y finds allies in Codices K, M, and Π. 
            The staff of the Cambridge Digital Library (which includes in its diverse collection Codex Bezae and a first-edition Gutenberg Bible) is to be congratulated for its high-quality presentation of this manuscript.  Not only are the photographs first-rate, but so is their magnification-method.  Visitors will learn much from an exploration of the “About,” “Contents,” Item Metadata” and other sub-sections of the site.  Codex Sinaiticus may still have the most thorough online presentation of any Greek New Testament manuscript (though the flaws in its on-site “translation” have not been addressed), but the presentation of Codex Macedonianus is not far behind.

Monday, August 14, 2017

eZooMS: Making Paleography Obsolete?

The York Gospels, shown here
(the beginning of Matthew),
has been sampled via eZooMS,
and so has 
The Hornby Bible.
PROBLEM:  One of the most frustrating aspects of manuscript-studies involves estimates.  Very often, unless a copyist has left a colophon that mentions the date when the manuscript was produced, the production-date of the manuscript can only be estimated via analysis of the handwriting-style.  Paleographers – analysts of ancient handwriting – may not always agree, however, and occasionally their estimates vary widely, not just by decades but by centuries. 
            Even worse is the variation in theories about the location where manuscripts were made.  Scholars have proposed that Codex Vaticanus, for instance, was made in Alexandria – or in Caesarea – or in RomeCodex Bezae has been thought to have been made in Italy in the 600’s – or in Beirut around 400.  Sometimes colophons mention the place where a manuscript was made, and sometimes illustrations, decorations, and other meta-textual features in the manuscripts provide clues – but all too often, determining the location where a manuscript was made is a matter of calculated guesswork.

SOLUTION:  Timothy Stinson, Associate Professor of English at North Carolina State University, realized in 2009 that with developing technology, scientists could analyze the DNA in the processed animal skins – that is, parchment – out of which most ancient manuscripts are made.  By 2015, a non-destructive method was developed to obtain genetic samples of parchment – enough to allow genetic analysis so detailed that it identifies the species of animal whose skin was used, and whether it was a male or a female.     
            Parchment is sometimes not the only DNA-source in a manuscript.  Beetles and other destructive pests sometimes left their DNA behind.  When venerated manuscripts were kissed, a DNA-sample left by the the kisser may survived to the present day.        
            A team of researchers consisting mainly of Matthew D. Teasdale, Sarah Fiddyment, Jiří Vnouček, Valeria Mattiangeli, Camilla Speller, Annelise Binois, Martin Carver, Catherine Dand, Timothy P. Newfield, Christopher C. Webb, Daniel G. Bradley and Matthew J. Collins recently refined a sample-gathering technique called electrostatic Zooarchacheology by Mass Spectrometry – conveniently known as ZooMS (or eZooMS with the electrostatic factor included).  EZooms involves “triboelectric extraction of protein” from the parchment’s surface (via the gentle application of an eraser),   
            The team applied this analysis-technique to a historically significant Latin Gospels manuscript, the York Gospels, which is kept at York Minster, in the city of York (the same place that was captured in 866 by Ivar the Boneless, the famous Viking). 

            For some idea of the usefulness of the data that can be gathered via the eZooMS sampling-technique, see the following articles and essays:

             Nowadays, New Testament textual critics group manuscripts according to their readings.  Theoretically, in the future it may be possible to group manuscripts according to parchment-sources distinct to specific areas.  It may be possible to identify locales of New Testament manuscripts, not just in theory by showing that the textual variants in a manuscript are shared with a patristic writer whose locale is known,  but by isolating organic abnormalities (from pollen, beetles, wax, etc.) of each manuscript and determining what other manuscripts, if any, share the same or similar abnormalities.
            For example, suppose that eZooMS shows that a particular group of manuscripts shares a special kind of parchment (parchment made from aurochs-skin, for example).  This could point researchers toward the next logical step:  studying those manuscripts’ texts to look for relationships among them.
            Or, suppose that different groups of Gospels-manuscripts, or different groups of lectionaries, are someday shown to be made of parchment from the skins of animals that lived only in particular places, or that a particular group of manuscripts consists of parchment which was prepared in a unique way, and which was as result either particularly resistant to, or vulnerable to, beetles.  Having that sort of data could help define the parameters for particular manuscript-groupings, along with data such as manuscripts’ page-size and rulings.   
            In addition, although I did not notice anything in the descriptions of eZooMS that would suggest that it can reveal the age of parchment, it seems to have the potential to reveal details about where a manuscript was made and what happened to it afterwards, and it does not seem unlikely that thoughtful analysis of that data (combined, perhaps, with ink-analysis) may be able to narrow down a manuscript’s provenance and production-date more precisely that what is possible via paleography.  
           The Ethiopic Garima Gospels, when subjected to radiocarbon tests of its parchment, was found to be about 500 years earlier than paleographers had thought.  Perhaps future applications of sampling-techniques such as eZooMS will yield many more such surprises in the future.  


Friday, August 11, 2017

Michael Brown and the Elephant in the Room

[Note for newcomers:  I am most definitely NOT a KJV-Onlyist.]

            Recently, Dr. Michael Brown, on The Line of Fire radio show, made an episode which, at first, targeted King-James-Onlyism, but which quickly shifted so as to target the King James Version itself.  His primary objection against the KJV was that its language is unfamiliar to people nowadays.  He stated, “What may have been an accurate translation then will not convey the same thing today.”  Of course that is true, where archaic terms and obsolete grammar is concerned – but that is an effect of the natural development of the English language in the past 400 years; it does not reflect upon the quality of the translation itself. 
            To illustrate:  if someone were to read about dissolving political bands in the Declaration of Independence and conclude that U2 was breaking up, the Declaration of Independence would not be to blame.  When readers approach a 400-year old text, it is their responsibility to take its age into consideration when interpreting it.  The chronological distance between 1611 and 2017 makes the King James Version more difficult to understand, but it does not necessarily make it erroneous.
            The accuracy of the KJV can only be measured fairly when it is measured in light of the meaning of words in 1611, and in light of the text upon which it was based.  Dr. Brown seemed to grant this when he provided two examples of terms in the KJV that meant one thing in 1611, but which mean something else in 2017:  the term “meat” – which could refer to food in general, including grains and fruits – and the term “study,” which was intended, just as Dr. Brown said, to mean, “Do your best,” or to exercise diligence when pursuing a particular goal. 
            So, when someone interprets the term “meat offerings” as if the cooked flesh of an animal must be involved, and when someone interprets “study” as if the word necessarily involves peering into a book, the error does not emanate from the KJV.  The error emanates from the reader’s failure to perceive what those particular words meant in 1611.  A simple glossary of the KJV’s archaic terms can greatly lower the risk of this sort of misimpression. 
            When Dr. Brown turned to the King James Version’s use of the word “Easter” in Acts 12:4, he called it an error.  However, a careful investigation shows that the term “Easter,” in the early 1600’s, was synonymous with “Passover.”  Dr. Brown said, “The Greek does not say ‘Easter.’  The Greek says ‘The Passover.’” 

            The term “Passover” was an invention of William Tyndale.  If you were to take in hand Tyndale’s translation of the New Testament, you would see that he freely interchanged the term “Easter” and his own new word “Passover.”  One example should suffice:  in Matthew 26:18-19, in Tyndale’s translation, Jesus tells His disciples to go into the city and deliver the message “I will kepe Myne ester at thy housse with my disciples,” – “and the disciples did as Iesus had apoynted them, and made redy the ester-lambe.” [Bold print added to make the reference super effective.] 
            So it should be plain as day that the KJV’s “Easter” in Acts 12:4 is not an error; it means the same thing that the versions which refer to the end of the Passover-feast mean:  that Herod intended to wait until after the last day of the Feast of Unleavened Bread before having Peter executed.  Dr. Brown said, “To say ‘Easter’ is 100% inaccurate and misleading.  It’s a mistake.”  However, what he has perceived as an error is really just another case of obsolete language.
            Likewise, Dr. Brown charged the KJV with error because the term didaskalos” is translated in the KJV as “Master,” rather than as “Teacher,” apparently unaware that in 1611, the term “master” was entirely capable of referring to a teacher.  (An echo of this usage is still retained in the term “schoolmaster.”)
            What were the other prime examples of errors in the KJV?  Dr. Brown said that instead of referring to “devils,” the KJV should refer to “demons.”  But does anyone imagine that the KJV’s use of the word “devils” is really confusing?  As confusing as using four English Bibles translated from four base-texts using four different translation-techniques?    
            A few of Dr. Brown’s other examples are more convincing:
            ● Readers could be spared some confusion if the King James Version’s translators had not used the term “unicorn.”  However, the KJV’s preface (The Translators to the Reader) specifically cautions readers against putting too much weight on their renderings of rare terms for animals, plants, and minerals.  Plus, the precise meaning of the Hebrew term re’em that is often translated as “wild ox” is still a matter of debate – it might be a wild ox, or the extinct buffalo-like animal known as the aurochs, or the rhinoceros.  From before the time of Christ, this term has been translated as if it refers to a one-horned animal (a real one, not a mythical bearded goat-horse thing), and the KJV’s translators deferred to the traditional understanding of the term, cautioning their readers not to treat this rendering dogmatically.
            ● Dr. Brown’s objection against the KJV’s artificially plural term “cherubims” seems entirely valid. 
            ● The text of First Kings 18:37 in the KJV could be made more literal by reading “Answer me” rather than “Hear me.” 
            ● Another inaccuracy in the KJV, Dr. Brown said, is found in Psalm 84:  “Psalm 84:  one of the verses that I grew up loving was, ‘Blessed be the Lord our God, who daily loadeth us with benefits’ in the King James.”  Dr. Brown was recollecting Psalm 68:19, not anything in Psalm 84.  His point (minus the mistaken reference) seems valid; more recent versions render the Hebrew phrase as “who bears our burdens,” or “who bears us up.”

            Dr. Brown put a microscope to the text, symbolically speaking, and found an error in how the KJV treats the Greek word exousian in Luke 10:19.  He also objected against translating the word ekklesia as “church.”  Another “major example” of errors he has found in the KJV is its use of two different words (“weakness” and “infirmities”) to represent the same Greek word in Second Corinthians 12:9.  
            In these cases, he may have a technical point, but it’s like watching an archer hit the bullseye, and having a referee say that the arrow didn’t hit the very center of the bullseye, so it’s not close enough and the archer might as well have missed the whole target.   If one were to put such a yardstick alongside the NASB, NIV, ESV, etc., then one could identify hundreds of such “errors,” every time there is no distinction between the singular and plural pronouns, and every time the word και (and) is not represented, and every time a proper name is put in place of a pronoun, and so forth.
            Dr. Brown also proposed that if a team of the King James version’s scholars had been able to sit down to improve the translation 20 years after its initial publication, they would have changed passages such as Acts 12:4 and Luke 10:19.  History stands in the way of Dr. Brown’s theory:  in 1638, some scholars who had served on the KJV’s translation-committees (John Bois and Samuel Warddid tidy up the text of the KJV – and they did not change those passages. 
            And then . . . 


            About 21 minutes into his presentation, Dr. Brown brought up manuscript-related issues, even though initially he had said that he would set that sort of thing aside.  
            In the fourth segment of the presentation, the manuscript-base of the KJV’s New Testament text came up again when Charles from Tennessee called the show and pointed out that “The bigger issue is actually the textual issue,” and that modern translations treat Mark 16:9-20 in ways that call it into question.  Charles also alluded to the enormous amount of manuscript-evidence in favor of Mark 16:9-20, and to the features in Vaticanus and Sinaiticus which indicate that their copyists were aware of Mark 16:9-20 “and willingly omitted it.”  Finally, Charles asked Dr. Brown, “Would you recommend any version that brackets or negates the ending of a Gospel, that basically removes the resurrection-account of Christ?”
            In reply, Dr. Brown first said that if one feels a certain way about the manuscripts, one should use the NKJV or MEV – but then he said (referring to Mark 16:9-20), “We know that that was not the original ending of Mark.  The vocabulary is totally different.”
            Charles responded that he had read Burgon’s book on the last 12 verses of Mark.
            Dr. Brown responded that Burgon’s book “has been refuted many times over.”  Yet he failed to name any specific refutation of Burgon’s book; instead, he recommended reading D. A. Carson’s book (The King James Version Debate: A Plea for Realism) and James White’s book (The King James Only Controversy), too – “They’ll help you there.”  Those who have read Carson’s book may wonder what Brown was talking about, since Carson specifically says in A Plea for Realism, on page 65, “I am not here arguing for or against the authenticity of Mark 16:9-20.”  White’s book is also remarkably unhelpful for people looking for accurate information about the external evidence pertinent to the ending of Mark – and concludes its discussion of Mark 16:9-20 with the affirmation that “Every translation should provide the passage” as well as mention “that there is good reason to doubt the authenticity of the passage as well.”  Well, that clears things up, eh.  (I contend in my book Authentic:  The Case for Mark 16:9-20 that with patristic support for Mark 16:9-20 from the 100’s, and with over 99.9% of the Greek manuscript-evidence supporting the inclusion of the passage, across all text-types and across many locales, such a definite maybe is not the best we can do.)  
            Dr. Brown then reaffirmed, “We don’t have the original ending to Mark’s Gospel.”  But he added that Mark 16:9-20 was “received by the church, and I personally am happy to use it.”
            Wait, WHAT?!  Dr. Brown just said that Mark 16:9-20 is not part of the original text, but he is happy to use it.  Happy to use it as what?  As if it is Scripture, or as if it is the work of some non-inspired person in the second century? 

            Dr. Brown recommended reading good commentaries on Mark to get the details about the ending; unfortunately he did not name any specific commentaries.  Then, after mentioning First John 5:7 again, he reminded himself that he had said that he wouldn’t be debating about the manuscripts – and again told his listeners that if they are at home with the Textus Receptus, “then by all means use the New King James, or the Modern English Version.”  (This seems a little inconsistent, since the NKJV has some of the same features – in Psalm 68:19 and Second Corinthians 12:9, for example – that Dr. Brown called errors.)
            Thirty-four minutes into the show’s video, Dr. Brown tried to reframe the narrative after Charles’ lively contribution to the discussion – but manuscripts were clearly still on his mind:  at one point he started a sentence with “Putting the manuscript debate aside” but continued, “don’t we want a translation . . . that has better manuscript evidence?”. 
            I don’t think that there is much of a chance that Dr. Brown will persuade any King-James-Onlyists that they are on the wrong track, as long as he pretends that the textual matters do not matter.  Most people who are willing to learn the archaic language of the KJV are not the sort of people who are going to be satisfied knowing that they have acquired the basic doctrinal message of the Bible; they want the full counsel of God, with no adulteration.  Few and far between are those individuals who would abandon the status of the KJV, the textual stability of the KJV, and the familiarity of the KJV, in order to be rid of the trivial inaccuracies listed by Dr. Brown.  One might as well invite people to kill their Cocker Spaniel in order to get rid of a few fleas.
            There are still some KJV-Onlyists who insist that the KJV’s translators themselves were as inspired as the apostles and prophets, but that is not where the momentum of the KJV-Only movement is going.  Increasingly, KJV-Onlyists (such as Samuel Gipp and David Sorenson) are making textual issues the centerpiece of their case.  To insist that all the essential doctrine is still there in the Alexandrian base-text of the NIV, ESV, NLT, etc., and that that makes it okay to select either the Textus Receptus or the Nestle-Aland compilation with as much consideration as one uses to select ice cream flavors, while it is spectacularly obvious that the differences between the two yield dozens of interpretive differences, is to insult the intelligence of one’s listeners.


Wednesday, August 9, 2017

The Comma Johanneum

First John 5:6b-9a in Codex A.
          In First John 5:7-8, there is a textual issue.  The King James Version reads, “For there are three that bear record in heaven, the Father, the Word, and the Holy Ghost: and these three are one.  And there are three that bear witness in earth, the Spirit, and the water, and the blood; and these three agree in one.”    The New American Standard Bible, however, has a shorter text:  “For there are three that testify:  the Spirit and the water and the blood; and the three are in agreement.”  The material that is in this passage in the KJV and its Greek base-text, but not in the NASB and its Greek base-text, is called the Comma Johanneum, which may be roughly translated as “John’s phrase” or “That phrase in John’s writings.”      
First John 5:7-8 in the 1611 KJV.
            The form of First John 5:7-8 in the KJV is based on a few late Greek manuscripts, plus hundreds (perhaps thousands) of copies of the Latin Vulgate, and some Latin patristic quotations.
            The form of First John 5:7-8 in the NASB is based on almost all Greek manuscripts of First John, including the early ones, plus hundreds of versional copies of First John in various languages, and many patristic quotations.   
            This difference in translations echoes a difference in the Greek base-texts used by the translators.  The King James Version is based on the Textus Receptus, or “Received Text,” a Greek compilation made in the 1500’s, beginning with Erasmus’ 1516 edition but continuing on throughout the 1500’s in various editions by Erasmus, Stephanus, and Beza, and issued with some adjustments in 1633.  The first edition of the Textus Receptus did not contain the Comma Johanneum, and Erasmus was criticized because of this.  In the course of his response to the accusation that he had acted irresponsibly by failing to include the Comma Johanneum (henceforth abbreviated as CJ), Erasmus wrote that he did not include the phrase because it was not in any of the Greek manuscripts that had been available to him when he had prepared his Greek compilation, and that if a single manuscript had contained the phrase, he would have included it.
The copyist of MS 1780 made a mistake when writing the
text of First John 5:7-8, accidentally repeating part of the
passage.  (Page 386 at the Rubenstein Library site.)
            Erasmus did not make a promise to include the CJ in the event that a manuscript was found that contained it.  The fictitious story that Erasmus made such a promise has been circulated far and wide; Bruce Metzger gave it wide popularity by presenting it in his handbook The Text of the New Testament as if it were true.  Even after Metzger retracted his claim – barely and timidly, in a footnote in the appendix! – it has proven to be a cockroach of a story, in the sense that it is hard to eradicate, even though Henk J. de Jonge efficiently refuted it in 1990.
            [The Legend of the Rash Promise is not the only fiction that commentators have spread about the CJ.  It is often claimed that the CJ was never cited at any church councils – which, though true as far as councils of Greek-speaking clerics are concerned, ignores the brief Council of Carthage (where Latin was prevalent) that took place in 484 (not to be confused with other councils that occurred there).  At this council, according to Victor Vitensis, Eugene of Carthage led a large delegation of African bishops and was prepared to confront the Vandal (and Arian) ruler Huneric with a citation of the CJ as “a shining light teaching the unity of the divinity of the Father and Son and Holy Spirit.”] 
            But there is something to the vague idea that Erasmus said something that induced him to include the CJ even on scant evidence.   After issuing the first two editions of his Greek compilation, Erasmus wanted to refine his work again.  Having stated previously that he would have included the CJ if he had found it in a single manuscript, he found himself in a predicament when someone brought to his attention the existence of a Greek manuscript from Britain in which First John 5:7-8 included the CJ. 
            Erasmus was capable of anticipating what his opponents would say if he continued to refrain from putting the CJ into the Greek text of his compilation (something like, “You claimed that you would have included it if it was in just one Greek manuscript, but now, after being shown one Greek manuscript that has the phrase, you still did not include it!  How inconsistent!”) and so, in 1522, Erasmus included the CJ in the third edition of his Greek compilation, and there it remained in the Textus Receptus throughout the 1500’s, and there it was in 1604 when the translation-work on the KJV began.
            Entire books have written about the subsequent history of the debate that has orbited the CJ in the 1600’s and onward; the names Isaac Newton, Richard Porson, Edward Gibbon, and Charles Forster should not escape the notice of anyone who wants to be thoroughly informed about all that.  My focus today is elsewhere:  I wish to share some reasons for maintaining that the CJ began as an allegorical comment about verse 8 in a branch of the Old Latin version, and that this can be demonstrated fairly concisely. 
            The earliest patristic evidence that is sometimes interpreted as evidence in favor of the CJ is a comment from Cyprian, bishop of Carthage, in the mid-200’s.  In his composition Treatise on the Unity of the Universal Church (1:6), Cyprian says: “Dicit Dominus, ‘Ego et Pater unum sumus,’ et iterum de Patre et Filio et Spiritu sancto scriptum est:  ‘Et tres unum sunt,’” that is, in English, “The Lord says, ‘I and the Father are one,” and again, it is written of the Father and Son and Holy Spirit, ‘And these three are one.’”  That final phrase, “And these three are one” is taken by defenders of the CJ as a reference to the end of First John 5:7.  However, depending on the arbitrary preferences of Latin translators, both verse 7 and verse 8 could end with the words Et hi tres unum sunt, or Et tres unum sunt (in the Vulgate, for example, as edited by Eberhard Nestle in 1906, both verses end the same way, Et hi tres unum sunt).  This reference does not rule out the idea that Cyprian was quoting verse 8, and interpreting it as a symbolic reference to the Father, Son, and Holy Spirit.
            But this naturally raises a question:  if Cyprians quotation is from 5:8 rather than 5:7, why did Cyprian say that it was something about the Father and Son and Holy Spirit?  If Cyprian’s text of First John did not have the CJ, how ever did he manage to read a text that meant, For there are three that testify:  the Spirit and the water and the blood; and the three are in agreement” and perceive therein a reference to the Father, Son, and Holy Spirit? 
            I contend that the Latin text of First John 5:8 used by Cyprian (and by several other Latin writers) contained a transposition.  Rather than refer to the spirit (or Spirit), and the water, and the blood, a form of First John 5:8 in a branch of the Old Latin text referred to the water, the blood, and the spirit, and it was only after this change to the word-order that the text of verse 8 elicited an allegorical interpretation, which a Latin writer expressed in a note that eventually was inserted into the text as the CJ.  The Old Latin text also shows that some Latin copyists altered the three witnesses in verse 8, so as to refer to the flesh (“caro”) as one of them.  But the transposition is the thing to see.  Consider these Latin utilizations of the passage:
            ● Liber Apologeticus (380’s, probably written by Priscillian or one of his associates):  Tria sunt quae testimonium dicunt in terra:  aqua caro et sanguis et haec tria in unum sunt.  Et tria sunt quae testimonium dicent in caelo:  Pater Verbum et Spiritus et haec tria unum sunt in Christo Iesu.” That is:  “There are three that bear witness in earth:  water, flesh, and blood, and these three agree in one.  And there are three that bear witness in heaven:  Father, Word, and Spirit.  And these three are one in Christ Jesus.”
            The order is different from what is seen in the Textus Receptus (in which the heavenly witnesses are mentioned before the earthly witnesses).  And the earthly witnesses themselves, and the order in which they are mentioned, are different in Priscillian’s quotation; instead of Spirit and water and blood, Priscillian mentions the water, flesh, and blood.  Priscillian also adds an extra phrase at the end, “in Christ Jesus.”
            ● Contra Varimadum Arianum (either from the late 300’s and written by Idacius Clarus, or from the late 400’s and written by Vigilius Thapsensis) cites First John 5:7-8 with the CJ and with the transposition in verse 8:  “John the Evangelist . . . says there are three who afford testimony on earth:  the water, the blood, and the flesh, and these three are in us; and there are three who afford testimony in heaven, the Father, the Word, and the Spirit, and these three are one.”  Notice that once again, the earthly witnesses are listed before the heavenly ones, and “flesh” is one of the earthly witnesses, and that the order of earthly witnesses in this Latin text is different (water, blood, flesh) – varying from the order used by Priscillian, but agreeing partly; water is listed first.
            ● Formulae Spiritualis Intelligentiae (from c. 440, by Eucherius of Lyons), chapter 10, states that the number three represents the Trinity; the author cites First John 5:8 as if it is a clear example:  “In the epistle of John, three are those who bear witness:  water, blood, and spirit.”  The CJ itself is not cited; rather, verse 8 is regarded allegorically as a reference to the Trinity – with the order of the witnesses rearranged so that water is listed first.
            ● Complexiones in Epostolis Apostolorum (from the 500’s, by Cassiodorus), utilizes the CJ as part of the text of First John:  “Cui rei testificantur in terra tria mysteria:  aqua sanguis et spiritus, quae in passione Domini leguntur impleta:  in coelo autem Pater, et Filius, et Spiritus sanctus, et hi tres unus est Deus.”  In English:  “And the three mysteries testify on earth:  water, blood and spirit.  The fulfillment of which we read about in the passion of the Lord.  And in heaven:  Father and Son and Holy Spirit.  And these three are one God.”  Notice again the word-order in the Latin text:  water is first; blood is second; spirit is third.  (Also notice that the variation in the wording of the CJ:  “Son” appears rather than “Word.”) 
            ● The Preface to the Canonical Epistles (found in the Latin Codex Fuldensis, which was produced in 546) was, for a long time, thought have been written by Jerome, and this of course gave its contents added weight in the 1500’s, and perhaps to the translators of the KJV as well.  Here is what the author says about the CJ:  “Much error has occurred at the hands of unfaithful translators contrary to the truth of faith, who have kept just the three words ‘water, blood and spirit’ in this edition, omitting mention of Father, Word and Spirit.”
            Notice that as far as the author is concerned, the CJ belongs in the text, and its absence is the effect of unfaithful translators.  Notice, too, the word-order in his citation of First John 5:8:  once again, water is listed first.  
            ● Adversus Elipandum (by Etherius of Osma in the 700’s), despite being written long after the Vulgate began to replace the Old Latin text, features a utilization of First John 7:8 with an Old Latin reading:  “the water and the blood and the flesh.”  Again, notice the word-order.

            The thing to see is that where the transposition goes, the CJ follows. 
            In the commentary of Scotti Anonymi – this moniker will have to do for the unknown author (possibly Augustinus Hibernicus) of a Latin commentary preserved in a single manuscript (Codex Aug. 233, kept at the Badische Landesbibliothek (Baden State Library) in Karlsruhe, Germany).  The manuscript itself was produced in the 800’s; the commentary was probably composed in the late 600’s.  In the relevant portion of Scotti Anonymi’s commentary, the CJ is not cited, but First John 5:8 is nevertheless interpreted by the commentator as if the three witnesses symbolize the three Persons of the Trinity, and the order of the witnesses in the citation is water, blood, and spirit. 
            Angland Shane has offered a summary of the gist of the part of the commentary that pertains to First John 5:8:  “The moral interpretation interpreted the three witnesses as baptism (water) martyrdom (blood) and the Spirit filled life (Spirit). Christ’s incarnation is presented as the prime example for this moral interpretation. The anagogical interpretation is Trinitarian. Water is said to speak of the Father (ingeniously Jeremiah 2:13 is cited as support). Blood speaks of Christ, especially His passion on the cross, and the Spirit is the Holy Spirit.”
            If the same sort of symbolic filter upon the text was applied to an Old Latin text of First John 5:8 in which the witnesses’ order was water-blood-spirit, it would explain a progression of events:
            (1)  An interpreter of the Old Latin text, upon reading the reference to water, blood, and spirit, is reminded of the Father, Son, and Spirit.  Recalling the testimony of the Father, Son, and Spirit in the Gospels, he composes a note mentioning this. 
            (2)  This note becomes the CJ, first in the margin of Old Latin manuscripts, and then in the Old Latin text itself in the 300’s – sometimes before verse 8 and sometimes after verse 8.  For a little while, it was exclusive to copies in which the order of the three witnesses in verse 8 was transposed.  
            (3)  By the 500’s, its doctrinal usefulness results in its adoption in Latin texts in which the witnesses in verse 8 were not transposed.

            Meanwhile in the Greek manuscripts, there is no external evidence of the existence of the CJ for over a thousand years, because in the entire Greek transmission-stream, the transposition of the witnesses in verse 8 never occurred.  In the late medieval era, some manuscripts were made in which each page contains two columns of text; the Greek text occupies one column, and the Latin text occupies the other one.  Sometimes the Greek text was altered so as to agree with the Latin text, and for this reason, the CJ appears in the text of First John 5:7 in a few late medieval Greek manuscripts (the earliest of which is (probably) minuscule 629, which is assigned to the 1300’s, though Daniel Wallace seems to think it is much younger).  But, unless one were to add to the equation Greek manuscripts that were copied with printed Greek copies of the Textus Receptus as their exemplars, the CJ seldom appears in exactly the same form twice in Greek. 
            For example, in Codex Montfortianus – the manuscript which was brought to Erasmus’ attention in the 1520’s as containing the CJ – it seems evident that the source of the CJ is Latin, not Greek.  The text of the CJ in minuscule 61 (as was noted by Orlando T. Dobbin in 1854) runs as follows [the sacred-name contractions that are underlined here are overlined in the manuscript]: 
            ὁτϊ τρεῖς εἰσιν οἱ μαρτϋροῦντ·¨ ἐν τῶ ουνω,
            πηρ, λογος, καί πνα αγῖον, καί οὗτοι οἱ τρεῖς εν εισϊ. 
            καί τρεῖς εἰσιν οἱ μαρτϋροῦντ·¨ ἐν τῃ γῃ, πνα, ὑδωρ, καί αιμα.

But this is different from the text found in the Greek column of minuscule 629, which says that the witnesses are απο του (“from the”) heaven rather than εν τω (“en the”) heaven. And although Erasmus’ third edition (1522) includes First John 5:7 in the same form found in minuscule 61, by 1556 the text was different: 
            ὁτϊ τρεῖς εἰσιν οἱ μαρτϋροῦντες ἐν τῶ ουρανω,
            πατὴρ, λογος, καί τὸ πνεῦμα ἅγῖον, καί οὗτοι οἱ τρεῖς εν εισϊ. 
            καί τρεῖς εἰσιν οἱ μαρτϋροῦντες ἐν τῃ γῃ,
            πνεῦμα καί ὑδωρ καί αιμα, καί οἱ τρεῖς ἐισ τὸ εν εἰσιν.  

It looks as if Erasmus, at some point, added two articles (ὁ, the), one conjunction (καί, and), and the final phrase.  In minuscule 629, the articles and the conjunction are absent, as in minuscule 61.  Thus, it seems difficult to maintain that the Greek base-text of First John 5:7 that is found in the KJV is extant in the text of any Greek manuscript not copied from a printed edition of the Textus Receptus.

The takeaway from all this is that the Comma Johanneum was not part of the original text of First John; it began as a Latin interpretative note on verse 8, after the word-order in verse 8 had been altered.  Furthermore, hardly any Greek manuscripts of First John contain the CJ, and the ones that have it in their text have variations indicating that the few late Greek manuscripts that support the CJ do so because of contamination from Latin copies, not because they echo earlier Greek copies.  The CJ does not belong in the New Testament. 


Scripture attributed to the New American Standard Bible® is Copyright © 1960, 1962, 1963, 1968, 1971, 1972, 1973, 1975, 1977, 1995 by The Lockman Foundation.  Used by permission.

Tuesday, August 8, 2017

Hand-to-Hand Combat! Sinaiticus vs. Minuscule 2474

The first page of the Gospel of John
in the Elfleda Bond Goodspeed Gospels
            It’s time for some hand-to-hand combat!  Today’s contest takes place in Matthew 24:23-30, a passage which contains some of Jesus’ instructions to His disciples regarding false prophets and signs of the end of the world.  The combatants:  Codex Sinaiticus – which is universally acknowledged as a very important manuscript, having been produced c. 350 – and minuscule 2474, known as the Elfleda Bond Goodspeed Gospels.
            Among New Testament manuscripts, Codex Sinaiticus is among the most famous.  As for minuscule 2474, few people knew of its existence until early in 1952, when it was purchased in Istanbul (Constantinople) by the American collector Harry Kurdian of Wichita, Kansas; it was acquired by Edgar J. Goodspeed later the same year.  He added it to the manuscript-collection at the University of Chicago in honor of his wife, who had died in 1949. 
            Let’s take a minute to learn something about Elfleda Bond Goodspeed.  Her unusual first name is the same as that of a British saint (and friend of Saint Cuthbert) of the 600’s.  Mrs. Goodspeed was born in 1880, at about the same time when her father, Joseph Bond, after being diagnosed with a debilitating condition, prayed to receive 20 years of life.  His prayer was answered; though still far from a state of strong physical health, he used that time to develop a highly profitable home-radiator business, and in 1901, one year before his death, his daughter Elfleda married Edgar Goodspeed.  Elfleda Bond Goodspeed was considered worthy not only of the honor of having a Greek Gospels-manuscript named in her honor, but in addition, on the campus of the University of Chicago, if one visits the Joseph Bond Chapel, one can see the exquisite stained-glass windows which her husband donated in her memory. 
            And now, on to the combat! 
            I will examine each manuscript’s text of these eight verses using the same standard of comparison:  each will be compared to the text in the 27th edition of the Nestle-Aland compilation; each non-original letter will be noted, each lack of an original letter will be noted; transpositions will be mentioned but not considered a loss or gain; contractions of sacred names and of the word και (“and”) will not be counted as omissions.  After the exhaustive comparison, another comparison will be made in which itacisms (minor interchanges of vowels) are removed from consideration.

Let’s see how the copyist of Codex Sinaiticus did.

23 – no variation.
24 – ﬡ has ψευδοπροφητε instead of ψευδοπροφηται (+1, -2)
24 – ﬡ has σημια instead of σημεια (-1)
24 – ﬡ does not have μεγαλα (-6)
24 – ﬡ has πλανηθηναι instead of πλανησαι (+3, -1)
25 – no variation.
26 – ﬡ does not have ουν (-3)
26 – ﬡ has ταμιοις instead of ταμειοις (-1)
27 – ﬡ has εξερχετε instead of εξερχεται (+1, -2)
27 – ﬡ has φαινετε instead of φαινεται (+1, -2)
27 – ﬡ has εστε instead of εσται (+1, -2)
[28 – ﬡ does not have the letter ο at the beginning of the verse (at the start of οπου); however a correction has been made, possibly by the proofreader of the manuscript, so this will not be included in the total.]
28 – ﬡ has σωμα instead of πτωμα (+1, -2); a correction has been made, but it is post-production, so this will be included in the total.
28 – ﬡ has εκι instead of εκει (-1)
28 – ﬡ has συναχθησοντε instead of συναχθησονται (+1, -2)
29 – ﬡ has εκινων instead of εκεινων (-1)
29 – ﬡ has σκοτισθησετε instead of σκοτισθησεται (+1, -2)
29 – ﬡ has δωσι instead of δωσει (-1)
29 – ﬡ has εκ instead of απο (+2, -3)
29 – ﬡ has δυναμις instead of δυναμεις (-1)
30 – ﬡ has φανησετε instead of φανησεται (+1, -2)
30 – ﬡ has σημιον instead of σημειον (-1)
30 – ﬡ does not have the second τοτε (-4); the word is added above the line but this appears to be post-production.
30 – ﬡ has κοψοντε instead of κοψονται (+1, -2)
[30 – ﬡ has πασε instead of πασαι.  A correction was made above the line; this correction looks like it was made by the proofreader, so this will not be included in the total.]
[30 – ﬡ has ε instead of αι.  A correction was made above the line; this correction looks like it was made by the proofreader, so this will not be included in the total.]
30 – ﬡ has οψοντε instead of οψονται (+1, -2)

            Thus, in the course of Matthew 24:23-30, Sinaiticus’ text displays 59 letters’ worth of corruption, consisting of the addition of 15 non-original letters, and the loss of 44 original letters.  This does not reflect well on the copyist.  However, many of the alterations in the text consist of small orthographic variations, within a word, he often wrote ι instead of ει, and at the end of a word he often wrote ε instead of αι.  If these orthographic quirks are removed from the equation, then the variations in ﬡ look more like this – 
            ● 24 – ﬡ does not have μεγαλα (-6)
            ● 24 – ﬡ has πλανηθηναι instead of πλανησαι (+3, -1)
            ● 26 – ﬡ does not have ουν (-3)
            ● 28 – ﬡ has σωμα instead of πτωμα (+1, -2)
            ● 29 – ﬡ has εκ instead of απο (+2, -3)
            ● 30 – ﬡ does not have the second τοτε (-4) –
which yields more respectable results:  if we ignore itacisms, Sinaiticus’ text in Matthew 24:23-30 has 25 letters’ worth of corruption, consisting of the introduction of six non-original letters and the loss of 19 original letters.

Now let’s see how the scribe of minuscule 2474 did.

23 – 2474 has πιστεύσηται instead of πιστεύσητε (+2, -1) 
24 – 2474 has has ψευδοπροφητε instead of ψευδοπροφηται (+1, -2)
25 – no variation.
26 – no variation.
27 – no variation.
28 – 2474 has γαρ after οπου (+3)
29 – no variation.
30 – 2474 has τω before ουρανω (+2)
30 – 2474 has οψοντε instead of οψονται (+1, -2)

            Thus, in the course of Matthew 24:23-30, 2474’s text contains 14 letters’ worth of corruption, consisting of the inclusion of nine non-original letters and the non-inclusion of five original letters.  If we remove itacisms from the equation, as was done with the text in Sinaiticus, then the corruptions in 2474 in Matthew 24:23-30 consist of:
            ● the inclusion of γαρ after οπου in verse 28, and
            ● the inclusion of τω before ουρανω in verse 30. 

            From this comparison, it may be concluded that in the transmission-stream of Codex Sinaiticus, 59 letters’ worth of corruption were introduced in the course of 280 years (positing the composition of the Gospel of Matthew in A.D. 70, and the production of Codex Sinaiticus in A.D. 350), which yields an ACR (Annual Corruption Rate) of .21 – that is, it implies that copyists in the Alexandrian transmission-stream were producing, on average, .21 letters’ worth of corruption each year.
            When itacisms are removed from the equation, over half of the corruptions in Matthew 24:23-30 in Sinaiticus are also removed; it then has only 25 letters’ worth of corruption and its transmission-stream’s ACR drops to .09.  
            Meanwhile, granting a production-date for 2474 around A.D. 950, the 14 letters’ worth of corruption in minuscule 2474’s text of Matthew 24:23-30 imply that as far as the text of Matthew 24:23-30 is concerned, its transmission-stream’s ACR is only .016.  When itacisms are removed from the equation, the ACR of the transmission-stream of 2474 drops to .0057.

To review:
            Sinaiticus’ text of Matthew 24:23-30 has 59 letters’ worth of corruption.  Removing itacisms from consideration, it has 25 letters’ worth of corruption.  This implies that, on average, copyists in the transmission-stream that produced this text added .09 letters’ worth of corruption each year, besides itacistic readings.
            Minuscule 2474’s text of Matthew 24:23-30 has 14 letters’ worth of corruption.  Removing itacisms from consideration, it has 5 letters’ worth of corruption.  This implies that, on average, copyists in the transmission-stream that produced this text added .016 letters’ worth of corruption each year, besides itacistic readings. 
            No matter which set of figures one uses (with, or without, itacisms), this analysis shows that in this contest, the much younger manuscript has a much better text, and that the Byzantine copyists in its transmission-line worked far more carefully than the Alexandrian copyists in the transmission-line of Codex Sinaiticus.  The copyists in the Byzantine transmission-stream that produced the text of 2474 were at least five times better at avoiding corruption than the copyists in the Alexandrian transmission-stream that produced the text of Codex Sinaiticus.  (When itacisms are removed from consideration, and the standard of comparison is the Robinson-Pierpont Byzantine Textform, rather than the Nestle-Aland compilation, the amount of corruption in the Elfleda Bond Goodspeed Gospels in Matthew 24:23-30 drops to zero.)

Saturday, August 5, 2017

More Cracks in Nestle-Aland 28 (Acts-Revelation)

            In the previous post, I described some passages in the Gospels where the rival variants may receive different treatment in future editions of the critical text of the Greek New Testament, and/or in English translations based on it – including a few passages where the editors may adopt readings with no Greek manuscript-support, as the editors of the 28th edition of the Nestle-Aland Novum Testamentum Graece recently did in Second Peter 3:10.  Today, let’s look at a dozen passages in the rest of the New Testament which may be similarly vulnerable to the effects of thoroughgoing eclecticism.

Acts 6:9 – The scholar Friedrich Blass (1843-1907), in the course of his detailed study of the Greek text of the Gospels and Acts, detected something abnormal about the mention of Libertines in Acts 6:9:  why did Luke resort to Λιβερτίνων, a Latin-based term, rather than simply write Ἀπελεύθεροι?  And why, followed by various geographically based terms, is this one not also geographically based? 
            Such questions elicited a search for answers.  Blass discovered that he was not the first reader to hum upon encountering the term Λιβερτίνων in this verse.  A long line of researchers, going all the way back to Beza, had sensed that something about this word was amiss. 
            Blass was informed by J. Rendel Harris that in the Armenian version, the reference was not to Libertines, but to Libyans.  How, though, could a reference to Libyans ever be misconstrued as Libertines?  And if the Armenian version’s base-text had referred to Libertines, how did the Armenian version end up with Libyans?  With remarkable determination, Blass dug a little deeper into this puzzle, and discovered, among the Latin poems of Catullus, the use of a rare term that satisfied his curiosity; transliterated into Greek, it is Λιβυστίνων – inhabitants of the area west of Cyrene.      
            This conjectural emendation resembles the extant text; it fits the context, it makes sense, and it introduces nothing problematic.  Future editors may reason that the rareness of the term Λιβυστίνων provoked early copyists to misread it, and also provoked the translators of the Armenian version to loosely approximate its meaning with a more familiar term.       

● Acts 7:46 – Working within the extant evidence, textual critics must choose between the statement that David asked to be allowed to find a dwelling-place for the God of Jacob (a reading supported by Codices A, C, E, 1739, the vast majority of manuscripts, and broad versional evidence), or a dwelling-place for the house of Jacob (which is the reading in the Nestle-Aland compilation, and which is supported by a small cluster of early manuscripts, including Papyrus 74, Vaticanus, Sinaiticus, and Bezae). 
            The Alexandrian reading is certainly more difficult, because it seems to say that David asked to build a house for a house.  Even when the second “house” is understood to refer to the nation descended from Jacob, the problem does not go away, since the temple was for God, not for the people, who were not looking for a new residence in the days of David. 
            The reading οἴκω (“house”) has been considered too difficult by some textual critics, including Hort, who wrote in 1881, “οἴκω can hardly be genuine,” but rather than accept the Byzantine reading, he proposed that probably neither reading is original.  Instead, he conjectured that the original text was τω Κυριω (“the Lord”), which was contracted to ΤΩ ΚΩ, which was misread by inattentive copyists as ΤΩΟΙΚΩ.  If future editors of the Nestle-Aland compilation are unwilling to adopt Hort’s conjecture, they might at least acknowledge the force of his admission of the implausibility of the Alexandrian reading, and adopt the other reading, for which the diversity of the external support is very impressive.        

Acts 12:25 – In the description of the action taken by Barnabas and Saul in this verse, there is a four-horse race, so to speak: 
            ἀπο Ἰερουσαλὴμ (“from Jerusalem”), supported by Codex D, Ψ, 614, several Old Latin copies, a significant minority of Byzantine manuscripts, the Vulgate, et al.
            εἰς Ἰερουσαλὴμ (“to Jerusalem”), supported by Codices א, B, and most Byzantine manuscripts.
            ἐξ Ἰερουσαλὴμ (“from Jerusalem”), supported by Papyrus 74, Codex A, et al.                       
            ἐξ Ἰερουσαλὴμ εἰς Ἀντιόχιαν (“from Jerusalem to Antioch”), supported by Codex E, 1739, the Peshitta, the Sahidic version, et al.
            The most difficult option is the variant with εἰς, because (1) when last seen in the narrative (in 11:30), Paul and Barnabas were already going to Jerusalem, and (2) in the very next scene (at the beginning of chapter 13), Paul and Barnabas are present at Antioch, not at Jerusalem.  Even though εἰς is in the Nestle-Aland compilation, some translators of modern versions have rejected this reading; for example, the NASB states, “And Barnabas and Saul returned from Jerusalem,” implying a base-text with either ἀπο or ἐξ.  The ESV reads identically (as of 10:00 p.m., August 4, 2017):  “And Barnabas and Saul returned from Jerusalem.”  The NIV also rejects εἰς, stating, “When Barnabas and Saul had finished their mission, they returned from Jerusalem.” 
            Nobody (other than avid advocates of the Peshitta) seems to think that the longest reading is original, because it looks like just the sort of textual adjustment that a copyist might make to alleviate a difficulty.  The contest, then, is between εἰς, ἐξ, and ἀπο.  Theoretically, if, in a scriptorium where a group of copyists worked from dictation, their supervisor read ἐξ, a copyist could mishear it as εἰς – but the theory works as well in the opposite direction. 
            Thoroughgoing eclecticism turns the race into a five-horse contest.  Hort suggested in 1881 that the original word-order has been garbled by copyists, and that the original text read τὴν εἰς Ἰερουσαλὴμ πληρώσαντες διακονίαν, so as to merely report that Barnabas and Saul returned, having completed their service in Jerusalem.  F. F. Bruce, whose confident comments about the reliability of New Testament manuscripts have been thoroughly recycled by many apologists, did not refuse to embrace a conjectural emendation in this passage; he held that in the original text there was no prepositional phrase at all, and that marginal glosses have impacted the text in Acts 12:25 in all extant manuscripts.
            Although textual critics often regard a higher degree of difficulty as a quality of the most-likely-original reading, it is possible that future editors may regard the currently printed reading here as simply too difficult, and either adopt ἐξ, or adopt ἀπο, or resort to conjectural emendation.     

Acts 20:28 – Bruce Metzger dedicated a full two pages of his Textual Commentary to a consideration of this passage.  The initial question is, did the original text refer to “the church of God,” or to “the church of the Lord,” or to “the church of the Lord and God”?  Like John 1:18, the contest between “God” and “Lord” is a contest amounting to the difference of a single letter, once one accounts for the contraction of sacred names:  Θεοῦ (“of God”) becomes ΘΥ and Κυρίου (“of the Lord”) becomes ΚΥ.
            If that contest is decided in favor of Θεοῦ (on the grounds that this is supported by ﬡ, B, the Vulgate, the Peshitta, and a significant minority of Byzantine manuscripts, including the Textus Receptus), then a second question arises:  did Luke report that Paul stated that God purchased the church with His own blood?  Many an apologist has used this verse to demonstrate Paul’s advocacy of the divinity of Christ, inasmuch it was neither the Father, nor the Spirit, whose blood was shed.  Hort, however, expressed a suspicion (which, it seems, was first expressed in 1797 by Georg Christian Knapp) that at the end of the verse, following the words διὰ τοῦ αἴματος τοῦ ἰδίου (“through His own blood”), there was originally the word υἱοῦ (“Son”).      
            It is possible that future editors, may decide that the inclusion of υἱοῦ in this verse is required by internal evidence, and that it is feasible that the word υἱοῦ was accidentally lost very early via a common parableptic error (when a copyist’s line of sight drifted from the letters ΙΟΥ at the end of ἰδίου to the same letters at the end of υἱοῦ.  Already, the Contemporary English Version, advertised as “an accurate and faithful translation of the original manuscripts,” has the word “Son” in its text of Acts 20:28b:  “Be like shepherds to God’s church.  It is the flock that he bought with the blood of his own Son.”  A footnote informs the CEV’s readers about the meaning of the extant text.
First Corinthians 6:5 – The Greek evidence, from Papyrus 46 to the Textus Receptus, is in agreement about how this verse ends.  However, the Peshitta – a Syriac version traceable to the late 300’s (followed by a period of standardization), but possibly earlier – disagrees.  The reading in the Peshitta implies that its Greek base-text included the phrase καὶ τοῦ ἀδελφοῦ (“and a brother”). 
            The momentum for this reading is drawn from a grammatical oddity in the usual Greek text.  The first part of Paul’s statement in this verse is something to the effect of, “Is there not even one person among you – just one! – who shall be able to judge between” – and that’s where the difficulty appears, because the Greek text just mentions one brother, whereas the idea of judgment between two parties seems to demand that more than one brother should be mentioned. 
            The KJV’s translators, although the Textus Receptus reads ἀνα μέσον τοῦ ἀδελφοῦ αὐτοῦ (“between his brother” – which is clearly singular), concludes the verse with “between his brethren” (which is clearly plural).  The NET is similar (“between fellow Christians”); the difference is due to the NET’s enlightened gender-neutral treatment of the term ἀδελφοῦ, not to any new feature in the Greek base-text.  The CSB, the NIV, and even the NASB likewise render the text as if the verse ends with a plural word rather than a singular one.  All such treatments of the text make the problem all the obvious:  the first part of the sentence, in Greek, anticipates two brothers, while the second part of the sentence mentions only one.             
            In light of such strong internal evidence, Michael Holmes, the compiler of the SBLGNT, recommended the adoption of a conjectural emendation at this point, so that καὶ τοῦ ἀδελφοῦ (“and the brother”) appears at the very end of the verse.  It is possible that future compilers of the Nestle-Aland text will concur.  If that happens, it will have hardly any effect on English translations, most of which already translate the passage as if the wording proposed by Dr. Holmes is extant in the manuscripts.   

Galatians 4:25 – For almost 300 years, a scholarly debate has orbited part of this verse.  The phrase “Now this Hagar is Mount Sinai in Arabia” is presently in the Nestle-Aland compilation (as τὸ δὲ Ἁγὰρ Σινᾶ ὅρος ἐστιν ἐν τῇ Ἀραβίᾳ, though this is contested by four slightly different rival forms).  However, it has been proposed that the entire phrase originated as a marginal note, and does not belong in the text.  This conjecture goes back at least as far as Richard Bentley (a gifted British cleric, 1662-1742, who advanced the field of New Testament textual criticism more than anyone else in his generation).  Recently Stephen Carlson, who has conducted a stemmatics-based analysis of the text of Galatians, has argued in favor of the same idea.  (Robert Waltz, however, retained the phrase in his compilation of the text of Galatians.)  If future editors of NTG concur with Carlson, the phrase might be exiled to the footnotes.

A Syriac manuscript at Saint
Catherine's Monastery displays
an adjustment of the text of
Hebrews 2:9.
(Charley Ellis pointed out this
feature of the MS to me.)
● Hebrews 2:9 – In Bart Ehrman’s 2005 book Misquoting Jesus, the author noted that instead of reading χάριτι θεοῦ (“by the grace of God”), a smattering of witnesses supports χωρὶς θεοῦ (“without God”).  Origen, in the 200’s, was aware of both variants.  Opposing an array of witnesses that includes Papyrus 46, Vaticanus, Sinaiticus, Claromontanus, the Byzantine Text, and broad versional support, Ehrman proposed that χωρὶς θεοῦ was the original reading, and that an early copyist altered the text so that it said something less provocative – and this unknown copyist was so influential that his alteration has affected almost all extant Greek copies of Hebrews.
            As Ehrman noted in his book, the existence of the variant χωρὶς θεοῦ has been accounted for by some textual critics via the idea that it was written in the margin by someone who intended for the phrase to be a qualification of the sentiment of the preceding verse – the idea being that all things except God were subject to the authority of Christ – an exception mentioned by Paul in First Corinthians 15:27.  Resisting this proposal, Ehrman objected that if this had been an annotator’s intent, “Would he not have written “except for God” (EKTOS THEOU – the phrase that actually occurs in the I Corinthians passage) rather than “apart from God (CHŌRIS THEOU – a phrase not found in I Corinthians)?” 
            Ehrman’s objection loses much force when one observes that the phrase EKTOS THEOU does not, in fact, occur anywhere in First Corinthians.  Ehrman also overstates the evidence when he claims that “Origen tells us that this [χωρὶς θεοῦ] was the reading of the majority of manuscripts in his own day,” for Origen cites this reading and then says that some copies have the other reading, χάριτι θεοῦ; nowhere does Origen say that his collection of manuscripts at Caesarea was typical of the manuscripts of Hebrews that existed throughout the world.  Most readers of Ehrman’s book, however, will probably not double-check his confidently worded assertions.
            Centuries ago, the devout scholar John Bengel (1687-1752) cautiously favored the reading χωρὶς θεοῦ and argued that its meaning is not scandalous, but theologically profound.  Bengel proposed that it was intended to mean that the Son of God, and not God the Father, tasted death for everyone – an idea that is consistent with the text of Hebrews 1:3, where, in Papyrus 46 and the Byzantine Text, Jesus is said to have made atonement by Himself
            Another theory, wounded but not killed by the grammatical quirk that it involves, proposes that χωρὶς θεοῦ is original, and that the author intended to thus qualify the words ὑπὲρ παντὸς (“for all”), so as to convey the idea that Christ tasted death for everyone except God.  This is how Origen interpreted this variant, without dogmatically deciding in favor of either variant.
            Meanwhile, F. F. Bruce proposed an alternative solution:  he conjectured that χωρὶς θεοῦ originated as a note in the margin, and that subsequently a copyist replaced that note with one that read χάριτι θεοῦ, and that both readings have slid into the text – in other words, Bruce suspected that neither phrase is original! 

● Hebrews 11:11 –As the list of variants in the textual apparatus of the STEP-Bible shows, the textual racetrack in Hebrews 11:11 is crowded with rival variants.  This sort of contest is difficult for the Coherence-Based Genealogical Method – the newly developed mapping-program used by some of the Nestle-Aland editors – to handle.  Researcher J. Harold Greenlee (a scholar in the same league as Bruce Metzger), proposed that the original text of this verse did not contain σπεῖρα or ἡ σπεῖρα (that is, it did not specifically say that Sarah was barren).  Michael Holmes’ SBLGNT likewise does not have σπεῖρα or ἡ σπεῖρα in its text. 
            This constitutes preference for a Byzantine reading (supported by ﬡ, A, and minuscules 33 and 1175, et al).  The effect of this textual decision (and some nuanced syntax-related translational decisions) can be seen via a comparison of the text of the NIV 2011 (which adopts it) and the text of the 1973 NIV (which does not):
            NIV 1973:  “By faith Abraham, even though he was past age – and Sarah herself was barren – was enabled to become a father because he considered him faithful who had made the promise.” 
            NIV 2011:  “And by faith even Sarah, who was past childbearing age, was enabled to bear children because she considered him faithful who had made the promise.”
            Whether future editors will continue this trend remains to be seen, particularly because the loss of σπεῖρα can be attributed to parableptic error, when a copyist’s like of sight drifted from the last two letters of Σάρρα (“Sarah”) to the same two letters in σπεῖρα (“barren”).    
● Hebrews 11:37 – In the list of the sufferings of spiritual heroes, one of those things is not like the others:  they are all somewhat unusual experiences, except for ἐπειράσθησαν, “they were tempted.”  Some textual critics have suspected that this word originated when a copyist committed dittography – writing twice what should be written once; in this case, the preceding word ἐπρίσθησαν (“they were sawn in two”), and that subsequent copyists, not realizing the mistake in their exemplar, changed it into something meaningful.  Others have thought that this relatively common term replaced one that was less common – perhaps ἐπάθησαν (“they were pierced”) or ἐπράσθησαν (“they were sold”). 
            Presently the Nestle-Aland compilation, deviating from the 25th edition, simply does not include ἐπειράσθησαν in the text, adopting instead the reading of Papyrus 46, which is very ancient.  Papyrus 13, however, is also very ancient, and appears to support the inclusion of ἐπειράσθησαν, in which case it has a very impressive array of allies.  I would advise readers to not get used to the NTG’s current form of this verse, for it seems to be merely a place-holder that might be blown away by the appearance of even the slightest new evidence, and even by an intrinsically appealing conjecture.    

● First Peter 3:19 – What may be the most popular conjectural emendation of all time was favored by the erudite textual expert J. Rendel Harris (1852-1941), who encountered a form of it in William Bowyer’s 1782 book Critical Conjectures and Observations on the New Testament.  The extant text of First Peter 3:19 says, “in which he went and proclaimed to the spirits in prison.”  Verse 18 refers to Christ, and nobody else is introduced into the text, so verse 19 has been interpreted to mean that during the time between Jesus’ death and resurrection, He visited the realm of the dead – specifically visiting the spirits of those who had been disobedient in the days of Noah, prior to the great flood – and delivered a message (ἐκήρυξεν) to them. 
            Harris, however, proposed that the original text was different.  He thought that Peter had in mind a scene that is related in the pseudepigraphical amalgamation known as the Book of Enoch (the first section of which is quoted by Jude in verses 14-15 of his epistle), in which Enoch is depicted delivering a message of condemnation to the fallen spirits who corrupted human beings so thoroughly that the great flood was introduced as the means of amputating the moral infection they had induced.
            Specifically, what Harris proposed was that the opening words of the original text of 3:19 were ἐν ᾧ καὶ Ἐνώχ (“in which also Enoch”), thus assigning the subsequent action not to Christ, but to Enoch.  (A variation on this idea is that the original text read Ἐνώχ instead of ἐν ᾧ καὶ.) 
            (It should, perhaps, be noted that Irenaeus, in Against Heresies Book 4,16:2, took for granted the veracity of the tradition that Enoch had brought God’s message to fallen angels; these fallen angels being the “sons of God” mentioned in Genesis 6:2-4.)
            How could the name “Enoch” have fallen out of the sentence?  In two ways:
            1.  If the original text were simply Ἐνώχ (without ἐν ᾧ καὶ), then, in uncial letters, the χ was susceptible to being misread as a και-compendium (that is, a common abbreviation for the word και (“and”)).  A copyist could easily decide to write the whole word instead of the abbreviation, and thus Enoch’s name would become ἐν ᾧ καὶ.
            2.  If the original text were ἐν ᾧ καὶ Ἐνώχ, a copyist, reading the χ as a και-compendium, could assume that the scribe who made his exemplar had inadvertently repeated three words, and, attempting a correction, remove “Ἐνώχ.”
            Against the charge that the introduction of Enoch’s name “disturbs the otherwise smooth context” (as Metzger claimed in 1963) the answer may be given that a reference to Enoch is not out of place, inasmuch as Enoch’s story sets the stage for the story of Noah and his family, whose deliverance through water Peter frames as a sort of pattern of the salvation of the church.
               Future compilers willing to engage in conjectural emendation might consider the internal arguments in favor of the inclusion of Enoch’s name in First Peter 3:19 to be too attractive to resist.  If that turns out to be the case, then it would certainly have some doctrinal impact, significantly diminishing the Biblical basis for the phrase “He descended into hell” found in the Apostles Creed.” 

● Jude verses 22-23 – Even though Tommy Wasserman has collated every known Greek manuscript of the book of Jude, plenty of questions remain about how that data should be interpreted.  Like Hebrews 11:11, verses 22-23 of the Epistle of Jude verse have multiple rival variants.  Here I shall spare readers the fine details of the case, and simply note that it is possible that future editors may discern here a threefold command, and that the first command from Jude is to refute (ἐλέγχετε) those who cause divisions.  Some copyists, with the earlier mention of mercy in verse 21 (ἔλεος) fresh in their minds, may have allowed their memory of it to complete the similarly started word in verse 22. 
            The reading with ἐλέγχετε is supported not only by Codices A and C* but also by members of manuscript-clusters represented by  minuscule 1739 and minuscule 1611; both of these clusters have special weight (as the echoes of distinct and ancient transmission-lines) in the Catholic (General) Epistles.  Even though the Nestle-Aland editors have already sifted through the text of Jude, they might take another look at this variation-unit. 

● Revelation 7:6 More than one commentator on the book of Revelation has admitted to being puzzled by a feature of the list of the twelve tribes of the children of Israel:  inasmuch as the tribe of Joseph is included, why is the tribe of Manasseh (Joseph’s son) also listed, but not Ephraim?  Another question:  why, in the extant Greek manuscripts of Revelation 7:6, is Manasseh’s name spelled in so many different ways?  
            A conjectural emendation answers both of those questions:  in the original text, the last portion of verse 6 did not refer to the tribe of Manasseh, but to the otherwise unmentioned tribe of Dan, and an early copyist misread ΔΑΝ as ΜΑΝ, and understood it to be an abbreviation of Manasseh’s name.  Thus ΜΑΝ originated, and different copyists with different orthography proceeded to spell out the name.  The Bohairic version supports this idea; in this verse the Bohairic text does not refer to Manasseh, but to Dan.  Perhaps future advocates of thoroughgoing eclecticism, on the strength of the Bohairic reading’s intrinsic appeal, will bring ΔΑΝ into the text.

In Conclusion . . . 

            Some readers, looking over these passages, and the passages from the Gospels described in the preceding post, may feel a measure of consternation, particularly because seven of them – in Matthew 1:16, Matthew 28:19, Mark 1:1, John 1:18, Acts 20:28, Hebrews 2:9, and First Peter 3:19 – have been used as a basis for establishing doctrine.  However, only in the case of First Peter 3:19, and the teaching that Christ visited imprisoned spirits, could it be argued that a doctrine stands or falls on the acceptance or rejection of a particular reading or conjecture (and even then, a case could be made that Paul teaches essentially the same doctrine in Ephesians 2:9-10, minus the specificity in First Peter).  
            (Apologists for Islam may sense a different sort of consternation, inasmuch as even with the allowance of conjectural emendation in the picture, the application of thoroughgoing eclecticism elicits nothing remotely close to the level of textual alteration that would bring the doctrinal teachings of the New Testament into harmony with the teachings of the Quran.  The charge, often made by Muslim apologists, that the New Testament agreed with the Quran until Christian copyists altered the text of the New Testament, simply lacks a historical foundation.)      
              It is sometimes said (because Bruce Metzger said it) that New Testament textual criticism is both an art and a science.  But it should be all science, and not art, because it is an enterprise of reconstruction, not construction.  Its methods may validly be creative and inventive – even intuitive – but not its product.  Conjectural emendation is the only aspect of textual criticism that involves the researcher’s artistic or creative skill. 
            No conjectural emendation should ever be placed in a compilation of the text of the Greek New Testament.  At the same time, the task of proposing possible readings which account for their rivals, or which otherwise resolve perceived oddities in the extant text, serves a valuable purpose:  to demonstrate the enormous weight of the intrinsic evidence in favor of such readings in the event that they are ever discovered in a Greek manuscript. 


Quotations from the ESV have been taken from The Holy Bible, English Standard Version. ESV® Permanent Text Edition® (2016). Copyright © 2001 by Crossway Bibles, a publishing ministry of Good News Publishers.

Quotations from the NIV (2011 edition) have been taken from the Holy Bible, New International Version®, NIV® Copyright ©1973, 1978, 1984, 2011 by Biblica, Inc.®  Used by permission. All rights reserved worldwide.

Scripture quotations marked CSB have been taken from the Christian Standard Bible®, Copyright © 2017 by Holman Bible Publishers.  Used by permission.  Christian Standard Bible® and CSB® are federally registered trademarks of Holman Bible Publishers.

Scripture quotations marked NASB taken from the New American Standard Bible®, Copyright © 1960, 1962, 1963, 1968, 1971, 1972, 1973, 1975, 1977, 1995 by The Lockman Foundation.  Used by permission.

Scripture quotations marked CEV are taken from the Contemporary English Version (CEV), Copyright ©  1995 by American Bible Society.

Quotations designated NET are from the NET Bible® copyright ©1996-2006 by Biblical Studies Press, L.L.C. All rights reserved.

The NIV (1973 edition, no longer in print) is Copyright © 1973 by New York Bible Society International, and published by The Zondervan Corporation, Grand Rapids, Michigan 49506, USA.

Misquoting Jesus:  The Story Behind Who Changed the Bible and Why, is Copyright © 2005 by Bart D. Ehrman, and published by HarperCollins Publishers, New York, New York.  All rights reserved.