무료

수료증 무료 발급 — 2027년 3월 31일까지.

BT604E Reading the Biblical Languages

Lecture 12-1: The Apparatus: Reading a Textual Note Without Being Misled

목차 37
BT604E Reading the Biblical Languages
제 0 장 · Introduction: Why Read the Originals
  1. Introduction: Why Read the Originals
제 1 장 · Chapter 1: Reading, Not Decoding
  1. Lecture 1-1: What This Course Assumes, and What It Is For
  2. Lecture 1-2: How to Work Through a Verse Without Drowning
  3. Lecture 1-3: The Tools on Your Desk: Lexicon, Grammar, Concordance, Apparatus
제 2 장 · Chapter 2: Hebrew Narrative - Genesis
  1. Lecture 2-1: The Shape of a Hebrew Sentence: Word Order and the Narrative Chain
  2. Lecture 2-2: Reading Genesis 22:1-14
  3. Lecture 2-3: Repetition, Wordplay, and What Translation Cannot Carry
제 3 장 · Chapter 3: Hebrew Verb and Discourse - 1 Samuel
  1. Lecture 3-1: The Verb System in Action: Aspect, Sequence, and the Perfect Problem
  2. Lecture 3-2: Reading 1 Samuel 17 as Connected Discourse
  3. Lecture 3-3: Direct Speech, Particles, and the Texture of Dialogue
제 4 장 · Chapter 4: Hebrew Law - Deuteronomy
  1. Lecture 4-1: Legal Hebrew: Casuistic and Apodictic Formulations
  2. Lecture 4-2: Reading Deuteronomy 6:4-9
  3. Lecture 4-3: Motive Clauses and the Rhetoric of Command
제 5 장 · Chapter 5: Hebrew Poetry - Psalms
  1. Lecture 5-1: Parallelism, and Why It Is Not Repetition
  2. Lecture 5-2: Reading Psalm 23 and Psalm 1
  3. Lecture 5-3: Poetic Vocabulary, Rare Words, and the Limits of the Lexicon
제 6 장 · Chapter 6: Hebrew Prophecy - Amos
  1. Lecture 6-1: Prophetic Forms: Oracle, Woe, Vision Report
  2. Lecture 6-2: Reading Amos 5:18-27
  3. Lecture 6-3: When the Text Is Hard: Corruption, Conjecture, and the Apparatus
제 7 장 · Chapter 7: Greek Narrative - Mark
  1. Lecture 7-1: Koine on the Page: Mark's Sentences and Why They Read as They Do
  2. Lecture 7-2: Reading Mark 4:35-41
  3. Lecture 7-3: Aspect Over Time: The Present, Aorist and Imperfect at Work
제 8 장 · Chapter 8: Greek Historiography - Luke and Acts
  1. Lecture 8-1: A More Literary Koine: Luke's Preface and the Septuagintal Style
  2. Lecture 8-2: Reading Luke 15:11-24
  3. Lecture 8-3: Participles: The Engine of Greek Narrative
제 9 장 · Chapter 9: Greek Letter - Galatians
  1. Lecture 9-1: Reading a Letter as Argument: Connectives and Signposts
  2. Lecture 9-2: Reading Galatians 2:15-21
  3. Lecture 9-3: Cases in Dispute: The Genitive and What Hangs on It
제 10 장 · Chapter 10: Greek Argument - Romans
  1. Lecture 10-1: Long Sentences: Subordination, Anacoluthon, and Keeping the Thread
  2. Lecture 10-2: Reading Romans 3:21-26
  3. Lecture 10-3: Conditional Sentences, Questions, and the Diatribe
제 11 장 · Chapter 11: The Greek of John
  1. Lecture 11-1: Simple Vocabulary, Difficult Meaning: The Johannine Style
  2. Lecture 11-2: Reading John 1:1-14
  3. Lecture 11-3: The Article, and What It Does and Does Not Prove
제 12 장 · Chapter 12: Text, Words, and Honest Use
  1. Lecture 12-1: The Apparatus: Reading a Textual Note Without Being Misled
  2. Lecture 12-2: Word-Study Fallacies and How to Avoid Them
  3. Lecture 12-3: Keeping the Languages After the Course Ends

핵심만 먼저 보기 ↓

52분

At the foot of every page of the Nestle-Aland Greek New Testament and the Biblia Hebraica Stuttgartensia there is a block of small type full of symbols. Most students who own these editions never learn to read it. They treat the printed text above the line as simply the text, and the block below as a technical matter for specialists.

That is the wrong way round. The block below the line is the editors telling you, in compressed form, that the text above it is a series of decisions — and inviting you to check them. A printed critical text is not a manuscript. No single surviving manuscript reads exactly what Nestle-Aland prints, and none reads exactly what BHS prints either. Both are reconstructions, and both are honest enough to show their working.

This lecture teaches you to read that working. It will not make you a textual critic; that requires palaeography, a knowledge of the manuscript families, and Coptic, Syriac and Latin, none of which this course has taught. What it will do is make you able to open an apparatus, see what the competing readings are, see what kind of evidence stands behind each, and follow — or refuse — the argument somebody else makes from it. That is a defensive skill above all. The single most common misuse of textual criticism in preaching and popular writing is a confident claim about "the oldest and best manuscripts" made by someone who has not opened an apparatus in their life.

The lecture then works four real cases to a judgement: one very small, one structural, and the two largest in the New Testament. Real cases are better than invented ones, because the invented case always comes out tidy.

이 52분으로 얻는 것

  • Identify the main classes of witness cited in a New Testament apparatus — papyri, majuscules, minuscules, lectionaries, versions, patristic citations — and say what each is good for.
  • Read a single entry in the apparatus of NA28 and of BHS, naming the reading printed in the text, the competing reading, and the witnesses adduced.
  • State the principal external and internal criteria of textual criticism, and give a case in which each has been shown to fail.
  • Set out the evidence on Mark 16:9-20 and John 7:53-8:11 and explain why the great majority of critics judge both to be later additions, while also explaining what that judgement does and does not settle.
  • Explain the difference between the textual situation of the Greek New Testament and that of the Hebrew Bible, and why the same method cannot simply be transferred from one to the other.

They went out, and fled from the tomb, for trembling and astonishment had come on them. They said nothing to anyone; for they were afraid. — Mark 16:8

World English Bible

The Greek:

καὶ ἀπὸ τοῦ , εἶχεν γὰρ αὐτὰς καὶ · καὶ , ἐφοβοῦντο γάρ.

kai exelthousai ephygon apo tou mnemeiou, eichen gar autas tromos kai ekstasis; kai oudeni ouden eipan, ephobounto gar.

Working translation: "And going out they fled from the tomb, for trembling and bewilderment had hold of them; and they said nothing to anyone, for they were afraid."

Two differences from the World English Bible are worth naming. The WEB has "trembling and astonishment had come on them," which is good English; the Greek verb is εἶχεν (eichen, imperfect of , "have, hold"), with τρόμος καὶ ἔκστασις (tromos kai ekstasis) as its subject — the trembling had them, rather than coming upon them. And I have kept "bewilderment" for ἔκστασις rather than "astonishment," because the noun covers a displacement of the mind that can be closer to being beside oneself than to admiration; the WEB's choice is defensible and the difference is small.

But look at the last two words, and notice something no English translation can reproduce: ἐφοβοῦντο γάρ. The clause, and with it Mark's Gospel in its earliest recoverable form, ends on the particle γάρ (gar, "for"). This lecture will come back to that ending. It is where the largest textual question in the New Testament begins.

1. Why an apparatus exists at all

We do not possess the original manuscript of any book of the Bible. What we possess are copies, copies of copies, translations made from copies, and quotations in the writings of people who owned copies. Every one of these was produced by hand, and hands make mistakes.

For the Greek New Testament this produces an unusual situation: an enormous number of witnesses. Somewhere in the region of 5,800 Greek manuscripts containing part or all of the New Testament are catalogued, plus manuscripts of the early versions in Latin, Syriac, Coptic and other languages, plus the quotations of the , which are numerous enough that large parts of the New Testament could be reconstructed from them alone. This wealth is a mixed blessing. More witnesses means more variants — the number of places where at least one manuscript differs from another runs into the hundreds of thousands — but it also means a much better chance that the original reading survives somewhere among them.

For the Hebrew Bible the situation is nearly the opposite, and this is the first thing to hold onto. The medieval Hebrew manuscripts are remarkably uniform: the Masoretic tradition was transmitted by professional scribes with an elaborate apparatus of counting and control, and the differences among the great medieval codices are small. The oldest complete one, the Leningrad Codex of 1008 or 1009, is the base text of BHS. Against this stands evidence of a quite different kind and much greater age: the biblical scrolls from Qumran and the other Judaean Desert sites, the Greek translation known as the (whose Hebrew source text was sometimes demonstrably different), the Samaritan , and the Syriac, Aramaic and Latin versions. So the Hebrew critic typically has one very stable Hebrew tradition and a small number of much older but indirect or fragmentary witnesses to something else — a completely different evidential problem from the New Testament critic's.

This asymmetry is why the two apparatuses look different, and why methods do not transfer. A New Testament critic can often compare dozens of independent Greek witnesses at a given point. An Old Testament critic frequently has to ask what Hebrew text a Greek translator was looking at in the third century BC, which is an inference, not an observation.

2. Reading a New Testament apparatus: the witnesses

An entry in NA28 names its witnesses by a system of sigla you should learn once and then never think about again.

Papyri. Cited by a Gothic capital P with a superscript number — printed in this lecture, for readability, as P45, P46, P66, P75. These are the oldest surviving copies, mostly fragmentary, mostly from Egypt, mostly second to fourth century. Their value is age, not completeness. P75, for instance, contains much of Luke and John and is an important early witness for both.

Majuscules (older books call them uncials). Written entirely in capital letters, on parchment, from roughly the fourth century onwards. The most important have both a letter and a number: Codex Sinaiticus is א (aleph) or 01, Codex Alexandrinus is A or 02, Codex Vaticanus is B or 03, Codex Ephraemi Rescriptus is C or 04, Codex Bezae Cantabrigiensis is D or 05. Sinaiticus and Vaticanus, both fourth-century, are the two great complete or near-complete early witnesses. Bezae is a bilingual Greek-Latin manuscript with a wild and distinctive text, especially in Acts, and is prized precisely because it is so different — an odd witness is often an informative one.

Minuscules. Written in a cursive hand from the ninth century onwards, cited by number alone (33, 565, 1739 and so on). Being late does not make a minuscule worthless: minuscule 1739 preserves an old text in Paul and is regularly cited, and the numbered families known as family 1 and family 13 preserve distinctive older readings. This is the first practical lesson in weighing rather than counting.

Lectionaries. Books of readings arranged for the church year, cited with a script letter l and a number. Their liturgical arrangement affects their text — a reading needs a beginning, so lectionaries add introductory phrases — and this must be allowed for.

Versions. Translations into Latin, Syriac, Coptic, Armenian, Georgian, Ethiopic and more. Cited by abbreviations such as latt for the whole Latin tradition, sy for Syriac, co for Coptic, with further letters for individual versions. Their evidence is powerful for date and geography, since a version made in the second century in Syria tells you what Greek text was circulating in that place at that time. Their evidence is limited for anything the target language cannot express: a Latin translation cannot show you whether its Greek exemplar had a definite article, since Latin has none.

Fathers. Named individually — Origen, Irenaeus, Eusebius, Jerome, Chrysostom. Their citations are datable and locatable, which is exactly what a manuscript usually is not. But a father quoting from memory, or paraphrasing in a sermon, is not the same evidence as a father commenting on the wording word by word, and a good apparatus distinguishes these.

The Majority text. A Gothic capital M, standing for the great body of later Byzantine manuscripts that read together. It is cited as a single witness precisely because those manuscripts are not independent of one another: a thousand copies descended from one exemplar are one witness with a thousand voices, not a thousand witnesses.

3. Reading a New Testament apparatus: the signs

NA28 marks the point of variation in the printed text itself, with small raised signs, and then repeats the sign in the apparatus below. The system is designed to save space and takes about an hour to learn:

  • A raised ° before a word means that word is omitted by the witnesses listed.
  • A raised ⸀ before a word means that word is replaced by another in the witnesses listed; the replacement is given in the apparatus.
  • ⸂ … ⸃ enclose a group of words replaced as a unit.
  • A raised ⸆ marks a point where the listed witnesses insert something.
  • txt in the apparatus introduces the witnesses supporting the reading actually printed above.
  • Square brackets in the printed text mean the editors were not confident: they have printed the word but regard it as uncertain. Double square brackets mean the editors regard the enclosed passage as a later addition that is nevertheless too important in the history of the church to leave off the page. Mark 16:9-20 and John 7:53-8:11 are both printed in double brackets, which is itself a judgement stated in punctuation.
  • In the Catholic Epistles, NA28 introduced a diamond to mark places where the editors could not decide between two readings and declined to pretend otherwise. This is a genuine change of practice and worth knowing about: the text of those letters in NA28 was produced by a different method from the rest, the Coherence-Based Genealogical Method developed at Münster by Gerd Mink, which uses computational comparison of the whole tradition rather than a prior classification into families.

BHS uses a different and older system. Small raised letters in the Hebrew text point to notes at the foot of the page written in Latin abbreviations, which is the main obstacle for most readers. The witnesses are given Gothic letters: the itself, the Septuagint, the Syriac Peshitta, the Targums, the Vulgate, the Samaritan Pentateuch, and the Qumran manuscripts (cited by their own sigla, such as 1QIsa-a). The Latin abbreviations you will meet constantly are mlt Mss ("many manuscripts"), l for legendum ("read" — that is, the editor proposes reading something else), prp for propositum ("it has been proposed"), dl for dele ("delete"), ins for insere ("insert"), frt for fortasse ("perhaps"), and cj for a conjecture.

Notes without witnesses

Here is the trap, and it is serious. In BHS, a large number of the notes are not reports of manuscript evidence at all. They are the editor's own proposals: prp, l, cj. A note saying that something should perhaps be read differently, with no witness cited, is one scholar's guess, printed in the same typeface as an actual variant supported by the Septuagint and a Qumran scroll. Lecture 6-3 told you to suspect an emendation three times before accepting it; this is where you will meet emendations, and the apparatus does not shout at you when it changes from reporting to proposing. Learn to check whether a witness is named. The successor edition, Biblia Hebraica Quinta, is being published fascicle by fascicle and improves matters considerably: its notes are in English, and it separates evidence from evaluation more clearly.

Alongside the apparatus, and quite distinct from it, BHS prints the Masorah: the marginal notes of the Masoretes themselves, in the outer margin (Masorah parva) and at the top and bottom of the page (Masorah magna). These are not text-critical notes in the modern sense but the scribes' own control apparatus — counts of how often a form occurs, warnings that a spelling is unique, and the ketiv/qere notation, in which the consonants written in the text (ketiv, "what is written") are to be read aloud as a different word given in the margin (qere, "what is read"). The most familiar case is the divine name, written with the consonants of one word and read as another.

4. Weighing, not counting

The single principle that organises everything else is easy to state and hard to apply: witnesses are weighed, not counted.

Three hundred manuscripts agreeing against three does not settle anything by itself, because the three hundred may all descend from a single ancestor. asks not "how many?" but "how many independent lines of transmission, from how early, from how many places?" The classic formulation is Fenton John Anthony Hort's, in the volume of introduction to the New Testament edited by him and Brooke Foss Westcott in 1881: knowledge of documents should precede final judgement upon readings. Know what kind of copy each manuscript is, and how they are related, before you start counting heads.

The criteria divide into two groups.

External evidence concerns the witnesses themselves: their date, their geographical spread, and their textual relationships. A reading attested in the second century in Egypt, in the third in Rome, and in an early Syriac version has three independent claims on you. A reading attested in five hundred fifteenth-century manuscripts from one region has one.

Internal evidence divides again. Transcriptional probability asks which reading a scribe was more likely to produce — that is, which variant best explains the origin of the others. Intrinsic probability asks which reading the author was more likely to have written, given his style and argument elsewhere.

From transcriptional probability come the two most quoted canons, both associated with Johann Jakob Griesbach in the late eighteenth century, though older:

Lectio difficilior potior — the harder reading is to be preferred. A scribe is more likely to smooth a difficulty than to create one, so a reading that is grammatically awkward, theologically uncomfortable or factually surprising has a claim to be original. Its known failure mode: some hard readings are hard because they are simply mistakes. A scribe who miscopies produces nonsense, and nonsense is very hard indeed. The must be limited to readings that are difficult but sensible, and the judgement of where that line falls is not mechanical.

Lectio brevior potior — the shorter reading is to be preferred, on the ground that scribes expanded, harmonised and explained more often than they cut. This is a point where readers will not agree, and here the disagreement has been settled largely in one direction within living memory. James R. Royse, in Scribal Habits in Early Greek New Testament Papyri (2008), examined the actual behaviour of the scribes of the major early papyri by comparing their corrections and their singular readings, and found that these scribes omitted more often than they added. That is the opposite of what the canon assumes. Royse's work built on earlier studies by Ernest Colwell and has been widely accepted; Peter Head and others have extended it. The consequence is not that the shorter reading is now always wrong, but that the canon has lost its status as a general rule and has to be argued for case by case. Where you meet an older commentary applying it confidently, register that the ground has shifted.

Reasoned eclecticism, the mainstream method behind NA28 and behind Bruce Metzger's A Textual Commentary on the Greek New Testament — the single most useful book for a student, since it reports the committee's reasoning case by case, including where the committee was divided — holds both kinds of evidence together and does not let either become mechanical.

This is a point where readers will not agree, and the alternatives are worth knowing by name.

J. Keith Elliott and, before him, George Kilpatrick, argued for thoroughgoing eclecticism: internal criteria — the author's style and usage above all — should decide every case, and the manuscripts serve only to supply the candidate readings. The strength of the position is that it refuses to be captured by a prior theory about which manuscripts are good. The weakness, as its critics press, is that an author's style is itself reconstructed from a text, so the method risks circularity.

Maurice A. Robinson and William G. Pierpont argued for Byzantine priority: that the great mass of Byzantine manuscripts preserves the original text, and that the Alexandrian witnesses represent an early, localised and defective branch. This position is a minority one and is often confused with the quite different King James Only movement, from which Robinson explicitly distances himself; it is a text-critical argument, made with text-critical evidence, about transmission history. Its central claim — that a text used continuously by the whole Greek-speaking church is unlikely to be a late corruption — is a real argument, and the standard reply is the one about counting: the Byzantine manuscripts' agreement is evidence of common ancestry, not of independent attestation.

Bart D. Ehrman, in The Orthodox Corruption of Scripture (1993), argued that a significant class of variants was produced by scribes adjusting the text in the direction of theological orthodoxy during the christological controversies of the second and third centuries, and in the popular Misquoting Jesus (2005) presented the state of the text to a general readership in terms that many readers took as undermining confidence in it altogether. The responses — from Daniel B. Wallace, Michael W. Holmes, Craig Blomberg and others — generally do not dispute Ehrman's catalogue of variants, which is largely the standard one; they dispute the inference. Their case is that the variants Ehrman treats as theologically motivated are a small proportion of the whole, that motivation is often unprovable, and that no article of any major creed depends on a disputed reading. Notice the shape of this: it is chiefly an argument about what the evidence implies, not about what the evidence is, and both sides work from the same apparatus. Metzger and Ehrman jointly produced the fourth edition of the standard handbook. That fact is worth more than a hundred pages of polemic in either direction.

5. Case one, small: Luke 15:16

Chapter 8 flagged this in passing. The prodigal son, hungry among the pigs, wants something. What?

The World English Bible reads:

He wanted to fill his belly with the husks that the pigs ate, but no one gave him any. — Luke 15:16 (WEB)

Modern critical editions read something else. The two candidates are:

τὴν αὐτοῦ (gemisai ten koilian autou, "to fill his belly") — an aorist infinitive of , "fill," with an object.

(chortasthenai, "to be fed full, to be satisfied") — an aorist passive infinitive of , a verb originally used of feeding animals and in Koine used generally of eating one's fill.

External evidence favours χορτασθῆναι: it is the reading of the oldest Alexandrian witnesses, including the early papyrus P75 and Codex Vaticanus, and it also has the support of Codex Bezae, which normally goes its own way — agreement between witnesses that usually disagree is worth a great deal. γεμίσαι τὴν κοιλίαν αὐτοῦ is the reading of the great mass of later manuscripts, and it is what the translators of the King James Version had in front of them; the World English Bible, which stands in that textual lineage, prints it.

Internal evidence points both ways, and this is why the case is worth doing. The argument for χορτασθῆναι from transcriptional probability is that "fill his belly" is the more vivid, more concrete expression, and scribes tend to make narrative more vivid. The argument against it is that Luke uses χορτασθῆναι just one chapter later, at 16:21, of Lazarus longing to be fed from the rich man's table — and harmonisation to a nearby parallel is one of the commonest things a scribe does. So the very fact that makes χορτασθῆναι feel Lukan is also a reason a scribe might have introduced it.

That is an honest tension, and it is not resolved by the internal criteria. What resolves it, for most editors, is the external evidence: the combination of P75, Vaticanus and Bezae is very strong, and the alternative has no early support. So the critical editions print χορτασθῆναι and note the other.

What follows for you as a reader? Almost nothing about the meaning of the parable. Both readings say the boy was starving. That is exactly the point worth taking away from a small variant: the overwhelming majority of textual variants are of this kind. They are real, they are worth getting right, and they change nothing that anyone preaches. A student who has done one small case properly is much less likely to be alarmed by a headline about hundreds of thousands of variants.

6. Case two, structural: where does the doxology of Romans belong?

Here is a case you can verify in your own Bible in two minutes, which makes it the best teaching example in the New Testament.

Romans ends, in most modern editions, with a doxology beginning "Now to him who is able to establish you according to my Good News" and running for three verses, numbered 16:25-27. But that doxology travels. In the manuscript tradition it appears after 16:23, after 14:23, after 15:33, in two of those places at once in some witnesses, and is absent altogether from others. P46, the oldest substantial manuscript of Paul's letters, places it after 15:33. The tradition that the heretic Marcion's edition of Romans ended at chapter 14 is part of the discussion, though what exactly Marcion's text contained is itself reconstructed from his opponents.

Now open the World English Bible. It follows the older textual arrangement and prints the doxology after 14:23, numbering it 14:24-26. The consequence is arithmetical and unmissable: the WEB's Romans 16 has no verses 25, 26 or 27. It ends at verse 24.

Go and look. If your English Bible has a Romans 16:25 and the WEB does not, you have just seen two editions embodying two different decisions about the same body of evidence, and you did not need a word of Greek to see it.

This case teaches something the other three do not. Variation is not only about words. It can be about arrangement — where a block of material sits, whether a verse is numbered, where a book ends. And the verse numbers themselves are not part of the text: they were imposed on the New Testament in the sixteenth century by the printer Robert Estienne, long after every manuscript in the apparatus was written. When two Bibles disagree about a verse number, no manuscript is being contradicted.

7. Case three, large: the ending of Mark

Return to this lecture's Scripture, and to the two words the Gospel ends on: ἐφοβοῦντο γάρ.

The evidence. Verses 9 to 20, the so-called Longer Ending, are absent from the two great fourth-century codices, Sinaiticus and Vaticanus — the oldest Greek witnesses that contain the end of Mark at all. Eusebius and Jerome, in the fourth and early fifth centuries, both state that the passage was missing from most of the Greek copies known to them, and Eusebius' discussion of how to reconcile the resurrection accounts proceeds on that basis. The Longer Ending is absent from the Sinaitic Syriac manuscript and from some Armenian and Georgian evidence. A number of manuscripts that do contain it mark it with asterisks or obeli — the scribal signs for "this is questioned."

Moreover there is not one competing ending but several. Some manuscripts have a Shorter Ending, a single sentence about the women reporting to Peter's companions. Some have the Shorter Ending followed by the Longer Ending. One manuscript inserts an additional passage inside the Longer Ending. A proliferation of endings is itself an argument: it is what you would expect if scribes independently felt that something was missing and supplied it, and not what you would expect if one original ending had simply dropped out of a few copies.

Internal evidence. The vocabulary and style of verses 9-20 diverge from Mark's elsewhere; a number of the words in the passage occur nowhere else in the Gospel, and the narrative joins badly — verse 9 reintroduces Mary Magdalene as though she had not been named in verse 1 of the same chapter, which is not how the same author writes twelve words later.

The judgement. Nearly all critics conclude that verses 9-20 are not by the author of Mark. NA28 prints them in double brackets. Metzger's Textual Commentary sets out the case fully.

Where Mark ends

This is a point where readers will not agree, and the disagreement is real, though not evenly balanced.

William R. Farmer, in The Last Twelve Verses of Mark (1974), argued that the Longer Ending has better claims than the consensus allows and that the internal stylistic arguments have been overstated. Advocates of Byzantine priority, Maurice Robinson among them, defend the Longer Ending on their general principles about the Byzantine tradition. And there is a broad position, held by many who accept that verses 9-20 are secondary, that the Gospel nevertheless cannot have ended at ἐφοβοῦντο γάρ, and that the true ending is lost — that the last leaf of an early codex was damaged. N. Clayton Croy argued this at length in The Mutilation of Mark's Gospel (2003), and there are respectable reasons for it: Mark 14:28 and 16:7 both promise a meeting in Galilee, and the Gospel as we have it never delivers it.

Against that stands a claim that used to be made confidently and has since been dismantled: that a Greek book cannot end with γάρ, because the particle is postpositive and needs something before it. P. W. van der Horst, in a 1972 article, assembled examples of Greek literary works and sections ending in exactly that way, and the linguistic objection has not recovered. Whether Mark would have ended a Gospel that way is a literary question, on which readers differ; whether he could is settled, and the answer is yes.

What the judgement does not settle. That verses 9-20 were not written by Mark is a claim about authorship and transmission. It is not, by itself, a claim about whether the events they narrate occurred, most of which are attested elsewhere in the New Testament. Nor does it settle what a church should read aloud: the Longer Ending has been part of the received text of the Latin, Greek and Syriac churches for many centuries, was affirmed as canonical Scripture at the for the Catholic Church, and continues to be printed in Catholic and Orthodox editions. Being able to distinguish a text-critical judgement from a canonical or liturgical one is part of what this lecture is for. A student who announces that "the last twelve verses of Mark are not in the Bible" has confused three different questions, and has done so in front of people whose churches read those verses on Sunday.

8. Case four, large: the woman taken in adultery

John 7:53-8:11, the story of the woman brought to Jesus by those who would stone her, is the other great case, and its evidence is unusually clean.

It is absent from P66 and P75, the two earliest substantial witnesses to John, and from Sinaiticus and Vaticanus. It is absent from the early Syriac and Coptic versions. No Greek Father comments on it before the twelfth century — Origen and John Chrysostom both work through John continuously and pass from 7:52 to 8:12 without a word, which for a commentator is loud silence. In the West it is a different story: Jerome knew it in Greek and Latin manuscripts and included it in the Vulgate, and Augustine and Ambrose both discuss it.

And it moves. In the manuscripts that contain it, it is not always in the same place. It stands after John 7:52 in most; in the group of manuscripts known as family 13 it stands in Luke, after Luke 21:38; one witness puts it after John 7:36; others put it at the very end of John. A passage that floats is a passage that entered the tradition after the tradition had a fixed shape.

Style. The vocabulary and narrative manner of the passage have long been observed to sit closer to Luke than to John.

The judgement is nearly unanimous: the passage was not part of the Gospel of John as it was written. NA28 double-brackets it.

And here the more interesting scholarship begins. That the passage is a later insertion has been the consensus for well over a century; what the passage is has become an active field. Chris Keith, in The Pericope Adulterae, the Gospel of John, and the Literacy of Jesus (2009), argued that the story was inserted deliberately and for a reason connected with the detail of Jesus writing on the ground. Jennifer Knust and Tommy Wasserman, in To Cast the First Stone: The Transmission of a Gospel Story (2019), traced the passage's whole history and pressed the case that the neat story of a late intrusion is more complicated than the standard account allows.

None of this makes the passage Johannine. It does mean that "a later addition" is a beginning of an account rather than the end of one, and that a very old story is not the same thing as a false one. As with Mark, the churches that read this passage liturgically are not making a text-critical error; they are answering a different question. What a student of the languages should be able to do is say precisely which question is being answered, and by what evidence.

9. The Hebrew side: two cases

Deuteronomy 32:8. The Masoretic text reads, at the end of the verse, (bene yisra'el, "the sons of Israel") — the WEB has "according to the number of the children of Israel." A manuscript from Qumran reads instead "sons of God," and the Septuagint has a corresponding phrase referring to God's angels or sons. The variant is not trivial: on the Qumran and Greek reading, the nations are apportioned according to the number of heavenly beings, which is a different picture. Here is the characteristic Hebrew-Bible situation in miniature: one stable Masoretic reading, and older but indirect evidence pointing elsewhere. Modern translations differ, and both the Revised Standard Version's successors and the English Standard Version have moved to the Qumran reading, while others retain the Masoretic. You now know enough to see why an English footnote at this verse exists.

1 Samuel 13:1. The Hebrew of this verse, as the Masoretes transmit it, is a formula that gives Saul's age at accession and the length of his reign — but the number for his age is missing and the number for his reign is unarithmetical. I give the consonants without vowel points deliberately, since the difficulty is in what is there rather than in how it is pointed: בן־שנה במלכו, "a son of a year was Saul when he became king." The World English Bible renders the verse "Saul reigned a year; and when he had reigned two years over Israel," which is one attempt at making the sentence do something. Some Septuagint manuscripts omit the verse. Some modern translations leave a blank or a footnote; some supply a number from Acts 13:21.

The lesson is the one from 6-3, in a new setting. The verse is not corrupt because it is theologically inconvenient; it is corrupt because it is arithmetically and grammatically broken, in a formula whose shape is known from elsewhere in the same books. That is the narrow class of case where the text really is damaged. Notice also what the Masoretes did: they copied what they had. A tradition that transmits a sentence it can see does not work is a tradition telling you it is copying rather than improving.

10. Rules of use

Six practical rules, in the order you will need them.

  1. Read the introduction to the edition you own. Every apparatus is a system, and the systems differ. NA28, BHS, BHQ and the UBS Greek New Testament all use different conventions and select their evidence on different principles. Ten pages of introduction will save you from most of the mistakes in this lecture.
  2. Check whether a witness is named. Particularly in BHS. A note with no witness is a proposal, not a variant.
  3. Ask what the variant would change. Most change nothing. If a variant does not alter any translation you can find in print, it is very unlikely to be worth a minute of a congregation's time.
  4. Never cite "the oldest and best manuscripts" as a phrase. Either you know which ones and why, in which case say so, or you do not, in which case you are borrowing an authority you have not earned.
  5. Distinguish the three questions. Is this reading original? Is it true? Should it be read in church? They have different evidence and different answers, and the confusion of them is the single most common failure in popular discussion of Mark 16 and John 8.
  6. When you build an argument on a word, check that the word is textually secure. This costs thirty seconds with an apparatus and occasionally saves an entire sermon.
Anatomy of a Textual Note What the block at the foot of the page is actually telling you In the printed text: καὶ ἐπεθύμει ⸀χορτασθῆναι ἐκ τῶν κερατίων … kai epethymei chortasthenai ek tōn keratiōn — "and he was longing to be filled from the pods" (Luke 15:16) ⸀ = this word is replaced in the witnesses listed At the foot of the page: ⸀ γεμισαι την κοιλιαν αυτου απο A Θ Ψ [Majority] | txt P75 ℵ B D L gemisai tēn koilian autou apo — "to fill his belly from" the competing reading · its witnesses || "txt" = witnesses for what is printed above Classes of witness Papyri (P45, P66, P75) — oldest, fragmentary Majuscules (ℵ 01, A 02, B 03, D 05) Minuscules (33, 565, 1739; fam. 1, fam. 13) Lectionaries — liturgically reshaped Versions (latt, sy, co) — date + geography Fathers — datable, locatable, quoted loosely Majority text = one ancestor, many copies Criteria — and where each fails External: age · spread · independence fails when an old copy is simply a bad copy Harder reading preferred fails when the hard reading is a mistake Shorter reading preferred Royse (2008): early scribes omitted more than they added — canon now argued case by case

The upper half of the diagram takes the one small case from section 5 and shows the mechanics: a raised sign in the printed line, the same sign repeated below, the competing reading, its witnesses, and then txt with the witnesses for what was actually printed. Every Greek form is repeated in transliteration underneath, so that the diagram is still readable where the Greek font is missing. The lower half sets the classes of witness beside the criteria — and note that the criteria panel names the failure mode of each rule rather than the rule alone. That is deliberate. A criterion you cannot break is a criterion you do not understand.

6. Further Reading

  • Bruce M. Metzger, A Textual Commentary on the Greek New Testament, 2nd edn (Stuttgart, 1994). The companion volume to the UBS edition, reporting the editorial committee's reasoning variant by variant, including the cases where the committee split. The most useful single book for a student learning to read an apparatus.
  • Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament: Its Transmission, Corruption, and Restoration, 4th edn (Oxford, 2005). The standard textbook, in an edition jointly produced by two scholars who disagree publicly about what the evidence implies.
  • Emanuel Tov, Textual Criticism of the Hebrew Bible, 3rd edn (Minneapolis, 2012). The indispensable book for the Old Testament, with a full account of the Qumran evidence and of why the Hebrew problem is not the Greek problem.
  • Ernst Würthwein, The Text of the Old Testament, revised by Alexander Achilles Fischer, 3rd English edn (Grand Rapids, 2014). Shorter than Tov and stronger on the practical business of reading BHS itself, with plates.
  • James R. Royse, Scribal Habits in Early Greek New Testament Papyri (Leiden, 2008). Technical, long, and the reason the shorter-reading canon can no longer be used as a general rule.
  • Jennifer Knust and Tommy Wasserman, To Cast the First Stone: The Transmission of a Gospel Story (Princeton, 2019). A full history of John 7:53-8:11 and a good demonstration that "a later addition" is where the questions begin.
  • David C. Parker, An Introduction to the New Testament Manuscripts and their Texts (Cambridge, 2008). Approaches the tradition as a history of living documents rather than a hunt for one original, which is a genuinely different framing and worth meeting.
  • Kurt and Barbara Aland, The Text of the New Testament, 2nd edn (Grand Rapids, 1989). Written from inside the Münster institute that produces Nestle-Aland; strongly opinionated, and the clearest statement of the position NA28 embodies.

핵심 요약

  • The New Testament and Hebrew Bible present opposite textual problems: roughly 5,800 Greek manuscripts give the New Testament critic many independent, comparable witnesses, while the Hebrew Bible's medieval manuscripts are remarkably uniform and the real evidence for variation — Qumran, the Septuagint's underlying Hebrew, the Samaritan Pentateuch — is older but indirect, which is why methods do not transfer between the two disciplines.
  • Witnesses are weighed, not counted (Hort's principle): three hundred late manuscripts agreeing against three early ones proves nothing if the three hundred all descend from one exemplar, which is exactly why the Majority text is cited as a single witness rather than as a numerical majority.
  • Royse's 2008 study of scribal habits in the early papyri overturned two centuries of assuming lectio brevior potior ("the shorter reading is preferred"): early scribes in fact omitted more often than they added, so the canon can no longer be applied as a general rule and must be argued case by case.
  • Mark 16:9-20 is absent from Sinaiticus and Vaticanus, unknown to Eusebius and Jerome in most copies they knew, exists in competing Shorter and Longer forms, and reads badly as a continuation (verse 9 reintroduces Mary Magdalene as though chapter 16 had not just named her) — nearly all critics judge it not authentically Markan, though the claim that a Greek book cannot end on the postpositive γάρ has been refuted by van der Horst's citation of parallels, so Mark could grammatically have ended at 16:8.
  • The lecture insists on distinguishing three separate questions that popular writing on Mark 16 and John 7:53-8:11 routinely collapses into one: is a reading original, is it true, and should it be read in church — a passage can be judged a later addition and still be centuries old, liturgically established, and not thereby false.

If you remember nothing else from this lecture: A critical edition's apparatus is not decoration for specialists; it is the editors showing their work, because no single surviving manuscript matches exactly what any printed Greek or Hebrew Bible prints. The governing principle is that witnesses are weighed by independence, age and geographic spread rather than simply counted, which is why a huge, late, single-ancestor tradition like the Majority text counts as one witness, not a thousand. The clearest demonstration is the ending of Mark: verses 9-20 are missing from the earliest and best manuscripts, contradict Mark's own narrative logic by reintroducing a character just named, and exist in multiple competing forms — evidence that points to a later addition — yet churches have read those verses for centuries, and answering whether a reading is original, whether it is true, and whether it belongs in worship are three different questions with three different kinds of answer, not one.

이 과목 요약 전체 보기 →

스스로 확인해 보세요

읽었는지가 아니라 설명할 수 있는지를 묻는 것입니다. 체크는 이 기기에만 저장됩니다.

묵상 및 토론

  • A reading exercise. Open your Greek New Testament at Mark 16 and read verses 1-8 aloud in Greek, then stop where the printed text stops before the double brackets. Then read verses 9-20 aloud. Without consulting any commentary, list three things about the Greek of the second passage that struck you as different. Then check your list against Metzger's *Textual Commentary*.
  • A reading exercise. Open your Hebrew Bible at any page of Genesis and find three notes in the apparatus. For each, decide whether it reports a witness or proposes a reading, and say which Latin abbreviation told you. If you cannot tell for one of them, that is the note to bring to your teacher.
  • Section 5 gives an internal argument on each side at Luke 15:16 and then lets external evidence break the tie. Reconstruct both internal arguments in your own words. If you had only the internal evidence, which way would you go, and what would your confidence be?
  • Verify the Romans doxology case for yourself in two English Bibles from different textual traditions. Write out exactly what each prints as Romans 16:24 and 16:25. What would you say to someone who told you that one of the two Bibles has "removed verses"?
  • The lecture insists on separating three questions about Mark 16:9-20: is it original, is it true, should it be read in church. Take a passage from your own tradition's lectionary and ask the three questions about it. Do they come apart in the same way?
  • Royse's finding that early scribes omitted more than they added overturned a canon that had stood for two centuries. What does it tell you about the status of the *other* canons that this one could be overturned by counting what scribes actually did? Is that reassuring or unsettling?

로그인하시면 여기에 답을 적고 보관하실 수 있습니다

다음 단계

원어 뜻풀이: Strong’s Concordance(1890/1894, 퍼블릭 도메인) · 어형 자료: MorphGNT/SBLGNT(CC BY-SA), Open Scriptures morphhb/WLC(CC BY 4.0)

여기까지가 강의입니다. 아래는 읽은 사람들이 남긴 글입니다.