Ancient Citation and Authorship Conventions: scholarly quotations

Jerome.  “Letter 57, To Pammachius, on the Best Method of Translating.”  Nicene and Post-Nicene Fathers, series 2, vol. 6.  Translated by W. H. Fremantle.  1893.  Jerome's defence, written in AD 395, of his own sense-for-sense translations against critics who wanted word-for-word renderings.  To make his case he lists New Testament quotations that differ in wording from both the Hebrew and the Greek, including Matthew 27:9, Mark 1:2 and Mark 2:26.

1. 1a.

   Section 7:  “This passage is not found in Jeremiah at all but in Zechariah, in quite different words and an altogether different order.”  “they may say that he has mistaken the author's name putting down Jeremiah when it should be Zechariah. Far be it from us to speak thus of a follower of Christ, who made it his care to formulate dogmas rather than to hunt for words and syllables.” ↩

2. 1b.

   Section 9:  “This quotation is made up from two prophets, Malachi that is to say and Isaiah.”  “it is not written in Isaiah at all, but in Malachi the last of the twelve prophets? Let ignorant presumption solve this nice question if it can, and I will ask pardon for being in the wrong.” ↩

3. 1c.

   Section 9:  “not to convict the evangelists of falsification — a charge worthy only of impious men like Celsus, Porphyry, and Julian — but to bring home to my critics their own want of knowledge”. ↩

Augustine of Hippo.  “The Harmony of the Gospels (De consensu evangelistarum).”  Nicene and Post-Nicene Fathers, series 1, vol. 6.  Translated by S. D. F. Salmond.  1888.  Augustine's defence of the agreement of the four Gospels, written about AD 400.  Book 3, chapter 7 is the fullest ancient discussion of Matthew 27:9, and it weighs the manuscript evidence, a slip of memory and a composite reading with Jeremiah 32.  The translation carries a bracketed note by its American editor inside section 30, which is not Augustine's.

1. 2a.

   Book 3, chapter 7, section 29:  “a majority of codices contain the name of Jeremiah, and that those critics who have studied the Gospel with more than usual care in the Greek copies, report that they have found it stand so in the more ancient Greek exemplars”.  Augustine first reports that "this ascription of the passage to Jeremiah is not contained in all the codices of the Gospels", then rejects omitting it. ↩

2. 2b.

   Book 3, chapter 7, section 30:  “it may have been the case, that when Matthew was engaged in composing his Gospel, the word Jeremiah occurred to his mind, in accordance with a familiar experience, instead of Zechariah”.  “all the holy prophets, speaking in one spirit, continued in perfect unison with each other in their utterances”. ↩

Gundry, Robert H.  “The Use of the Old Testament in St. Matthew's Gospel, with Special Reference to the Messianic Hope.”  Supplements to Novum Testamentum 18.  Leiden:  E. J. Brill.  1967.  A doctoral study comparing every Old Testament quotation and allusion in Matthew with the Hebrew, Greek, Aramaic and Syriac texts.  Its treatment of Matthew 27:9-10 includes the main published argument for a Zechariah-Jeremiah composite and a list of ten other explanations.

1. 3a.

   Page 125:  “The naming of one author in a composite allusion is not unknown elsewhere. For example, the allusive quotation in II Chron 36:21 is verbally drawn from Lev 26:34 f., yet ascribed to “Jer” (25:12; 29:10), from which the number of years, “seventy,” is drawn.” ↩

2. 3b.

   Page 125:  “Mt, then, sees two separate prophecies, one typical and one explicit, fulfilled in one event, and makes the ascription to Jer because the manifestness of the quotation from Zech and the lack of verbal resemblance to Jer would cause the Jer-side of the prophecies to be lost.” ↩

Metzger, Bruce M.  A Textual Commentary on the Greek New Testament.  London and New York:  United Bible Societies.  1971.  The committee's explanation of its choices for the United Bible Societies' Greek New Testament (third edition), written by the most influential New Testament textual critic of the twentieth century.  This is the first edition;  a second followed in 1994 with different page numbers.

1. 4a.

   Page 33:  “more than one scribe might have been prompted to insert the name of the best known prophet”.  Pages 66-67:  “The reading Iepeutovu is firmly established”.  ““Tsaiah’’—perhaps because, as the most prominent of the”.  Pages 66-67 are on Matthew 27:9 (OCR garbles the Greek name and "Isaiah");  page 33 on Matthew 13:35, where the commentary also notes that Jerome reports Porphyry citing "Isaiah" there "as showing the ignorance of Matthew." ↩

Wallace, Daniel B.  “Mark 2:26 and the Problem of Abiathar.”  Bible.org.  Mar 22, 2006.  A survey of five kinds of solution to the Abiathar problem by an evangelical New Testament textual critic at Dallas Theological Seminary, written to show how presuppositions shape exegesis.  He states the case for error fairly and gives his own preference in an appendix.  Mirror:  Archive.org

1. 5a.

   Section Conclusion:  “Option 1 (text-critical) is clearly out.”  “It must be admitted that views 2-5 all have a certain plausibility.”  Section Appendix:  “My own preference is for view 5: I believe that we have interpreted the text incorrectly.” ↩

2. 5b.

   Section 5. Hermeneutical:  “I do not yet know of any texts in which the construction ἐπί + genitive of personal name + anarthrous title indicates the general “in the time of,” though one or two seem to come close (see appendix). To be sure, I have not searched very diligently for this construction.” ↩

Ehrman, Bart D.  Misquoting Jesus:  The Story Behind Who Changed the Bible and Why.  San Francisco:  HarperSanFrancisco.  2005.  A popular account of New Testament textual criticism by a leading scholar of early Christianity, opening with the story of how he moved away from belief in an inerrant Bible.

1. 6a.

   Page 9:  “Maybe Mark just made a mistake.”  The comment of Cullen Story, his teacher at Princeton Theological Seminary, on Ehrman's student paper arguing that Mark 2:26 refers to "the part of the scriptural text that has Abiathar as one of the main characters." ↩

The Dead Sea Scrolls Study Edition.  Leiden:  Brill.  Edited by Florentino García Martínez and Eibert J. C. Tigchelaar.  1999.  The standard two-volume edition of the non-biblical Dead Sea Scrolls, with Hebrew and Aramaic text and facing English translation, by two leading scrolls scholars.  The translators write the divine name as four asterisks.

1. 7a.

   Page 17, 1QpHab column VII, lines 1-5:  “And God told Habakkuk to write what was going to happen to 2 the last generation, but he did not let him know the consummation of the era.”  “Its interpretation concerns the Teacher of Righteousness, to whom God has made known 5 all the mysteries of the words of his servants, the prophets.”  The numerals are the scroll's line numbers. ↩

Kaiser Jr., Walter C.  The Uses of the Old Testament in the New.  Chicago:  Moody Press.  1985.  An evangelical Old Testament scholar's defence of the view that New Testament writers used Old Testament texts in line with their original authors' meaning, worked through hard cases including Hosea 11:1.

1. 8a.

   Page 49:  “The whole community was pointedly designated by a singular noun, thereby allowing for the many to be represented in the One and the One to stand for the many”. ↩

Enns, Peter.  Inspiration and Incarnation:  Evangelicals and the Problem of the Old Testament.  Grand Rapids:  Baker Academic.  2005.  An evangelical Old Testament scholar's argument that the New Testament writers interpreted the Old Testament by the methods of their Second Temple Jewish setting, reading it from the resurrection backward.  Chapter 4 treats Matthew's use of Hosea 11:1.  The book cost him his post at Westminster Theological Seminary, which makes it the clearest statement of the view from within the evangelical world.

1. 9a.

   Page 133:  “Hosea 11 is not looking to the future at all but simply alluding to the past”.  “What motivates Matthew to handle Hosea’s words as he does is not a desire to do grammatical-historical exegesis.”  Pages 133-134:  “in writing to a Jewish audience, such a handling of the Old Testament would not have seemed strange but very familiar”. ↩

2. 9b.

   Page 155:  “I do not hesitate to point out that the christotelic and ecclesiotelic dimensions do not explain absolutely everything the New Testament writers do with the Old Testament.” ↩

Philo of Alexandria.  “On Mating with the Preliminary Studies (De congressu eruditionis gratia).”  The Works of Philo Judaeus.  London:  H. G. Bohn.  Translated by C. D. Yonge.  1854.  An allegorical commentary on Genesis 16 by Philo, a Jewish philosopher of Alexandria who lived from about 20 BC to AD 50, in which Sarah stands for virtue and Hagar for the general schooling that prepares for it.

1. 10a.

   Section 9-11:  “we are not as yet capable of becoming the fathers of offspring of virtue, unless we first of all have a connection with her handmaiden; and the handmaiden of wisdom is the encyclical knowledge”.  “of which Hagar, the handmaid of Sarah, is an emblem”. ↩

Galen.  De libris propriis (On My Own Books), in Claudii Galeni opera omnia, vol. 19.  Leipzig:  Cnobloch.  Edited by Karl Gottlob Kühn.  1830.  The standard nineteenth-century edition of Galen, with Greek text and facing Latin translation.  On My Own Books lists the physician's genuine works because books under his name were being sold that he had not written.

1. 11a.

   Page 9:  “non est ista dictio Galeni et falso hic liber inscriptus est”.  Kühn's facing Latin, with long s and ligatures normalised:  "non est ista dictio Galeni et falso hic liber inscriptus est" ("this is not Galen's diction and this book is falsely titled"). ↩

Eusebius of Caesarea.  Historia ecclesiastica, Greek text in the First1KGreek TEI.  2016.  A digital transcription of the Greek text of Eusebius's Church History, used here for Serapion's own word for "falsely ascribed."

1. 12a.

   Book 6, chapter 12, section 3:  “τὰ δὲ ὀνόματι αὐτῶν ψευδεπίγραφα ἔμπειροι παραιτούμεθα”.  English from McGiffert's translation (NPNF, 1890). ↩

Eusebius of Caesarea.  “Church History, Book 6.”  Nicene and Post-Nicene Fathers, series 2, vol. 1.  Translated by Arthur Cushman McGiffert.  1890.  Book 6 of Eusebius's history preserves passages of Serapion of Antioch, Clement of Alexandria and Origen on disputed and falsely attributed books, in their own words.

1. 13a.

   Book 6, chapter 14, section 2:  “He says that the Epistle to the Hebrews is the work of Paul, and that it was written to the Hebrews in the Hebrew language; but that Luke translated it carefully and published it for the Greeks, and hence the same style of expression is found in this epistle and in the Acts.”  Eusebius summarising Clement of Alexandria's lost Hypotyposes. ↩

2. 13b.

   Book 6, chapter 25, section 11-14:  “the verbal style of the epistle entitled 'To the Hebrews,' is not rude like the language of the apostle”.  “the thoughts are those of the apostle, but the diction and phraseology are those of some one who remembered the apostolic teachings, and wrote down at his leisure what had been said by his teacher”.  “But who wrote the epistle, in truth, God knows.”  Origen's lost Homilies on Hebrews, quoted by Eusebius. ↩

Eusebius of Caesarea.  “Church History, Book 7.”  Nicene and Post-Nicene Fathers, series 2, vol. 1.  Translated by Arthur Cushman McGiffert.  1890.  Book 7 preserves the most detailed ancient comparison of style for authorship:  Dionysius, bishop of Alexandria in the mid-third century, comparing Revelation with John's Gospel and first letter.

1. 14a.

   Book 7, chapter 25, section 7-26:  “Therefore that he was called John, and that this book is the work of one John, I do not deny.”  “For I judge from the character of both, and the forms of expression, and the entire execution of the book, that it is not his.”  “without even a syllable in common with them”.  “I perceive, however, that his dialect and language are not accurate Greek, but that he uses barbarous idioms, and, in some places, solecisms.” ↩

Jerome.  “Lives of Illustrious Men (De viris illustribus).”  Nicene and Post-Nicene Fathers, series 2, vol. 3.  Translated by Ernest Cushing Richardson.  1892.  Jerome's catalogue of Christian writers, written in AD 392, which records the doubts about several New Testament books and the reasons given for them.

1. 15a.

   Chapter 1:  “He wrote two epistles which are called Catholic, the second of which, on account of its difference from the first in style, is considered by many not to be by him.” ↩

Jerome.  “Epistulae LXXI-CXX.”  Corpus Scriptorum Ecclesiasticorum Latinorum 55.  Vienna:  Tempsky.  Edited by Isidor Hilberg.  1912.  The standard critical edition of Jerome's letters 71 to 120.  Letter 120 answers twelve questions from a woman named Hedibia;  the eleventh concerns 2 Corinthians 2:12-13.  English renderings here are this article's own.

1. 16a.

   Pages 507-508:  “habebat ergo Titum interpretem sicut 5 et beatus Petrus Marcum”.  “denique et duae epistulae, quae feruntur”.  The full sentence on p. 508:  "denique et duae epistulae, quae feruntur Petri, stilo inter se et caractere discrepant structuraque uerborum;  ex quo intellegimus pro necessitate rerum diuersis eum usum interpretibus."  The "5" inside the first quotation is the edition's line number. ↩

Tertullian.  “On Baptism.”  Ante-Nicene Fathers, vol. 3.  Translated by S. Thelwall.  1885.  The older English translation of Tertullian's On Baptism, made from an edition without the word "Acta" in chapter 17, so it speaks of "writings which wrongly go under Paul's name" where the oldest manuscript names the Acts of Paul.

1. 17a.

   Chapter 17:  “the presbyter who composed that writing, as if he were augmenting Paul's fame from his own store, after being convicted, and confessing that he had done it from love of Paul”. ↩

Anonymous.  “The Muratorian Fragment.”  Early Christian Writings.  Translated by Bruce M. Metzger.  n.d.  A Latin list of New Testament books, surviving in an eighth-century copy;  most scholars date the original to the late second century, though a minority argue for the fourth.  Metzger's translation numbers the lines of the Latin.

1. 18a.

   Lines 64-67:  “forged in Paul's (65) name to [further] the heresy of Marcion, and several others (66) which cannot be received into the catholic Church (67)— for it is not fitting that gall be mixed with honey.” ↩

Iamblichus.  Life of Pythagoras.  London:  J. M. Watkins.  Translated by Thomas Taylor.  1818.  An English translation of Iamblichus's On the Pythagorean Life, written about AD 300 by a Neoplatonist philosopher who admired the Pythagoreans of the sixth to fourth centuries BC.

1. 19a.

   Chapter 31, pages 103-104:  “This also is a beautiful circumstance, that”.  “they referred every thing to Pythagoras, and called it”.  Preface:  “obscured by false and spurious writings”. ↩

Salvian of Marseilles.  “Epistula IX, in Salviani presbyteri Massiliensis opera omnia.”  Corpus Scriptorum Ecclesiasticorum Latinorum 8.  Vienna:  Gerold.  Edited by Franz Pauly.  1883.  The critical Latin edition of Salvian, a fifth-century priest of Marseilles.  His ninth letter answers Bishop Salonius's question why recent books "to the church" carried the name of Timothy, and it survives in a single manuscript.  English renderings here are this article's own.

1. 20a.

   Page 217:  “cur libellis nuper”.  “Timothei nomen inscriptum sit”.  “carent enim apocryfa suspicione, qui agnoscuntur”.  Page 221:  “legunt non tam considerent quid legant quam cuius legant”.  "Why the name of Timothy was written on books recently made by a man of this time to the church";  "for those are free of the suspicion of the apocryphal which are recognised not to have been the apostle Timothy's";  readers "consider not so much what they read as whose it is." ↩

Ehrman, Bart D.  Forged:  Writing in the Name of God, Why the Bible's Authors Are Not Who We Think They Are.  New York:  HarperOne.  2011.  A popular book arguing that several New Testament letters, including 2 Thessalonians, the Pastorals and 2 Peter, are forgeries in the ancient sense, and that ancient readers condemned forgery.  Ehrman distinguishes anonymous books later misattributed from books whose authors falsely claimed a name.  His technical version is Forgery and Counter-Forgery (Oxford, 2013).

1. 21a.

   Page 271, endnote 27:  “the Acts of Paul is not a book that claims to be written by Paul; it claims to be a true account about Paul. The author was punished not for lying about his identity, but for fabricating a fictitious account and trying to pass it off as a historical record.” ↩

2. 21b.

   Page 137:  “Even the upper classes would sometimes instruct a secretary to spin off a quick stereotyped letter for them to someone, as is evidenced on several occasions by Cicero.”  “But drafting a brief stereotyped letter is completely different from composing a long, detailed, finely argued, carefully reasoned, and nuanced letter like 1 Peter or Ephesians.” ↩

3. 21c.

   Page 36:  “in virtually every instance in which an ancient author mentions forgery, he condemns it”.  Page 39:  “That forgery was not a transparent fiction is evidenced as well by the negative things people say about it in the ancient sources”. ↩

4. 21d.

   Page 24:  “In writings of this sort, which are wrongly attributed to a well-known person, the author is not trying to deceive anyone.”  “He or she remained anonymous. It is only later readers who claimed that the author was someone else.”  “My definition of a forgery, then, is a writing that claims to be written by someone (a known figure) who did not in fact write it.” ↩

5. 21e.

   Page 138:  “I don’t know of a single piece of evidence or a single analogy to suggest that Peter or Paul used a secretary who significantly—or insignificantly, for that matter—added to the contents of the letter.”  “Since secretaries did not produce the contents of letters (at least letter-essays of this sort), a secretary could not be responsible for the difference.” ↩

Bauckham, Richard J.  “Jude, 2 Peter.”  Word Biblical Themes.  Dallas:  Word Publishing.  1990.  A short thematic study of Jude and 2 Peter by a British New Testament scholar known for defending eyewitness testimony behind the Gospels.  He argues that 2 Peter is a "testament" written after Peter's death and that its pseudonymity was a convention its readers recognised.  It is not his longer 1983 commentary in the Word Biblical Commentary.

1. 22a.

   Page 44:  “In other words, Petrine authorship is a fiction, but it is one which the author does not feel obliged to maintain throughout his work. In that case, it must be a deliberately transparent fiction, a literary convention which the author expected his readers to recognize as such, just as modern readers are not likely to mistake a historical novel for factual history.” ↩

Guthrie, Donald.  “New Testament Pseudonymity?  A review.”  Themelios 13(2).  1988.  A review, by a conservative New Testament scholar who defended the traditional authorship of the disputed letters, of David Meade's Pseudonymity and Canon (1986), which argued that pseudonymous attribution claimed authoritative tradition rather than literary origin.

1. 23a.

   Section Opening paragraph; paragraph beginning The real Achilles heel:  “literary attribution is primarily an assertion of authoritative tradition, and not of literary origins”.  “The real Achilles heel of Meade’s case is when he is obliged to admit that there was an element of deception in the process.” ↩

Wilder, Terry L.  Pseudonymity, the New Testament, and Deception:  An Inquiry into Intention and Reception.  Lanham:  University Press of America.  2004.  A conservative study of how ancient Jewish, pagan and Christian readers received pseudonymous writings, concluding that apostolic pseudepigrapha, if any were in the New Testament, would have deceived their readers.

1. 24a.

   Page 253:  “in none of the references above are pseudonymous apostolic letters ever mentioned in connection with this concept. Further, a lie is nonetheless still a lie.” ↩

2. 24b.

   Page 217:  “the first-century church probably also would have rejected pseudonymous letters written in their names”. ↩

Cicero, Marcus Tullius.  Letters to Atticus (Epistulae ad Atticum), Latin text of the Perseus Digital Library.  2026.  Cicero's private letters to his friend Atticus, 68 to 44 BC, the largest surviving body of ancient correspondence about how letters were actually written and sent.  English quotations are from E. S. Shuckburgh's translation (London, 1899-1900), cited by the standard letter numbers.

1. 25a.

   Book 3, chapter 15, section 8:  “si qui erunt quibus putes opus esse meo nomine litteras dari, velim conscribas curesque dandas”.  Book 11, chapter 3, section 3:  “ad quos aliquid scribendum a me existimes ipse conficias”.  Book 11, chapter 5, section 3:  “tu velim et Basilo et quibus praeterea videbitur, etiam Servilio conscribas, ut tibi videbitur, meo nomine”. ↩

Richards, E. Randolph.  “Silvanus Was Not Peter's Secretary:  Theological Bias in Interpreting διὰ Σιλουανοῦ ... ἔγραψα in 1 Peter 5:12.”  Journal of the Evangelical Theological Society 43(3):417-432.  Sep 2000.  A study of every known use of the formula "I write through someone" in Ignatius, Polycarp and the Greek papyri, by an evangelical specialist in ancient letter-writing who had earlier written the main monograph on Paul's secretaries.  He argues the formula names the carrier, against a reading his own side has used to defend Peter's authorship.

1. 26a.

   Page 418:  “It will be argued here that this formula identi˜ed only”.  “the letter-carrier and was not used to identify the secretary.” ↩

2. 26b.

   Page 432:  “Silvanus certainly could have”.  “We do not need 1 Pet 5:12 to argue that there was secre-”.  “1 Pet 5:12 does not address the role of the secretary.” ↩

3. 26c.

   Page 427, note 45:  ““no one has yet proved that gravfw diav tinoÍ can”.  Kümmel, Introduction to the New Testament (1975) p. 424, as quoted by Richards;  the OCR renders the Greek font as "gravfw diav tinoÍ". ↩

Eder, Maciej.  “Does Size Matter?  Authorship Attribution, Small Samples, Big Problem.”  Digital Humanities 2010, King's College London.  2010.  A conference report testing how much text standard word-frequency attribution needs, by cutting known-author texts in English, Polish, German, Hungarian, French, Latin and ancient Greek into samples from 600 to 20,000 words and measuring how often each sample was attributed correctly.  Its thresholds are the clearest published guide to minimum sample size.

1. 27a.

   Section Experiment I:  “samples shorter than 5000 words provide a poor "guessing", because they can be immensely affected by random noise. Below the size of 3000 words, the obtained results are simply disastrous”.  “the critical point of attributive success could be found between 5000 and 10000 words per sample (and there was no significant difference between inflected and non-inflected languages)”.  “its minimal effective sample size was of some 2500 words”.  Section Conclusions:  “using 2500-word samples will hardly provide a reliable result, to say nothing of shorter texts”.  The first figures are for 63 English novels;  the 2,500-word figure is for Latin prose. ↩

Stover, Justin A., and Mike Kestemont.  “The Authorship of the Historia Augusta:  Two New Computational Studies.”  Bulletin of the Institute of Classical Studies 59(2).  2016.  A computational study of a Latin collection of imperial biographies that claims six authors, using a validated verification method on 1,000-word chunks, with a review of two earlier studies that reached opposite answers.  Locators are to the authors' accepted manuscript.

1. 28a.

   Pages 1-2:  “the use of sentence length as a criterion of authorship, which is no longer considered an effective stylometric feature even for modern texts, and should definitely not be used for ancient texts, where the punctuation is due to the modern editor”. ↩

Morton, A. Q., and James McLeman.  Paul, the Man and the Myth:  A Study in the Authorship of Greek Prose.  London:  Hodder and Stoughton.  1966.  The first computer study of the authorship of Paul's letters, by a Scottish minister and statistician, based on sentence length and the frequency of common Greek words.  It concluded that only four letters were Paul's.  Its sentence-length measure depends on modern editors' punctuation.

1. 29a.

   Page 90:  “Three of the Epistles, II Thessalonians, Titus and Philemon are excluded from the analysis as they are not large enough to provide reliable samples in tests of this type.” ↩

Neumann, Kenneth J.  “The Authenticity of the Pauline Epistles in the Light of Stylostatistical Analysis.”  SBL Dissertation Series 120.  Atlanta:  Scholars Press.  1990.  A doctoral study testing many stylistic measures on known-author samples of Greek, then applying the best combinations to Ephesians, Colossians and 2 Thessalonians, with the Pastorals and 1 Peter among the comparison groups.  Its candour about its own failures makes it useful.

1. 30a.

   Page 195:  “All disputed letters, including all three Ephesian sam- ples, are assigned to the Pauline group with a rounded-off probability of 1.0”.  “These results, despite the Pauline classification, may still indicate a non-Pauline authorship. Before discussing this further, a disturbing result needs to be treated.”  “for VARSET1 the Galatian sample is classified with Ignatius with a probability of 0.9910”. ↩

2. 30b.

   Page 218:  “the authors of the Pastorals and 1 Peter since they are classed with non-Pauline groups with a high degree of probability”.  “Only in cases where all the potential authors for disputed docu- ments are known may discriminant analysis or other stylistic analyt- ical methods decide authorship”. ↩

van Nes, Jermo.  “Hapax Legomena in Disputed Pauline Letters:  A Reassessment.”  Zeitschrift für die neutestamentliche Wissenschaft 109(1):118-137.  2018.  A regression test of whether each disputed Pauline letter uses more once-only words than its length predicts, using Paul's seven undisputed letters as the baseline.  It is the main length-controlled test of the classic vocabulary argument.

1. 31a.

   Page 128:  “Colossians, Ephesians, and 2 Thessalonians neatly fit the undisputed Pauline pattern”.  “Titus among the Pastorals also fits this pattern, be it close to the maximal prediction value. At this stage, it appears that only the numbers of hapaxes in the Timothy correspondence are significantly higher than in the other Paulines.” ↩

Harrison, P. N.  The Problem of the Pastoral Epistles.  Oxford:  Oxford University Press.  1921.  The classic vocabulary study arguing that the Pastoral Epistles were written in the second century around genuine fragments of Paul.  Its counts are per page of printed text and are not adjusted for the length of each letter.

1. 32a.

   Page 20:  “The vocabulary of the Pastorals consists of some 902 words, of which 54 are proper names. Of the remaining 848, 306 or over 36 per cent. are not to be found in any one of the ten Paulines.” ↩

Kestemont, Mike et al.  “Authenticating the Writings of Julius Caesar.”  Expert Systems with Applications 63:86-96.  2016.  A computational authorship study that first validates its verification method on benchmark sets, including Latin authors cut into 1,000-word slices, and then applies it to the Caesarian war commentaries.  Locators are to the authors' accepted manuscript.

1. 33a.

   Section 4.2, Latin Data; 5, Caesar's Writings:  “cut up in non-overlapping slices of 1000 words; while this constitutes a challengingly limited document size”.  “Hirtius wrote Book 8 of the Gallic Wars and the remainder of the Alexandrian War.” ↩