Exposing Textual Corruption: Community as a Stabilizing Aspect in the Circulation of the New Testament Writings during the Greco-Roman Era

Because few manuscripts of the NT writings are preserved from the first three centuries of the Christian era, scholars have debated the extent that modern critical editions of the NT reflect the text in circulation during these early centuries. In order to answer this question, this article will set out the evidence for ancient publication through community transmission. It will consider examples from Cicero, Martial, Quintilian, Pliny the Younger and Galen. These authors reveal that they preferred social networks rather than commercial dealers to circulate their writings. These same communities that copied and distributed an author’s works inadvertently created an environment in which significant alterations and plagiarizing of these same writings became known. Matthew D.C. Larsen, who has recently approached the same problem addressed in this article by examining ancient publication conventions, is engaged with throughout. The conclusions drawn here press hard against Larsen’s assertions.


Introduction
Though there are many manuscripts that preserve portions of the NT writings, very few of these manuscripts date to within the first millennium of the Christian era, and only a handful within the second and third centuries. 1 This paucity in early manuscript attestation, along with other evidence, has continued to pressure scholars into re-evaluating the traditional goals of NT textual criticism. 2 Rather than asserting that textual critics must recover a specific document's 'original text' (however defined), some scholars contend that each particular reading is intrinsically valuable as a 'window' into the various Christian communities that copied, circulated and read these texts. 3 In addition, others remain skeptical towards the idea that the earliest recoverable text (that of the late second century) actually represents the forms these writings took when they were first released for circulation and copying (Larsen 2018: 4-5;2017: 376-79;Parker 2012: 24-25). Helmut Koester articulated this criticism well when he wrote that, New Testament textual critics have been deluded by the hypothesis that the archetypes of the textual tradition which were fixed ca. 200 ce -and how many archetypes for each gospel? -are (almost) identical with the autographs. This cannot be affirmed by any evidence. On the contrary, whatever evidence there is indicates that not only minor, but also substantial revisions of the original texts have occurred during the first hundred years of the transmission (Koester 1989: 37). According to Koester, the evidence is clear, 'New Testament textual critics have been deluded' into a false textual confidence of the form of the archetypes of the gospels in the first century.
Conveying equal skepticism, William L. Petersen, after examining the quotations of the gospels in the church fathers, and looking at the scarcity of early manuscript attestation, determined that, 'We know next to nothing of the text of the gospels in the first century, for we have no manuscript evidence and few (if any) Patristic writings' (Petersen 2002: 53-54). A few pages later in the same work Petersen expressed his conclusions more fully: To be brutally frank, we know next to nothing about the shape of the 'autograph' gospels; indeed, it is questionable if one can even speak of such a thing. This leads to the inescapable conclusion that the text in our critical editions today is actually a text which dates from no earlier tha[n] about 180 ce, at the earliest. Our critical editions do not present us with the text that was current in 150, 120 or 100 -much less in 80 ce (Petersen 2002: 62).
In a similar manner, D.C. Parker conveyed this uncertainty by pointing to the quotations of the NT writings by the early church fathers. These allusions and 1. The Liste on the Institut für Neutestamentliche Textforschung website offers over 5,000 Greek manuscripts for the NT writings (http.//ntvmr.uni-muenster.de/liste). 2. For an evaluation of the traditional goals of NT textual criticism, see Epp 1999. For a more recent assessment of the traditional goal of NT textual criticism, see Holmes 2013. 3. For a discussion of textual readings as windows into church history, see Parker 1997;Ehrman 2013aEhrman , 2011 quotations are 'frequently different from those found in later manuscripts', and this and other evidence 'press hard upon anyone wishing to claim that the oldest forms of the manuscript tradition of the NT writings are also forms of text dating from the beginning of the second century and earlier' (Parker 2008: 117-118). Not only does Parker voice uncertainty for the text of the gospels, he also expands this doubt to broadly encompass the NT writings as a whole. Eldon Epp communicated comparable views (earlier using nearly identical language as Parker later used for his work The Living Text of the Gospels) when he wrote that 'the text of the New Testament in its earliest stage was a vibrant, living text that functioned dynamically within the developing church' (Epp 1989: 75).
As a solution to the scarcity of early manuscript support, Michael W. Holmes has suggested that the 'trends, patterns, and tendencies from a later period for which we have evidence' be projected 'back into the earliest period for which we lack evidence' in order to understand the earliest shape of the text (Holmes 2011: 78). This is not unlike Petersen's approach, the conclusions of which were already mentioned above. 4 This method has led Holmes to conclude that 'all the variation during the time period in view affects a verse or less of the text' indicating that the text of the NT writings is 'characterized by macro-level stability and micro-level fluidity' (Holmes 2011: 78). This is in contrast to Petersen, who, using a similar approach, determined that 'our critical editions do not present us with the text that was current in 150, 120 or 100 -much less in 80 ce' (Petersen 2002: 62).
Despite the disparity of these conclusions, each of the studies mentioned above fails to adequately consider the effects of ancient publication and circulation practices upon the transmission of the text. 5 More recently, Matthew Larsen has published a monograph that attempts to fill this gap and apply ancient composition and publication practices to the transmission of the gospel texts (Larsen 2018). After engaging with a host of Greco-Roman and Jewish primary sources that discuss note taking (ὑπομνήματα) and demonstrate an 'accidental publication', Larsen concludes that 'the reality is we do not have access to the textual tradition we now call the Gospel according to Mark as it existed in the first century' (Larsen 2018: 151). 6 As a way of moving beyond the impasse, this article will survey Greco-Roman 'publication' conventions by examining several Greek and Roman figures (both 4. Petersen stated that 'the only route of inquiry is to take what we have discovered thus far, from our study of the second century, project these trends and tendencies back into the first century, and see what they suggest ' (2002: 54). 5. Holmes briefly references Greco-Roman publication conventions when defining the 'earliest recoverable stage(s) of the text's transmission' (2013: 657-58). 6. Chapter 3 is titled 'Accidental Publication and Postpublication Revision'. Larsen made similar comments in 2017: 379-80. pagan and Christian) from the beginning of the Roman imperial age through to its decline. 7 Whenever appropriate, this article will dialogue with Larsen's work with regard to Greco-Roman publication. In tension with Larsen's work, this article proposes that, because the NT writings were transmitted and circulated through Greco-Roman conventions of community distribution, this naturally produced a condition in which the plagiarizing and 'macro-level' alteration of these writings would have been exposed within these same community circles.

The Circulation of Books in the Greco-Roman World
The term 'book' is used in this article to describe the Greco-Roman medium for texts and refers to both the book roll and the codex, the ancient predecessor to the modern book. The book roll was written entirely by hand in columns of continuous text on long sheets of papyrus or parchment and then rolled up with the text on the inside. The codex, in contrast, was constructed from leaves of parchment or papyrus (usually cut from a blank roll) stacked together, folded down the center and then stitched along the folded spine with the hand-written text on both sides of the pages (Johnson 2009: 256-77). Contemporary notions of publication with the mass production of books, marketing, copyright laws and rules of plagiarism did not exist in the Greco-Roman world. Any theories that concern the transmission of the NT writings in the stages before our earliest manuscripts must incorporate the techniques of book production available to the authors of these Christian writings and the early communities who circulated them. A study of the procedures used in the ancient Mediterranean culture to disseminate works of literature may illuminate the processes by which the NT documents were first circulated and continued to disseminate on into the following centuries of the Roman Imperial age.
There have been several studies over the past one hundred years and more that have surveyed classical, Hellenistic and Greco-Roman primary sources, as well as ancient material remains that shed light on publication in antiquity. 8 Older 7. In order to ensure that the article remains at a manageable length, Latin authors, residing primarily in Rome, are cited more frequently than authors present in other locations in the Mediterranean. Of course, there is risk of misrepresenting the writing cultures of the Mediterranean more broadly. It is highly probable, however, that the writing and circulation practices operating in Rome are representative of practices across the Empire. After examining the Oxyrhynchus papyri for the time and distance it took to send and receive letters and documents, Eldon Epp concluded that documents could 'in a matter of a few weeks, have moved anywhere in the Mediterranean area ' (1991: 55-56). Brian J. Wright has exhaustively documented that the dissemination of literature through communal reading (a form of 'publication') spans the entire Mediterranean basin (2017: 37-38, 208-209). See also the comments in n. 27. 8. White 2009;Iddeng 2006;Gamble 1995: 83-93;Starr 1987;Phillips 1986;Quinn 1982; treatments of the subject anachronistically compared the ancient circulation of literature with modern commercial publication. 9 More recent studies, however, have either completely minimized the commercial production of books in Imperial Rome or have taken a mediate position (Starr 1987: 221;Iddeng 2006: 63-64, 77-78).
Both the commercial and private (or, more accurately, communal) production and distribution of literature can be seen in the letters of Pliny the Younger. Writing to the historian Suetonius, Pliny urged him to publish his work, declaring that he wanted to 'hear that my friend's books are being copied, read and sold' (Ep. 5.10). 10 Here, each avenue of circulation is alluded to. Copying appears to be a reference to the distribution through scribal duplication of the writing amongst Suetonius's associates. Selling is likely an allusion to the placement of copies to be sold in book shops. Reading is likely a reference to the reading out of the work in a communal setting at dinner parties and other social events. 11 Despite the risk of anachronism, in the interests of simplicity, the term 'publication' is used in this article to refer to this ancient practice of circulating a written work.
The commercial and communal distribution of books is evident in the remains of a letter found in the Roman City of Oxyrhynchus. 12 Dating to approximately the second century ce, P.Oxy 18.2192 (TM29029) is a fragmentary papyrus in which the main text of the letter no longer survives. All that remains are two postscripts written in a different hand to the text of the letter. The first postscript was written by the sender and was written in a different hand likely because the sender, though obviously literate, used a secretary to compose the letter. The first postscript reads, Make and send me copies of Books 6 and 7 of Hypsicrates' Characters in Comedy [or Topics in Comedy]. For Harpocration says they are among Polion's books. But it is likely that others too have got them. He also has prose epitomes of Thesagoras' Myths of Tragedy, 71(?).
̔ Yψικράτους τῶν κωμωιδουμένων ς̅ ζ̅ ποιήσας μοι πέμψον. φησὶ γὰρ Ἁρποκρατίων ἐν τοῖς Πωλίωνος αὐτὰ βιβλίοις ε ί͂ ναι. εἰκὸς δὲ καὶ ἀλλους αὐτὰ εσχηκέναι. καὶ λόγου ε̣ ̓ π̣ ιτο̣ μὰς τῶν Θερσαγόρου τῶν τραγικῶν μύθων ἐχει. 13 Schubart 1962;Reichmann 1938;Kenyon 1932;Sommer 1926;Dziatko 1900;Birt 1882. 9. See, for example, Birt 1882. 'Patere me videre titulum tuum, patere audire describi legi venire volumina Tranquilli mei.' The Latin text of this quotation and the following texts and English translations of Pliny's letters are taken from Radice 1969. See also Sherwin-White 1966. On communal reading, see Wright 2017. 12. Roberts 1941: 150-52. See also Hatzilambrou 2007. Translations of both post-scripts are taken from Hatzilambrou 2007: 284-85. The second postscript is in a different hand and was made by the recipient of the letter who possibly returned it to the sender in reply (Hatzilambrou 2007: 283). It reads, According to Harpocration Demetrius the bookseller has got them. I have instructed Apollonides to send me some of my books, which you will hear about from him. And of Seleucus' work on the Tenses make copies and send me as many (books) as you find, apart from those I possess. Diodorus (and his circle?) also have some that I do not possess. This fragmentary papyrus gives a glimpse into the circulation and copying of books amongst social networks. Here two circles of educated acquaintances are exchanging and borrowing books in order to procure copies of them. A bookseller is briefly mentioned, but the exchange of literature occurs mainly by appropriating through a community of scholars and making personal copies from these borrowed books (Johnson 2009: 270-71). The informal nature of distribution often led to a convoluted composition, editing and releasing process. Pliny gave a detailed description of this type of editing and releasing process.
First of all, I go through my work myself; next, I read it to two or three friends and send it to others for comment. If I have any doubts about their criticisms, I go over them again with one or two people, and finally I read the work to a larger audience; and that is the moment, believe me, when I make my severest corrections, for my anxiety makes me concentrate all the more carefully.
Ac primum quae scripsi mecum ipse pertracto; deinde duobus aut tribus lego; mox aliis trado adnotanda, notasque eorum, si dubito, cum uno rursus aut altero pensito; novissime pluribus recito, ac si quid mihi credis tunc acerrime emendo; nam tanto diligentius quanto sollicitius intendo. (Ep. 7.17) 14 During these draft and re-writing stages the author assumed that the work in progress would not be circulated beyond this immediate group of friends. 15 This trust was sometimes violated, however, and portions of an unfinished writing could circulate without the author's consent. The author had no control over the text once the writing began to circulate beyond the immediate acquaintances of the author. Writing in the last half of the first century bce, Horace warned that, 'What you have not published you can destroy; the word once sent forth can never come back . 16 This occurred in the case of Pliny's friend Octavius to whom Pliny wrote, Some of your verses have broken free in spite of you and have become more widely known; unless you recall them to be incorporated in the whole, like runaway slaves they will find someone else to claim them.
Enotuerunt quidam tui versus, et invito te claustra sua refregerunt. Hos nisi retrahis in corpus, quandoque ut errones aliquem cuius dicantur invenient. (Ep. 2.10) Octavius attempted to control the exchange of his poems, probably limiting the circulation to trusted associates who provided constructive criticism. 17 Despite his efforts, some of the verses circulated beyond this intimate group and were in danger of being falsely attributed to another author. Even though Octavius was losing control over his work, the community of literate acquaintances was intimate enough for Pliny to notify his friend of the unauthorized distribution of the verses.
The epigrammatist Martial, writing in the last decades of the first century, often alluded to the publication of his work throughout his poems. 18 Similar to Pliny, Martial recited draft versions of his poems to friends and associates before releasing them for circulation. 19 As a result, his work was susceptible to plagiarism. 20 Martial alludes to this when he wrote, You are mistaken, greedy purloiner of my books, in thinking that it costs no more to become a poet than the price of copying and a cheap length of papyrus … You must look for private, unpublished work, poems known only to the parent of the virgin sheet, which he keeps sealed up in his book-box … A well-known book cannot change 16. 'delere licebit quod non edideris; nescit vox missa reverti.' Latin text and English translation taken from Fairclough 1942: 483. See also Horace's comments in Epist. 1.20. 17. Pliny mentioned having a glimpse at some of Octavius's work in Ep. 1.7. 18. Martial mentions publication in Epig. 1.2, 3, 16, 29, 66;2.8, 20;3.38;5.16;11.2;3;14.194 (Iddeng 2006: 64 n. 21). For a brief discussion on the dating of Martial's epigrams, which were published in the 80s and 90s ce, see the introduction in Williams 2004: 4-5. 19. Epig. 5.80;Gurd 2012: 108-10. See also Sage 1919: 168-76. 20. I use the term 'plagiarism' in a general sense as defined by Scott McGill: 'the culpable reuse of earlier texts, customarily described in terms of stealing, in which a person wins false credit by presenting another's work as his own' (McGill 2012: 3). Although the term 'plagiarism' has its source in Greek (πλάγος, πλάγιος), modern concepts of plagiarism are far removed from ancient ideas of literary borrowing and authorship, though Olcott argues that in the earliest period of Classical Greece there are more similarities with modern ideals than in the Christian era (2002: 1048, 1052). See also the discussion on Galen below in the section "5. Implications for Textual Criticism." author … Whoever recites other men's productions and seeks fame thereby, ought to buy -not a book, but silence.
Erras, meorum fur avare librorum, fieri poetam posse qui putas tanti, scriptura quanti constet et tomus vilis: … quas novit unus scrinioque signatas custodit ipse virginis pater chartae, … mutare dominum non potest liber notus … aliena quisquis recitat et petit famam, non emere librum, sed silentium debet. (Epigr. 1.66) (Bailey 1993: 87-89) Notice here what Martial is asserting, once a piece was circulating in a community of readers, that writing could not be easily plagiarized without the theft becoming exposed within this same community. The theft of Martial's work was caught presumably due to its wide circulation. Therefore, an aspiring author who wished to publish another author's work as their own would have to do so from writings not yet in circulation. 21

Accidental Publication
For this phenomenon of unauthorized circulation Larsen uses the phrase 'accidental publication' which he defines as when 'the author claims to have no knowledge of how a text became public and such that publication was against his will' (Larsen 2018: 38;2017: 370). Larsen cites several ancient figures such as Cicero, Diodorus, Horace, Josephus and Arrian in order to illustrate this occurrence. 22 In many of these examples, the authors and copyists communicate that a work of some kind -a speech, poem, history or lecture -was circulated without their consent, and this unauthorized distribution was made known to them. The following example is taken from Larsen's work in order to illustrate his argument. Cicero wrote to his life-long friend Atticus in response to some disturbing news he had heard from him.
Then you have given me a blow about the circulation of that speech. Try, as you say, to patch up the damage if you can. I did write it long ago in a fit of annoyance with him because he had written against me, but I suppressed it and never expected it to leak out. How it did get out I do not know.
percussisti autem me etiam de oratione prolata. cui vulneri, ut scribis, medere, si quid potes. scripsi equidem olim iratus quod ille prior scripserat, sed ita compresseram ut numquam emanaturam putarem. quo modo exciderit nescio. (Att. 3.12) 23 Though Cicero does not know how the speech began to circulate, Atticus is the one who first informed him that the oration was making the rounds. Though he 21. See Larsen's comments on this passage in 2018: 18. 22. See Chapter 3 of Larsen 2018: 37-58. 23. This and the following translations are taken from Bailey 1999. attempted to disown the personally damaging speech, he was ultimately unsuccessful, and portions of this discourse are extant to this day. 24 Though Larsen defines 'accidental publication' as the author having 'no knowledge of how a text became public', there are instances in which Cicero learns the means by which a particular work was prematurely released. In a letter to his long-time friend Atticus, Cicero chastised him for allowing a segment of his unfinished manuscript to circulate.
Come now, in the first place do you approve of publishing without my instructions? … And another thing: do you think it was right to give the book to anyone before Brutus, to whom at your suggestion I address it? Balbus writes to me that he has copied 'On the Limits' [De Finibus], Book V, from your manuscript, a book in which I have made changes, not many to be sure, but still some.
Dic mihi, placetne tibi primum edere iniussu meo? … quid illud? rectumne existima<s> cuiquam <ante quam> Bruto, cui te auctore προσφωνῶ? scripsit enim Balbus ad me se a te quintum de finibus librum descripsisse; in quo non sane multa mutavi, sed tamen quaedam. (Att. 13.21a) Atticus must have presumed that Cicero's De Finibus was completed and had sent a draft version of Book 5 to their mutual friend Balbus. 25 Even though Cicero depended completely upon the faithfulness of his inner group of friends to keep tight control over his draft composition, Cicero inevitably learned when this trust was broken. Their mutual friend Balbus notified Cicero by letter when he had received Book 5. This was the means by which Cicero learned that his work was 'accidentally published'.

Postpublication Revision
Larsen also makes reference to Cicero (along with other Greco-Roman authors) in order to draw attention to the ancient practice of 'postpublication revision' (Larsen 2018: 49). Larsen highlights a request that Cicero made to Atticus concerning an error that remained uncorrected in his Orator. 26 To Atticus Cicero wrote, 'Chremes, so little of your own to do' that you are actually reading the Orator? Bravo! I am pleased to hear it, and shall be still better pleased if you will get your clerks to substitute Aristophanes for Eupolis both in your own copies and in other people's. 24. See the discussion in Larsen 2018: 39 (Bailey 1999: 269) Here Cicero requested Atticus to correct a name entered mistakenly in the text of his Orator, from Eupolis to Aristophanes. Though the book had already been released for copying, only the corrected text is present in all of our surviving manuscripts (Starr 1987: 29). In this instance it is likely that the circulation was small enough for changes to be made. Another example, not referenced by Larsen, reveals that Cicero also edited his work for style or clarity and not just for the correction of mistakes. To his friend Atticus Cicero wrote, To come to business, I now do not like one bit your word inhibere, which I had thought the very thing … So will you please put it back in the book as it was, and tell Varro [the dedicatee of the work] in case he has made the alteration?
Nunc, ad rem ut redeam, 'inhibere' illud tuum, quod valde mihi adriserat, vehementer displicet … qua re facies ut ita sit in libro quem ad modum fuit. dices hoc idem Varroni, si forte mutavit. (Att. 13.21) (Bailey 1999: 133) Seemingly, Cicero ordered these changes after the book had already been completed and a presentation copy given to the dedicatee Varro. Though the work had been released for transmission, the initial circulation must have been small enough for Cicero to direct changes in all the known copies.

Exposing Textual Corruption
Larsen contends that factors such as 'accidental publication' and 'postpublication revision' call into question the modern conception that '[a] text that has survived to the present must have been published and therefore be in its final form as intended by the author' (Larsen 2018: 37). Though Larsen is correct to push back against anachronistic perceptions of publication, the mechanics of ancient circulation and copying tended to subject a textual fixity to a released document. Though an exhaustive analysis of ancient authors from the beginning of the Roman Imperial age through to its height is outside the scope of this article, a survey of ancient writers from the period in view should paint a picture of the effects of publication on the text of a circulating document.

Greco-Roman Writers 27
Speeches and lectures delivered to the public provided an ideal situation for unauthorized copying, especially if they were delivered by a famous orator embroiled 27. Classifying the following authors as 'Greco-Roman' creates the risk of anachronistically drawing together potentially disparate elements in date and culture. A sharper distinction in a high-profile incident (White 2009: 279-80). These circumstances occurred when Cicero delivered an address defending his friend Milo against charges of murder, and this speech was then transcribed and circulated before Cicero could release an edited version (White 2009: 279). Plutarch wrote that when Cicero 'came out of his litter to plead Milo's cause and saw Pompey stationed on the heights as in a camp, and arms flashing all around the forum, he was confounded and could scarcely begin his speech, for his body quivered and his voice faltered' (Cic. 35). 28 According to Asconius, Cicero's original speech was bumbling and filled with errors, and he later supplanted this earlier raw transcript by releasing a polished edition and it is this subsequent version that survives today (White 2009: 279;Marshall 1987: 731-32). He wrote that, What he actually said was taken down and also survives, but the speech that we are reading is what he composed in writing, and with such consummate skill that it may rightly be reckoned his finest.

Pro. Mil. 42C) 29
Though Cicero had two forms of his speech circulating -the first being copied without his consent by stenographers, and the second edited and polished version subsequently released -decades later, Asconius was able to differentiate between the two versions and preferred the authorized edition. 30 In this instance, once the speech was taken down and circulated, Cicero had no control over its fate, and the text was inadvertently 'finalized' at this point. For Cicero to make any could be made between Roman figures of the Republic such as Cicero and Greek figures of the Empire such as Galen. A common unifying factor, however, between these potentially disparate figures is the common educational practices throughout the Mediterranean basin. Whether in Rome or in the provinces, a similar method of Greek classical education was employed (Cribiore 2001: 6-7). Broadly speaking, this would result in a commonality in writing and reading culture. Though the fall of the Roman Republic and the rise of the Empire brought a slew of economic and cultural changes in places like Egypt, there was a broad uniformity in education from the Republican age of Cicero through to the height of the Imperial age of Galen. See the observations of Cribiore 2001: 8. For a thorough treatment of the interplay between Christian and Pagan education for the period under discussion, see Gemeinhardt 2007. 28. τότε δ᾿ οὖν ἐπὶ τὴν τοῦ Μίλωνος δίκην ἐκ τοῦ φορείου προελθὼν καὶ θεασάμενος τὸν Πομπήϊον ἀνω καθεζόμενον ὥσπερ ἐν στρατοπέδῳ, καὶ κύκλῳ τὰ ὁπλα περιλάμποντα τὴν ἀγοράν, συνεχύθη καὶ μόλις ἐνήρξατο τοῦ λόγου, κραδαινόμενος τὸ σῶμα καὶ τὴν φωνὴν ἐνισχόμενος … (Perrin 1919: 172-73). 29. Lewis 2006: 84-85. See also the comments of Dio Cassius 40.54.3-4. 30. Centuries later, Dio spoke only of the published speech being in existence but knew of a different version (40.54.2). Marshall 1987: 732. See also Settle 1963: 269-80. changes to the text, it was necessary for another more edited version of the speech to be subsequently released.
As was the case with public speeches delivered in court by Cicero and other orators, lectures presented to students provided another opportunity for unauthorized copying and distribution. Near the end of the first century, the prominent rhetorician Quintilian began to compile his magnum opus, Institutio Oratoria, which was a 'lengthy dissertation on the finer points of raising a gentleman to the art'. 31 In the preface to Book 1, while dedicating the work to his friend Marcellus Victorius, Quintilian indicated that some of the material in his Institutio might be found in an unedited form circulating under his name.
[T]wo books on the Art of Rhetoric are already circulating in my name, though they were never published by me nor prepared for this purpose. One is a two days' lecture course which was taken down by the slaves to whom the responsibility was given. The other lecture course, which spread over several days, was taken down by shorthand (as best they could) by some excellent young men who were nevertheless too fond of me, and therefore rashly honoured it with publication and wide circulation. In the present work, therefore, there will be some things the same, many things changed, and very many things added, and the whole will be better written and worked up to the best of my ability.
atque eo magis quod duo iam sub nomine meo libri ferebantur artis rhetoricae neque editi a me neque in hoc comparati. Namque alterum sermonem per biduum habitum pueri quibus id praestabatur exceperant, alterum pluribus sane diebus, quantum notando consequi potuerant, interceptum boni iuvenes sed nimium amantes mei temerario editionis honore vulgaverant. Quare in his quoque libris erunt eadem aliqua, multa mutata, plurima adiecta, omnia vero compositiora et quantum nos poterimus elaborata. (Inst. Or. 1. Pr. 7-8) (Russell 2002a: 54-57) Quintilian learned that his lectures had been copied and were circulating widely in an unedited form and thus he could no longer direct the quality of the text or its content. He knew that Marcellus had either read these transcribed lectures already or would likely come across them in the future and was careful to indicate that the Institutio was a new work that supplanted these crude notes. In addition to his lectures, several of his speeches had also been copied down by scribes who were expert in shorthand. In Book 7 of Institutio he wrote.
My speech in this Cause is the only one I have so far published, and I admit that I was induced to do so by a youthful desire for fame. The other speeches circulating under 31. Johnson 2010: 26. Quintilian enjoyed unrivaled fame as an orator as he was the first to be appointed Rome's official rhetorician and tutored Emperor Domitian's heirs (White 2009: 278). See also Martial's comments in Epig. 2.90. For a detailed discussion on the Institutes, see also Adamietz 1986. my name, corrupted as they are by the negligence of the shorthand-writers who took them down to make money, have very little of me in them.
Cuius actionem et quidem solam in hoc tempus emiseram, quod ipsum me fecisse ductum iuvenali cupiditate gloriae fateor. Nam ceterae quae sub nomine meo feruntur neglegentia excipientium in quaestum notariorum corruptae minimam partem mei habent. (7.2.24) (Russell 2002b: 198-99) As Quintilian notes, the circumstances by which his speeches were copied by shorthand were different than for the lectures that were the foundation for his Institutio. His speeches were specifically transcribed without his consent in order to steal them and be sold for a profit. 32 Unlike his lectures on the Art of Rhetoric, he disowned these crudely copied speeches, yet, because they were already circulating, he could not guarantee their quality. The physician Galen of Pergamum flourished in the last half of the second century, practicing medicine for Rome's social elite and authoring hundreds of treatises. 33 Because many of these treatises circulated haphazardly and without Galen's consent, he wrote two works discussing the composition and publication of his many writings, De libris propriis and De ordine librorum suorum (Mattern 2008: 12). In De libris propriis, Galen provided a unique glimpse into the circulation practices of Greco-Roman society when he complained that 'my books have been subject to all sorts of mutilations, whereby people in different countries publish different texts under their own names, with all sorts of cuts, additions, and alterations' (De libr. propr. 19.9). 34 Galen wrote that the corruptions of his books occurred because several of his writings were given without inscription to friends or pupils, having been written with no thought for publication, but simply at the request of those individuals, who had desired a written record of lectures they had attended. When in the course of time some of these individuals died, their successors came into possession of the writings, liked them, and began to pass them off as their own [...] Taking them from their owners, they returned to their own countries, and after a short space of time began to perform 32. Quintilian's fame created such a demand for his works that in the preface to the Institutio, Quintilian wrote that the bookseller Tryphon had been pressuring him to complete it (White 2009: 278). Tryphon appears to have been the same bookdealer who sold Martial's works (Epigr. 4.72.2;13.3.4;Gamble 1995: 86). 33. For a detailed study of his life works, see Mattern 2008. For a full list of Galen's works, see Fichtner 2012. Peterson 1977. διά τε οὖν αὐτο τοῦτο καὶ διότι πολυειδῶς ἐλωβήσαντο πολλοὶ τοῖς ἐμοῖς βιβλίοις, ἀλλοι κατ' ἀλλα τῶν ἐθνῶν ἀναγιγνώσκοντες ὡς ἰδια μετὰ τοῦ τὰ μὲν ἀφαιρεῖν, τὰ δὲ προστιθέναι, τὰ δὲ ὑπαλλάττειν. This and the following English translations of Galen are taken from Singer 1997. The Greek text is taken from Boudon-Millot 2007. the demonstrations in them, each in some different way. All these were eventually caught, and many of those who then recovered the works affixed my name to them. They then discovered discrepancies between these and copies in the possession of other individuals, and so sent them to me with the request that I correct them.
Apparently, many of Galen's compositions were not meant to circulate as formal works of literature with his name affixed. Instead they were merely Galen's unedited lecture notes. 35 Once these transcripts were given to his students, he lost control over their fate and they were then misappropriated and altered by his student's successors. Although he does not mention the persons by name, or the nature of their relationship to Galen, the community of students, followers and readers of his work were the persons responsible for notifying him when his lecture notes had been stolen and corrupted. They clearly distinguished between the authored text and the alterations made by others after the notes began to circulate and preferred the text authored and edited by Galen. A length of time must have elapsed because the plagiarizing of his notes occurred after 'some of these individuals died' (De libr. propr. 19.10). This indicates that the community played a role in exposing textual corruption long after a document began to first circulate. 36

Christian Writers
The writings of the NT and the Apostolic Fathers allude to similar avenues of circulation that were employed in the wider Greco-Roman world. In Col. 4.16, the author directed the recipients: 35. See Larsen's comment on this passage, 2018: 29. 36. Even when a book circulated through booksellers, an author often received reports on the status of their writings by acquaintances who came across their work being sold in bookstores across the Empire. Pliny learned from a friend that his books were for sale in Lugdunum Gaul (Ep. 9.11), and Galen learned from an acquaintance that someone in a Roman bookstore was attempting to sell an inferior work with a title attached reading 'Galen the Doctor' (De libr. propr. 19.8-9).
And when this letter has been read among you, have it also read in the church of the Laodiceans, and see that you also read the letter from Laodicea.
καὶ ὁταν ἀναγνωσθῇ παρ' ὑμῖν ἡ ἐπιστολή, ποιήσατε ἵνα καὶ ἐν τῇ Λαοδικέων ἐκκλησίᾳ ἀναγνωσθῇ, καὶ τὴν ἐκ Λαοδικείας ἵνα καὶ ὑμεῖς ἀναγνῶτε. 37 Two avenues of distribution are in play here, private copying and communal reading. Presumably, the Colossians would send the original or a copy of the letter on toward Laodicea in order to read it out before the congregation (Gamble 1995: 97-98). The author of 1 Timothy directed the reader to 'devote yourself to the public reading of scripture' (4.13) and John assumed that Revelation would be read out to a Christian gathering when he wrote, 'Blessed is the one who reads aloud the words of this prophecy, and blessed are those who hear' (1.3), 38 that is, one reading and many hearing what was being read (Wright 2017: 195). Similar methods of circulation were employed in the writings of the Apostolic Fathers. Polycarp, Bishop of Smyrna, sent a letter to the Philippian church. 39 He wrote: We are sending to you the letters of Ignatius that were sent to us by him together with any others that we have in our possession, just as you requested. They are appended to this letter; τὰς ἐπιστολὰς Ἰγνατίου τὰς πεμφθείσας ἡμῖν ὑπ ̓ αὐτοῦ, καὶ ἀλλας ὁσας εἰχομεν παρ ̓ ἡμῖν, ἐπέμψαμεν ὑμῖν, καθὼς ἐνετείλασθε· αἵτινες ὑποτεταγμέναι εἰσὶν τῇ ἐπιστολῇ ταύτῃ· (Pol. Phil. 13.1-2) 40 Polycarp reveals that Ignatius's letters were being copied and distributed throughout Asia Minor as churches requested copies. This appears to be a similar practice to that mentioned in Col. 4.16.
Another glimpse into the publishing practices of the early Christians can be seen in the Shepherd of Hermas. At one point the narrative features an 'elderly woman' (who represents the church) giving Hermas specific instructions to distribute a message contained in a 'little book'. Therefore you will write two little books, and you will send one to Clement and one to Grapte. Then Clement will send it to the cities abroad, because that is his job. But Grapte will instruct the widows and orphans. But you yourself will read it to this city, along with the elders who preside over the Church.
γράψεις οὖν δύο βιβλαρίδια καὶ πέμψεις ἑν Κλήμεντι καὶ ἑν Γραπτῇ. πέμψει οὖν Κλήμης εὶς τὰς ἐξω πόλεις, ἐκείνῳ γὰρ ἐπιτέτραπται. Γραπτὴ δὲ νουθετήσει τὰς χήρας καὶ τοὺς ὀρφανούς. σὺ δὲ ἀναγνώσῃ εἰς ταύτην τὴν πόλιν μετὰ τῶν πρεσβυτέρων τῶν προϊσταμένων τῆς ἐκκλησίας. (Herm. Vis. 2.4) (Holmes 2007: 468-69) Once again two primary avenues of distribution are illustrated. Circulation by copying and distributing through social networks was to be performed by Clement who was to distribute the work throughout the cities. And Grapte was directed to teach (likely involving communal reading), and Hermas was to read the message to the inhabitants of 'this city' (most likely Rome). Another example is found at the end of the Martyrdom of Polycarp in a series of colophons. 41 These colophons illustrate how scribes, presumably in the late second or early third century, acquired copies of important Christian works through church communities and scribal networks.
Gaius [who lived in the same city as Irenaeus] transcribed this account from the papers of Irenaeus, a disciple of Polycarp. And Ι, Socrates, wrote it down in Corinth from the copies of Gaius. Grace be with everyone.
Ταῦτα μετεγράψατο μὲν Γάϊος ἐκ τῶν Εἰρηναίου, μαθητοῦ τοῦ Πολυκάρπου, ὁς καὶ συνεπολιτεύσατο τῷ Εἰρηναίῳ. ἐγὼ δὲ Σωκράτης ἐν Κορίνθῳ ἐκ τῶν Γαίου ἀντιγράφων ἐγραψα. ἡ χάρις μετὰ πάντων. (Mart. Poly. 22.2) (Holmes 2007: 330-31) The methods employed by the Christians of the first and second centuries to distribute their literature were very similar to that found in Roman writers of the same era. Consequently, in the same manner as Greco-Roman communities, we should be able to see a similar phenomenon among the early Christians with regard to the alteration of texts, that is, a correlation between the wide circulation of a writing and the exposure of significant textual alterations.
The unauthorized publication of texts plagued Christian authors as well. The apologist and theologian Tertullian of Carthage was the victim of theft during the composition of his treatise Adversus Marcionem. 42 At the beginning of the 41. For a discussion on the implications of this passage, see Haines-Eitzen 2000: 80-81. 42. For a survey of Tertullian, see Ferguson 2009. For a detailed analysis, see Alexandre 2012;Tränkle 1997. treatise Tertullian alerted the reader to the circumstances surrounding its publication.
Nothing I have previously written against Marcion is any longer my concern. I am embarking upon a new work to replace an old one. My first edition, too hurriedly produced, I afterwards withdrew, substituting a fuller treatment. This also, before enough copies had been made, was stolen from me by a person, at that time a Christian but afterwards an apostate, who chanced to have copied out some extracts very incorrectly, and shewed them to a group of people. Hence the need for correction. The opportunity provided by this revision has moved me to make some additions. Thus this written work, a third succeeding a second, and instead of third from now on the first, needs to begin by reporting the demise of the work it supersedes, so that no one may be perplexed if in one place or another he comes across varying forms of it. Apparently, there were three separate editions of the treatise that were released over a period from around 198 to 212 ce. The second edition was actually an unauthorized release (Gamble 1995: 119;Larsen 2017: 375). Particularly relevant to the discussion here is the manner in which portions of the second edition were stolen, incorrectly copied (perhaps maliciously) and then circulated without Tertullian's consent. The identity of the thief must have been known, for Tertullian gave unique details of the copyist's later apostasy and he knew which sections of the composition were stolen and the extent of their corruption. Christian authors were also victims of the deliberate alteration of their writings. Origen of Alexandria, in a letter to his friends in Alexandria, recounts an instance in which a debate between himself and a theological opponent was copied down: 44 For a certain author of a heresy, when a discussion was held between us in the presence of many persons and was recorded, took the document from those who had  Faye 1923Faye , 1927Faye , 1928 conspicuously on the things that he himself had composed. The brethren who are in Palestine were indignant over this. They sent a man to me at Athens who was to receive from me the authentic copy. Prior to this I had not even re-read or revised the work, but it was lying there in such a neglected state that it could hardly be found. In this occurrence of outright textual change, it was Origen's acquaintances in Palestine who notified him of the corruption. An authentic copy was obtained from him and the additions were then brought to light. Origen goes on to describe how the same opponent, too timid to face him in a discussion, instead forged the text of a phantom debate between Origen and himself. Though he does not mention specifically how he came to learn of the forgery, Origen states that the composition 'reached a large number of our own people' in the city of Antioch (De adult. libr. 7). He continues: But when I came there in person, I convicted him in the presence of many; and, when without any shame he persisted in the impudent defense of his forgery, I demanded that the book be brought out in public, so that my style would be recognized by the brethren, who of course knew the things which I customarily discuss, and the kind of teaching I employ. When he did not dare to produce the work, he was convicted by everyone of forgery and was silenced. And thus the brethren were convinced not to give ear to the accusations.
Sed ubi adfui, multis eum praesentibus argui; cumque iam sine ullo pudore pertenderet impudenter adserere falsitatem, poposci ut liber deferretur in medium, (ut) Scheck 2010: 129, Latin Amacker andJunod 2002: 304) In this account there is a direct correlation between the wide circulation of the forgery and its exposure. The community of readers and followers of Origen participated in evaluating the content of the forgery and the changes made to his doctrine and teaching style.

New Testament Writings
The discussion will now turn to a few places in the NT writings which suggest that macro-level textual alteration and plagiarizing to the writings which were already in circulation was revealed within these same Christian communities. In 2 Thess. 2.1-2, Paul warns his readers 'not to be quickly shaken in mind or alarmed, either by a spirit or a spoken word, or a letter seeming to be from us …' 46 Here Paul expresses concern over the possibility that someone could pen a letter and falsely attribute it to him or another apostle. 47 It is impossible to know whether this had occurred or not, but Paul was able to send a letter to the Thessalonians warning them of this possibility. 48 Conversely, Paul could have learned whether the Thessalonians had received such a letter through his emissary Timothy. In 1 Thess. 3.1-6 we learn that it was Timothy who brought word concerning the status of the Thessalonians' faith. As a result, a falsely attributed letter and teaching that contradicted that of Paul would have been exposed through the same networks that circulated his letters. 49 46. εἰς τὸ μὴ ταχέως σαλευθῆναι ὑμᾶς ἀπὸ τοῦ νοὸς μηδὲ θροεῖσθαι, μήτε διὰ πνεύματος μήτε διὰ λόγου μήτε δι' ἐπιστολῆς ὡς δι' ἡμῶν. 47. Though there are some scholars that do not consider 2 Thessalonians as genuinely Pauline, it appears that the growing consensus is in favor of Pauline authorship; see Foster 2012. Despite the authorship issues, the passage remains as a reflection of Greco-Roman praxis. 48. The very thesis of this article works against the concept of pseudepigraphical apostolic or authoritative works circulating within the early Christian communities as accepted authentic writings. It would be difficult to widely circulate a letter falsely attributed to, say, Paul during his life or soon after his death and it be accepted by the Christian community as authentic. See Richards 2004: 146 for an example of a forged letter from Caesar to Cicero which was exposed by him. There is another example from Tertullian, De bapt. 17.4-5, where the author of a writing falsely attributed to Paul (The Acts of Paul) was discovered and the falsehood exposed (Metzger 1987: 160;Ehrman 2013b: 379-80). Despite this, scholars note that in the Christian era pseudepigraphal writing was not always seen as simply false attribution but also as a practice of devotional imitation, Stang 2012: 51-80. For a similar view in a Jewish milieu, see Morczek 2016. 49. In an earlier portion of the letter to his Alexandrian friends quoted above, Origen references 2 Thess. 2.1-2 as an example of pseudepigraphy (De adult. libr. 7). For a discussion on the various roles of letters carriers, see Head 2009;Epp 1991. Next, consider the implications of 2 Tim. 1.13-14 where the author wrote: Follow the pattern of the sound words that you have heard from me, in the faith and love that are in Christ Jesus. By the Holy Spirit who dwells within us, guard the good deposit entrusted to you.
Of course, this seems to be referencing the guarding of Paul's (or the Pauline school's) preaching and teaching rather than guarding against textual corruption. Yet, in 1 Tim. 5.18 the author makes a reference to Deuteronomy and possibly Luke as scripture, and 2 Tim. 3.16 references scripture as well. Thus, it would not be a stretch for the readers of 2 Timothy to apply this exhortation to preserve the text of the authoritative writings they had received (Kruger 2012a: 67.).
Concerning the corruption of the Pauline collection, the author of 2 Peter wrote: And count the patience of our Lord as salvation, just as our beloved brother Paul also wrote to you according to the wisdom given him, as he does in all his letters when he speaks in them of these matters. There are some things in them that are hard to understand, which the ignorant and unstable twist to their own destruction, as they do the other Scriptures.
(2 Pet. 3. [15][16] The author is referring to a body of Paul's letters that were considered to be scripture (Kruger 2012b: 204-205;Metzger 1987: 259). He was declaring that he knew of those who were twisting Paul's writings. Now, it is most likely that the author was referring to the interpretational or exegetical twisting of Paul's teaching. As was the case with Origen above (De adult. libr. 7), it could be that doctrinal tampering would lead to textual tampering. To be sure, 2 Peter does not mention the means by which this knowledge was obtained; it was likely learned through the same networks that circulated Paul's letters. Also note that at 2 Pet. 1.12, 3.1 and 3.17 the author himself functioned as the avenue by which others within the community gained knowledge of the distortion of Paul's epistles.

Community as a Textual Stabilizer
This selection of Greco-Roman and Christian authors indicates a larger trend: that the corruption and misappropriation of a text already in circulation would likely become exposed within the same circles that were transmitting the text. As discussed above, the methods used for circulating Christian scripture were similar to that used in the wider Greco-Roman culture. As a result, a similar occurrence of exposing the corruption of Christian scripture is already hinted at within the writings of the NT. With the rise of differing theological perspectives in the following centuries came the potential for the textual alteration of the NT. In the second and early third centuries, there were a few instances of deliberate textual tampering of the Christian scriptures that contemporary writers were discussing.

The Theodotians
Eusebius preserves an account in Hist. eccl. 5 of Asclepiodotus and Theodotus, who were followers of Theodotus the Cobbler, who denied the deity of Jesus. This occurred when Zephyrinus was Bishop of Rome at the end of the second and into the first years of the third century. Not only were Asclepiodotus and Theodotus accused of leading the faithful astray, they were charged with altering and 'correcting scripture' in order to support their divergent Christology. As evidence for this accusation, Eusebius's source points to the manuscript tradition: if any are willing to collect and compare with each other the texts of each of them, he would find them in great discord, for the copies of Asclepiades do not agree with those of Theodotius … the copies of Hermophilus do not agree with these, the copies of Apolloniades are not even consistent with themselves, for the copies prepared by them at first can be compared with those which later underwent a second corruption, and they will be found to disagree greatly.

Marcion
One of the most famous instances of the textual corruption of NT writings can be found in the character of Marcion of Sinope, who flourished in the first half of the second century (Foster 2010). He is most infamous in the history of the church for his alleged alteration of Paul's epistles and the Gospel of Luke (Roth 2015). Tertullian wrote concerning Marcion: Corruption of the Scriptures and of their interpretation is to be expected wherever difference in doctrine is discovered … Marcion openly and nakedly used the knife, not the pen, massacring Scripture to suit his own material. For if the apostolic gospels have come down to us in their integrity, while the gospel of Luke, in the form in which we have it, is in such agreement with the standard of those others that it is retained in the churches along with them, it is at once evident that Luke's also came down in integrity until Marcion's act of sacrilege. In fact it was only when Marcion laid hands upon it, that it became different from the apostolic gospels, and in opposition to them.
Si enim apostolica integre decucurrerunt, Lucae autem, quod est secundum nos, adeo congruity regulae eorum ut cum illis apud ecclesias maneat, iam et Lucae constat integrum decucurrisse usque ad sacrilegium Marcionis. Denique ubi manus illi Marcion intulit, tunc diversum et aemulum factum est apostolicis. (Marc. 4.5) (Evans 1972: 272-73) The specific contours of Marcion's textual alteration, whether he received an already abbreviated version of Luke, or edited it himself, has been hotly debated in recent years (Roth 2008). Without becoming entangled in the debate, it is instructive to note Tertullian's reference to the copies of the gospels already in 52. English translation is from Greenslade 2006: 58-59, Latin text is from Refoulé and de Labriolle 1957. circulation among the churches as standards by which to compare Marcion's text of Luke. Tertullian was pointing to the community which circulated Luke's gospel as the standard by which Marcion's text was to be judged. This falls in line with Martial's comment that 'A well-known book cannot change author ' (Epigr. 1.66). 53 In other words, because Luke had been circulated and was well known among the churches, Marcion could not easily alter its form. 54 Other examples of documented textual change can be cited as well; for instance, Irenaeus of Lugdunum appealed to the oldest and best manuscripts as well as the wider community of those who knew John (the one believed to be the author of Revelation) in order to support the reading 666 in Rev. 13.18 (Haer. 5.30.1) (Donaldson 2009: 94-96). 55 Once a manuscript was circulating, it was impossible to make any changes to the text, and this worked in both directions. Any who wished to tamper with the text could only do so in their own copies, thus, there would be significant differences between these tampered copies and the copies already in circulation. Those who noticed these textual changes could only raise the alarm and had no way to enforce quality controls. Nevertheless, this inadvertently created an environment of transparency and textual stability. 56 After an extensive study of communal reading practices of Greco-Roman, Jewish and Christian communities in the first century, Brian J. Wright observed a similar phenomenon. He noted that 'the overall reading culture creates the expectation that these events were a viable quality control of the Christian tradition'. 57 Though this article examines the process of ancient publication in its entirety, the reading out of a text to a group represented a form of publication, 53. 'mutare dominum non potest liber notus.' 54. In support of the view that Marcion edited Luke to form his gospel, see Moll 2010: 89-102. After examining the Oxyrhynchus papyri for the time and distance it took to send and receive letters and documents, Eldon Epp concluded that the 'text-types' of the NT writings (contained in physical manuscripts) could 'in a matter of a few weeks, have moved anywhere in the Mediterranean area ' (1991: 55-56 and thus his findings undergird the arguments presented here. Wright does not suggest that this early Christian quality control governed the 'text' of these traditions, only the teachings. Yet, as has already been discussed above, the safekeeping of traditions often involved guarding the texts of those traditions. 58 In addition, Wright has documented the widespread economic and political stability of the Roman Imperial age (Wright 2017: 23-38). This stability led to technological advances as well as increased infrastructure, such as roads, that resulted in a high level of travel and mobility. 59 A thorough treatment of these topics is beyond the scope of this article; however, at the risk of oversimplifying complex issues, this stability would foster fluidity of boundaries and interchange between various communities (Thompson 1998). Church leaders had contact with Jewish communities who also had access to Christian writings and scriptures. 60 It is apparent that a similar phenomenon occurred with regard to Christians and pagan philosophers. 61 Though economic, geographic, social and cultural differences certainly affected the interchange of ideas, Christian writings appear to have circulated widely across the Mediterranean despite these sociocultural and socio-economic barriers.

Implications for Textual Criticism
The unauthorized circulation of Cicero's speech (Att. 3.12), and the premature release of his De finibus (Att. 13.21a) reveal that once his works were copied beyond his trusted circle of friends, these same works were inadvertently 58. Earlier, Wright does note that '[t]he author of Mark's Gospel highlights that audiences often seemed prepared to critically evaluate someone who read, recited or mentioned texts communally -especially sacred texts -and how they responded in various ways … Mark reveals that the teacher was assuming the audience had already heard the text(s) he read, recited, or mentioned, as in 10.19 ("you know the commandments"). The assumption also suggests that whatever copy of the Scriptures they have read or heard is similar and stable enough to permit such a statement, and it assumes they know it well enough to recall it either from reading or hearing it enough times. Moreover, depending on the familiarity the teacher expected the audience to have with the text(s), there is necessarily a correlation between the number of times an audience would have had to hear the particular text(s) read aloud for such familiarity -especially if illiteracy was pervasive ' (2017: 123 Kruger 2018: 63-64); Justin Martyr and Tatian (Justin's student) were philosophers who had access to Christian thinking and writings and were converted by the 'philosophy' of Christianity (Kruger 2018: 66-69). The second-century philosopher and critic of Christianity Celsus had access to Christian writings, read them and engaged thoroughly with Christian theology (Hill 2010: 155-56). See also Mills 2019 and Hoffmann 1987: 35-37. finalized. In the case of his De finibus it was the same acquaintances that circulated and read his work, Balbus and Atticus, who informed Cicero of the accidental publication. Both the sanctioned (Inst. Or. 1. Pr. 7-8) and the unapproved release of Quintilian's lectures (Inst. Or. 7.2.24) show that, once his orations left the author's circle and began to make the rounds, Quintilian could no longer make changes to the text and thus these texts were involuntarily finished. It was necessary for Quintilian to write a completely revised edition in which he felt necessary to inform his readers that he considered his earlier crudely copied lectures as inferior and should be regarded as a distinct work. A similar attitude can be seen in the work of Tertullian, Adversus Marcionem, in which he clearly distinguished between the text he authored and those additions and corruptions made by the thief responsible for the premature release of portions of his polemic (Marc. 1.1-2). This presses hard against Larsen's assertions: Unless we can determine that a text was finished, closed and published, which would exclude many of the texts now called the New Testament, traditionally conceived modes of textual criticism may be a square peg for a round hole (Larsen 2017: 379-80).
Once a text began to circulate, whether sanctioned by the author or released without their consent, practically speaking, a text became finished at this point. Authorized changes were noted by the ancients in the opening remarks of a newly released edition, and readers notified the community when corruptions and plagiarizing occurred. In essence, modern textual criticism continues this ancient practice of carefully mapping the history of the NT text in stages from authors, to scribes, to readers. Therefore, the moment at which a manuscript began to circulate was the definitive starting point for the textual critic to begin their work. Though the modern textual critic may have difficulty discerning between the authorial text and later additions and corruptions made to the text after the document began to circulate, this does not mean that the ancients did not make these distinctions or that the modern textual critic should not. The circumstances surrounding the circulation of Galen's medical notes, commentaries and informal speech transcriptions tell a similar story (De libr. propr. 19.10). Once his writings began to circulate beyond his circle, he had no say over the manner in which they would be used or the quality of the text being copied. Galen's circle of friends and students recognized that these additions and corruptions were not by the hand of Galen, clearly distinguishing between his authored text and the additional text. This challenges Larsen's claim that, with the writing of notes, '[o]penness is left in the text for another to rework them and attach their name to them to "author"' (Larsen 2018: 36). Galen never intended for his notes to be an open textual tradition, and his students and readers assisted Galen in locating these 'reworked' notes for him to correct. This same attitude can be seen in the writers already observed above.
With regard to the NT writings, the passages already mentioned, 2 Thess. 2.1-2, 2 Tim. 1.13-14 and 2 Pet. 3.15-16, hint at similar forces in play. As the account concerning Marcion and the Theodotians reveal, these same factors continued to operate into the second and third centuries, the period from when the earliest fragments of the NT writings were preserved. Thus, when compared with the views of Parker, Larsen, Koester, Petersen and others, Holmes's assertions become more probable, that 'all the variation during the time period in view affects a verse or less of the text' indicating that the text of the NT writings is 'characterized by macro-level stability and micro-level fluidity' (Holmes 2011: 78). Galen may give some insight into the level of textual change that was acceptable to the scribes and readers of the first and second centuries ce. In the midst of critiquing the work of previous Hippocratic scholars who were active in the first part of the second century Galen wrote that, A second book written in place of one formerly written is said to be revised when it has the same hypothesis and most of the same words; some (of the words) taken out from the former work; some added; some altered.
Ἐπιδιεσκευάσθαι λέγεται βιβλίον ἐπὶ τῷ προτέρῳ γεγραμμενῳ τὸ δεύτερον γραφέν, ὁταν τὴν ὑπόθεσιν ἐχον τὴν αὐτὴν καὶ τὰς πλείστας τῶν ῥήσεων τὰς αὐτὰς τινὰ μὲν ἀφῃρημένα τῶν ἐκ τοῦ προτέρου συγγράμματος ἐχῃ, τινὰ δὲ προσκείμενα, τινὰ δ ̓ ̔ ὑπηλλαγμενα· (Hipp. vict. acut. 120.5-14) 62 Galen is speaking specifically in reference to editing an ancient author's work in order to remove textual corruptions. Yet, it is clear that Galen saw that a composition was still considered to be that of the ancient author's as long as 'most of the same words' were still from the original writer and that the editing process did not alter the 'hypothesis' of the original work. This would allow for some 'micro-level fluidity' during the copying process when scribes 'added' or 'altered' some of the words yet retained 'macro-level stability' at the 'hypothesis' level. However, there is no way to know for sure if Galen's comments are representative of broader attitudes.

Conclusion
Though these and other authors did perform post-publication revision and released new versions and editions of their works, the authors themselves and the communities that distributed them drew lines of distinction between the different versions in circulation. Though the readers of these works may have had difficulty discerning between the various editions being transmitted, the community 62. English translation is from Scherbenske 2013: 39, Greek text is from Helmreich 1914: 120. of authors and readers worked together to make sense of the text in circulation. The author informed the readers when new additions or corrections were being made, and the readers informed the authors and other readers when these texts where being corrupted or plagiarized. 63 With this phenomenon in view, because the NT writings were transmitted and circulated through Greco-Roman conventions of community distribution, this naturally produced a condition in which the plagiarizing and 'macro-level' alteration of these writings would have been exposed within these same community circles. 64