The Birth of the Book (Part 2)

From Issue: R&R – February 2022

[Editor’s Note: This article is the second installment in a three-part series pertaining to the integrity of the biblical text through the centuries. AP auxiliary writer Dr. Rogers serves as an Associate Professor of Bible at Freed-Hardeman University. He holds an M.A. in New Testament from FHU as well as an M.Phil. and Ph.D. in Hebraic, Judaic, and Cognate Studies from Hebrew Union College-Jewish Institute of Religion.]

Long before Johannes Gutenberg (1400-1468) introduced movable type and printed the Bible (a copy of the Latin Vulgate) in the 1450s, the biblical text had already been copied by hand for centuries. The word “manuscript” itself means “hand-written” and reflects a reality that many Christians—especially after the Printing Revolution of the 15th century and the 19th-century Industrial Revolution—have forgotten, namely, that the Bible was hand copied until recent history.1 But what was this copying process like? What can one know about the scribes who copied the Old and New Testaments? How good a job did they do in copying the text? What kinds of variations in copying are evident from the manuscript tradition? Has the text of Scripture been faithfully and accurately copied through the years? There is great debate over these questions, thus in this article attention will be given to each question, especially with regard to both testaments and each historical period.

The Copying Process: A Brief Historical Overview

Before the actual copying of the text could occur, one would had to have first found a suitable exemplar (the manuscript from which the copy was being made). Interestingly, one of the more recent developments in the study of biblical manuscripts is the development of stemmata (family trees), which illustrate the relationships that exist between manuscripts and their ancestors, which might even include their exemplars.2 Then papyrus, parchment, or paper materials, the ink, and writing instrument(s) had to either be made or purchased. Once the scroll or codex was assembled and prepared for writing, the scribe would begin to copy. In preparation the selected material would be ruled both horizontally and vertically to designate rows and columns.3

The Renaissance-like setting that many have imagined of scribes sitting in a scriptorium with good light, nice desks, fresh ink with quill pens, and constant efforts to correct the text is not an accurate picture of the conditions in which many of the scribes labored. While transmission eventually leads to standardization (as evidenced by the Vulgate [4th cent.] in Latin, or Peshitta [5th cent.] in Syriac), the transmission of the biblical text was not generally standardized until a later period. For the Old Testament this would have been around A.D. 500 with the Masoretes and, for the New Testament later, with the development of the Byzantine text tradition around A.D. 700-800.4 Historically, the standardization of the text in a New Testament setting would have also been made difficult before the early 4th century because of the status of Christianity as illegal, which the emperor Constantine (A.D. 274-337) would change.

The variety of exemplars and scribes is evident even in the manuscripts themselves, given the differences between the illumination of manuscripts (the way they were decorated), calligraphy (handwriting styles), marginal glossa (notes that became much more frequent in the Middle Ages), ligatures (the joining of two or more letters into a single sign), and other paleographic figures. Furthermore, the size of manuscripts varied from personal amulet-like copies of Scripture to large manuscripts like the 12th-century Codex Gigas (36″ long x 20″ wide x 8.7″ thick).

Picture of Codex Gigas with Illumination of Devil on One of Its Folios. Source: Wikimedia Commons (Creative Commons license)

With the variation of scribes, exemplars, skill levels, and contexts, unintentional errors sometimes resulted from a misreading of the text of the exemplar. In both the Old and New Testament text traditions, errors of the eye seem to be more common than errors from mishearing the text being read aloud. With all of these variables, those interested in the transmission of the text of Scripture would be well served to speak of the characteristics of particular manuscripts, scribes, and correctors, rather than assuming universal qualities that all of these facets of the transmission of the text shared. In both the Old and New Testaments, the earlier the period, the less one can know definitively about scribes or the copying process.

The vastness of the time and the sheer volume of material is much greater for the Old Testament than the New. Beginning with the first Old Testament books to be written, the text was hand-copied for nearly 3,000 years.5 Scribal interest likely arose after the Babylonian exile, in the days of Ezra and Nehemiah, when an emphasis on the Law led to a need to know and teach the Law. While it is possible that “the (Jewish) scribes” referenced in the New Testament arose out of the leadership and influence of these sopherim (scribes), there is little textual evidence for the Old Testament before 300 B.C. From 300 B.C. to A.D. 135, there is extant manuscript evidence, including biblical manuscripts from Qumran. It is in this time period that those examining the manuscripts themselves can begin to see evidence of varying text families.6

While some manuscripts align in style with the later standardized Masoretic text, not every manuscript fits this style. Likely, there were concurrent texts at this time of Samaritan and Babylonian origin that were still read and revered in certain settings. Most significantly, the Old Testament text was divided into paragraphs and verses early in this period, which along with other adaptations aided readers in using the Old Testament text in liturgical settings. It was later during the textual activity of the Masoretes (5th-9th centuries A.D.) that the Old Testament text became more standardized, including the development of written vowel pointing, written symbols for pronunciation, and other specialized notes. Scribes sought to preserve the text as evident through the changes made. The period from A.D. 1000-1450 was basically a time to preserve and maintain Masoretic readings.7 All printed editions of the Old Testament contain the Masoretic text, though it differs from the Samaritan Pentateuch and many other texts from the Judean desert. 

While the history of the text of the New Testament is not as long, it does follow a similar pattern. Not much is known about scribal practices among the first Christians—given the lack of many extant witnesses from the 1st and 2nd centuries (there are a few fragmentary witnesses like p52 dating back to A.D. 125-150). But beginning in the 3rd century, the nature of textual transmission becomes much clearer. From the time the New Testament writings were first produced, until the time of the “conversion” of the emperor Constantine (around A.D. 325), the text was freely copied in a number of diverse ways. This diversity was due in large part to the varieties of copying processes, scribes, materials, exemplars, and perhaps even the emerging recognition of Scripture by Christians throughout the Mediterranean world. Most variant readings (a place where manuscripts present at least two options for a reading in a given text) in the New Testament come about in this earliest period because of a lack of a professional copy process. Furthermore, oral tradition was still strong and, as late as the middle of the second century, some still preferred the oral tradition in a setting only one generation removed from an apostle. Certain scribal traits were already being developed at this time which one can see in the Alexandrian text that took shape around the end of the second century in Egypt, as evident in the text of p4 (A.D. 200), p75 (A.D. 200), and Codex Vaticanus (A.D. 325-350). Professional scribal activity arose after Christianity was legalized, while most copies of Scripture in this earliest period were transmitted through the hands of scribes who wanted to copy Scripture as carefully as possible.

After Christianity was legalized, there is a period in which the text of the New Testament text began to converge from A.D. 325-700. The emperor Constantine unified the Roman empire politically through his move towards Christianity, but also ushered in a period where many religious structures were built, aspects of the faith were discussed in open council meetings, and Christians found themselves benefitting from more political freedoms than ever before. One illustration of the benefits of this Constantinian shift comes when the emperor ordered 50 copies of Scripture, which resulted in a limited number of standard copies for urban churches throughout the empire.

As noted earlier, at this same time translations of the New Testament in other languages began to be standardized as with Jerome’s Vulgate and the Peshitta for the Syriac versions. The Greek manuscripts did not consistently carry a text tradition form of the New Testament text at this point, though textual traditions like the Alexandrian, Western, and Byzantine began to be more pronounced at this time. While a lot of evidence points to Egypt as the primary consolidation point (namely, through the influence of Alexandria), John Chrysostom (A.D. 349-407) is a major figure in pushing the Byzantine text to the forefront, which became the standard form by A.D. 700-800. The influence of the Byzantine text tradition and the natural movement towards standardization really defines the history of the transmission of the New Testament text from A.D. 700-1500.

The Scribes

In greater Greco-Roman society the task of copying texts was not an honorable one.8 There was a wide range of scribal training and experience in the New Testament world. Literary copyists worked for the book trade in producing, reproducing, and disseminating the texts. Bookstores would keep exemplars on hand and literary scribes would copy them as requested. While ancient libraries employed these literary copyists, private copyists were also prevalent as personal libraries were frequently a way of demonstrating one’s social status. Private scribes were contracted by those who could not read or write. Public administrative scribes maintained official archives, oversaw financial and agricultural workings, religious temple documents, and legal documents. Some scribes were not highly literate. Thus, a distinction existed between those scribes who could simply copy as compared to those who could actually compose. Some scribes were multifunctional and could take dictation, edit, and keep copies of letters. Nonprofessional copies were also common as they were made by those who used the text themselves. In the 4th century, the evidence for the copying and dissemination of texts increased dramatically. There was a notable shift as scribes went from being viewed as low class to highly spiritual for displaying their religious stature. It is highly possible that scribes of all kinds of training converted to Christianity and used their skills to copy the biblical text.

To summarize, scribes differ in their abilities and tendencies as they represent a diverse group, ranging from professionals, who were paid by the line for what was copied (stichometry was the practice of counting lines in the text to measure the length of a book and sometimes to calculate the payment for a scribe), to novices who wanted to copy the text of Scripture but were barely literate enough to copy it—much less read the text. Gamble recommended that in the tumultuous pre-Constantine setting of the church, manuscript production in Christian settings was not about profit, but rather about simply distributing writings that were readable and usable. He also recommended—by means of internal evidence from the letters of Paul, the Gospel accounts, and Revelation—that Christian materials were widely distributed through private channels as well.9

Female scribes were also present in antiquity. While female copyists most often worked for female masters, they were more than mere secretarial help. Eleven Latin inscriptions from Rome identify women as scribes.10 While some female scribes were free and some slaves, it seems that all of these scribes primarily worked in urban areas. In early Christianity, the scribe Melania the Younger (A.D. 383-439) was raised as a scribe in a monastic setting. An Arabic note ascribes the copying of Codex Alexandrinus to a certain female scribe named Thecla.11 Nowhere in the ancient traditions is the possibility of a female scribe questioned, though there was generally a gender separation with males working for males and females working for females.

The scribes of the Old and New Testaments were predominantly those who believed they were copying holy words and sought to give careful attention to the sacred task before them. Their familiarity with the text is evident based on their tendency to harmonize the text not only to the immediate context, but also to parallel contexts. They desired to copy the text for accuracy and readability. Human error was inevitable (as with parablepsis “eye jumps”) but, in general, Old and New Testament scribes were trustworthy in the task set before them.

Westcott and Hort recognized the introduction into the text of accidental, or “clerical,” errors by scribes even when transcribers were attempting to copy accurately the text. In their discussion of the value of internal evidence for evaluating manuscripts, they argued that a knowledge of the manuscripts themselves (based on external and internal criteria) would provide a “sure foundation” for determining the “original” reading. Westcott and Hort also suggested that sometimes manuscripts are affected “by the blunders of a careless scribe,” but one must be sure to evaluate scribal traits as associated with particular manuscripts, rather than ascribing traits to “scribes as a class.”12 Readers should be thankful for scribal devotion to the task and the apparent success with which they handled the Word of God. And we must also avoid the tendency to describe these diverse scribes, who copied manuscripts of varying texts in different historical contexts, as all being the same.

An Overview of the Types of Variations

The primary goal for God-fearing scribes was to copy the biblical text accurately, but none of them did so perfectly. Their number one problem was that they were human. Old Testament scribes sometimes accidentally erred because of the confusion that resulted from the similarities between Paleo-Hebrew letters, or with the square letters that eventually replaced them.13 In the Hebrew, the consonantal nature (the lack of vowels) of the language likely led to variations, since divisions between words in the text were sometimes not easily discerned. Similar sounding vowels led to many “errors of ear” (orthographical shifts) in the copying of the New Testament text as well. Of course, illegible handwriting, colophons (omitted text), and damage to exemplars could have also contributed to changes in the Old and New Testament texts. Tov correctly noted that these types of variations are evident in both the proto-Masoretic and Masoretic text traditions of the Old Testament, while New Testament textual critics observe the same pattern in manuscripts of all text traditions from the 3rd to 16th centuries. 14

Scribes of the Old and New Testaments sometimes accidentally omitted material (minuses), added material (pluses), confused letters for one another, while also unintentionally missing word divisions, vowels, or abbreviations. These same scribes sometimes committed homoeoteleuton and homoeoarkton (commonly called parablepsis), which occurred when the identical ending or beginning of words caused material to be skipped due to an eye-jump. Similarly, scribes occasionally were guilty of haplography (“writing once”) which omitted neighboring words or letters that were similar, or dittography (“writing twice”) which doubled letter(s) or word(s) in the text that should have only been written once, while also transposing words on occasion which reversed or rearranged words in a clause or phrase. Sometimes scribes created “doublets” or “harmonization” because, in their familiarity with the text, they conflated more than one reading.

Are there occasions when scribes intentionally altered (emended) the text? Yes, but rarely if ever with malicious intent. Bart Ehrman famously challenged the historicity of the New Testament because of a tendency of scribes to expand divine names out of reverence (for example, from “Jesus” to “Lord Jesus Christ”). Ehrman argued that on occasion scribes made changes to the text in order to make readings more orthodox so that the text would be more difficult to use by Christians with differing perspectives.15 On other occasions scribes attempted to “correct” difficult readings as with the difficult expression “unique God” (μονογενὴς θεὸς) in John 1:18, which many scribes changed to the more familiar “unique Son” (μονογενὴς υἱὸς). Ehrman used Walter Bauer’s (1877-1960) perspective on the history of early Christianity to argue that textual emendations before the 4th century resulted from an opposition to teachings labeled as heretical by the “winners” of these theological battles.16 Ehrman argued many variants in the text that resulted from an intentional “corruption” introduced by proto-orthodox scribes. Most of the “corruptions” that Ehrman noted, however, have little manuscript support.

In the copying of the New Testament text, Dan Wallace estimated that there are 300,000 to 400,000 variations existing in a testament that only has 140,000 words total.17 The sheer number of variants can be discouraging until one considers the nature of the variants. If one focuses on the number of variants, without also considering the number of variables, this number can be misleading. When the number of manuscripts and text traditions, the diverse training of scribes, and other contextual factors are taken into account, it is obvious that the number of textual variants will naturally increase based on the number of manuscripts and scribes. Furthermore, it is comforting to consider that no human organization oversaw or had control of the copying process, which might have lessened the number of variants but also could have prevented the autograph texts from being represented more fully in the manuscripts themselves.18


As noted above, in recent years, Bart Ehrman and others have begun to suggest that these copies of the text indicate that the process of copying Scripture points to both a lack of standard text and set of scribal controls in the earliest period of Christian history. Though one cannot know everything about the copying process of the Old and New Testaments, through the careful evaluation and comparison of manuscripts, the text of Scripture can be established. The copying process was diverse depending on the qualifications of scribes and the qualities of their exemplars, but there is no evidence of a widespread effort to “corrupt” the biblical text. Most scribal errors were accidental or intended to “correct” a reading that could be misunderstood because of thematic, grammatical, or theological difficulties. The scribes who copied the text of the Old and New Testaments were not perfect, but the texts they were handling reflected perfect autographs. In other words, most of these scribes believed that they were copying a holy text that communicated God’s will for those who would read the text of the scrolls or codices or hear it read. When one weighs the evidence, evaluating the types of textual variations that have been introduced into the text as it has been transmitted, it is difficult not to see that the text has been faithfully transmitted. Are there differences between biblical manuscripts? Yes, but these differences have resulted from the work of human scribes, not the God who inspired the autograph text when it was first written.

The reader is urged to sit down sometime and try to write out by hand a biblical book, a group of books, or even a testament. Sixty-five members of the Concord Rd. church of Christ in Brentwood, TN recently copied the New Testament by hand. In the preface to their work (“The 260 Project,” which began in 2008), it was acknowledged how copying the text of Scripture helped them to read, review, and remember the biblical text, while also developing a greater appreciation for those who sacrificed so much to copy the Word of God by hand. Let us never forget the simple blessings of having a copy of the Bible that we did not have to write out—in our own language—that we can read. The transmission of the text shows that the demand for the Word was high even when the supply was low. Some apologists for world religions have used textual variants as an opportunity to attack the credibility of the Bible, while having destroyed the discordant copies of the books they value to avoid the same critique. Yet, these copies of Scripture are an opportunity to praise God for His faithfulness, while acknowledging the tireless efforts and sacrifices of so many to transmit the Word of God faithfully from generation to generation.


1 Though handwritten manuscripts drastically slowed down in production with the printing press, a few manuscripts from the 16th and 17th centuries exist as well.

2 In his work The Profile Method for Classifying and Evaluating Manuscript Evidence (Studies and Documents) [(1982), ed. Irving Sparks (Grand Rapids, MI: Eerdmans), Vol. 44], Frederick Wisse argued that when several hundred witnesses of a text exist, the best way to group them is to select test passages, select significant test readings from those test passages, collate the test readings against a Majority Text base, classify them according to agreement, and choose representative texts to be represented in critical apparatuses. Gerd Mink’s Coherence-Based Genealogical Method (CBGM) was generated in an effort to improve a stemmatic understanding of the genealogical tradition. In response to the circular nature of the witness-variant discussion, adherents of this method seek to establish a hypothetical reconstruction of an “initial text” (Ausgangstext) by which textual influences in certain text streams are evaluated. For more on the CBGM, see Gerd Mink, “The Coherence-Based Genealogical Method—What Is It All About?” University of Münster Institute for New Testament Textual Research, Genealogical_method.html.

3 Emanuel Tov (2001), Textual Criticism of the Hebrew Bible (Minneapolis, MN: Fortress Press, second revised edition), p. 206.

4 Porter and Pitts state that there is no evidence of the Byzantine text tradition nor citations of the Byzantine text by church fathers earlier than the fourth century. See Stanley E. Porter and Andrew W. Pitts (2015), Fundamentals of New Testament Textual Criticism (Grand Rapids, MI: Eerdmans), p. 78.

5 Ellis Brotzman (1994), Old Testament Textual Criticism: A Practical Introduction (Grand Rapids, MI: Baker), p. 38.

6 Brotzman, p. 43.

7 One should note that the Masoretic text tradition (which was solidified as a single text-tradition in the Middle Ages), much like the Byzantine text tradition in the Greek New Testament, is not uniform. Thus, to argue that this is the original text tradition that should be used to establish the biblical text, one must first decide which readings of the Masoretic or Byzantine texts are to be preferred. For more on this reality, see Tov’s discussion of the “editions” of the Masoretic text on pp. 22-49,77-79.

8 Kim Haines-Eitzen (2000), Guardians of Letters: Literacy, Power, and the Transmitters of Early Christian Literature (New York: Oxford University Press), pp. 22-23.

9 Harry Y. Gamble (1995), Books and Readers in the Early Church: A History of Early Christian Texts (New Haven: Yale University Press), pp. 140-143.

10 Haines-Eitzen, pp. 41-52.

11 Garrick Allen (2016), “The Apocalypse in Codex Alexandrinus: Exegetical Reasoning and Singular Readings in New Testament Greek Manuscripts,” Journal of Biblical Literature, 135:869, n. 39.

12 Brooke Foss Westcott and Fenton John Anthony Hort (1882), Introduction to the New Testament in Original Greek (Peabody, MA.: Hendrickson, 1988 reprint), pp. 230-50. See especially pp. 231-32.

13 Tov, p. 8; Brotzman, p. 39.

14 Tov, p. 9. For more on global patterns of scribal traits in New Testament manuscripts, see Doug Burleson (2012), “Case Studies in Closely Related Manuscripts for Determining Scribal Traits.” PhD Diss. New Orleans, LΑ: New Orleans Baptist Theological Seminary.

15 Bart Ehrman (1993), The Orthodox Corruption of Scripture (New York: Oxford University Press), p. xi.

16 Walter Bauer (1971), Orthodoxy and Heresy in Earliest Christianity, ed. Robert Craft and Gerhard Krudel (Philadelphia, PA: Fortress).

17 Daniel B. Wallace (2011), Revisiting the Corruption of the New Testament: Manuscript, Patristic, and Apocryphal Evidence (Grand Rapids, MI: Kregel), p. 26.

18 Wallace, p. 39.


A copied sheet of paper

REPRODUCTION & DISCLAIMERS: We are happy to grant permission for this article to be reproduced in part or in its entirety, as long as our stipulations are observed.

Reproduction Stipulations→