Before we discuss the various theories and methodologies of New Testament textual criticism, it is important to recognize that nearly all textual critics first begin their task by identifying transcriptional errors. These are scribal errors caused by faulty copying. These are identified by the following names: dittography, haplography (or, scribal leap), homoioarchton (or homoeoarchton), homoioteleuton (or homoeoteleuton), and transposition. 

Homeoteleuton, also spelled homoeoteleuton and homoioteleuton, is a form of copyist error in ancient texts. A scribe would be writing out a new copy of a frequently reproduced book, such as the Bible. As the scribe was reading the original text, his eyes would skip from one word to the same word on a later line, leaving out a line or two in the transcription. When transcripts were made of the scribe’s flawed copy (and not the original) errors are passed on into posterity.

Haplography (from Greek: haplo- ‘single’ + -graphy ‘writing’), also known as lipography, is a scribal error where a letter or group of letters that should be written twice is written once. It is not to be confused with haplology, where a phoneme is omitted to prevent two similar sounds from occurring consecutively: the former is a textual error, while the latter is a phonological process.  It refers to a scribe’s, copyist’s, or translator’s inadvertently skipping from one word or phrase to a similar word or phrase further on in the text, and omitting everything in between. It is considered to be a form of parablepsis [A circumstance in which a scribe miscopies text due to inadvertently looking to the side while copying, or accidentally skips over some of it.]

Dittography is the accidental, erroneous act of repeating a letter, word, phrase, or combination of letters by a scribe or copyist. The term is used in the field of textual criticism. The opposite phenomenon, in which a copyist omits text by skipping from a word or phrase to a similar word or phrase further on, is known as haplography. Papyrus 98 in Rev 1:13 has περιεζωσμμενον instead of περιεζωσμενον (doubled μ). Codex Vaticanus in John 13:14 word διδασκαλος is repeated twice. In Codex Vaticanus in Acts, a book of the Bible, verse 19:34, the phrase “Great is Artemis of the Ephesians” appears twice while it only appears once in other manuscripts.

A homophone is a word that is pronounced the same (to a varying extent) as another word, but differs in meaning. A homophone may also differ in spelling. The two words may be spelled the same, for example rose (flower) and rose (past tense of rise), or differently, as in rain, reign, and rein. 

The following passages contain significant textual variations. Click on the bold ones to read a textual discussion on that variant.

Matt. 6:13 Rom. 5:1 Matt. 17:21 1 Cor. 2:1 Mark 1:1 Luke 17:36

Phil. 3:3 Mark 16:9–20 1 Thess. 2:7 Luke 4:4 1 Tim. 3:16 Luke 23:17

John 1:18 James 2:20 John 7:53–8:11 1 John 2:20 Rev. 1:5 Luke 22:43-44

Textual critics also recognize a few other variants as being telltale signs of purposeful scribal alteration. Two of the most common are “conflated readings” and “interpolations.” A conflation is the scribal technique of resolving a discrepancy between two or more variant readings by including all of them. This phenomenon is more prevalent in later manuscripts because the scribe was confronted with a greater variation among the extant witnesses. Interpolations are scribal additions to the manuscript that attempt to clarify the meaning of the text. These account for a host of variants.

As for the rest of the textual variants, textual critics have developed theories and methodologies for selecting the reading that is most likely original. These theories and methodologies generally fall into two categories: (1) those that pertain to external evidence (with a focus on the classification of manuscripts or studies of the documents themselves) and (2) those that pertain to internal evidence (with a focus on discovering the one “original” reading from which all others deviated). Theories and methodologies concerning external evidence focus on establishing criteria to evaluate variant readings present in various manuscripts. Various New Testament textual critics have posited canons for determining the original wording on the basis of external evidence—the evidence of the documents themselves. This endeavor began in the early 1700s, when scholars became dissatisfied with the accuracy of the Textus Receptus. At first, scholars appended variant readings to the Textus Receptus; then they began to abandon the Textus Receptus. In 1707 John Mill of Oxford produced a critical edition of the Textus Receptus with an extensive critical apparatus and a thorough prolegomena that detailed several principles of textual criticism pertaining to genealogical method. Though he did not change the Textus Receptus, he laid the foundations for modern textual criticism. In the 1730s Bengel became the first man to categorize manuscripts according to their age and location, and to formulate the significant principle that textual witnesses must be weighed and not merely counted.

HOW DO WE DETERMINE THE ORIGINAL READING THROUGH The Principles and Practice of New Testament Textual Studies?

In 1751 J. J. Wettstein published a Greek text, Novum Testamentum Graecum, and from 1775 to 1807 J. J. Greisbach published three editions of the Greek New Testament. Both these men provided canons of external criticism that are still recognized as useful today. Karl Lachmann, a classical philologist, produced a fresh text (in 1831) that presented the Greek New Testament of the fourth century. Soon thereafter, Samuel Tregelles produced a text on nearly the same principles as did Lachmann, without knowing it. His text reflects the work of a man who rigidly followed the rule that the reading supported by the earliest manuscripts is preferable. Tischendorf also produced a text, but his work was too heavily influenced by Codex Sinaiticus.

Westcott and Hort then made known their theory (which was chiefly Hort’s) that Codex Vaticanus and Codex Sinaiticus (along with a few other early manuscripts) represent a text that most closely replicates the original writing. Based on this theory, they developed a genealogical tree that traced back from extant witnesses (such as Vaticanus and Sinaiticus) to the original autographs. According to their theory, Vaticanus was almost a perfectly transmitted text from the original. It was a “Neutral Text”—i.e., a text void of textual corruption. Their theory was revolutionary, and their text was responsible for overthrowing the Textus Receptus.

However, this was not the opinion of many textual critics, who became skeptical of recovering the original text through genealogical means. It was judged by several scholars that Westcott and Hort had made a subjective selection about the purity of Codex Vaticanus and then used that to determine the impurity of other manuscripts. Thus, Westcott and Hort’s theory was no longer heartily endorsed. Left without a solid theory for making external judgments, textual critics turned more and more to internal evidence. They began to endorse the canon that the reading that is most likely original is the one that best explains the variants. This canon is a development of Bengel’s maxim, proclivi scriptoni praestat ardua (the harder reading is to be preferred), a maxim he formulated in responding to his own question as to which variant reading is likely to have arisen out of the others.

This overarching canon for internal criticism involves several criteria, which one scholar or another has posited and/or implemented during the past three hundred years of New Testament textual criticism. Having made a thorough historical survey of the development of canons for internal criticism, Eldon Epp summarized all the criteria as follows:

1. A variant’s status as the shorter or shortest reading.
2. A variant’s status as the harder or hardest reading.
3. A variant’s fitness to account for the origin, development, or presence of all other readings.
4. A variant’s conformity to the author’s style and vocabulary.
5. A variant’s conformity to the author’s theology or ideology.
6. A variant’s conformity to Koine (rather than Attic) Greek.
7. A variant’s conformity to Semitic forms of expression.
8. A variant’s lack of conformity to parallel passages or to extraneous items in its context generally.
9. A variant’s lack of conformity to Old Testament passages.
10. A variant’s lack of conformity to liturgical forms and usages.
11. A variant’s lack of conformity to extrinsic doctrinal views.

The primary canon that the shorter reading is usually to be preferred is borne out by the fact that the longer reading is often the result of scribal gap-filling and expansion. Though this is explained at length in a previous chapter, by way of review it can be said that scribes often encountered perceived gaps in the text (especially narratives), which prompted some kind of filling. This gap-filling was often done cognitively in the reading process. However, scribes also took the liberty to insert the gap-filler in writing, so as to make the text more lucid for their readers. Such insertions, whether one word or one sentence, account for the ever-expanding text of the New Testament throughout the course of its transmission. By way of example, gap-filling is nowhere more evident than in the interpolations that occur in the D-text of Acts. Thus, while it cannot be said that the longer reading is always suspect, it can be affirmed that any reading which looks like an attempt to fill in textual gaps is suspect as a scribal addition. This understanding adds new light to the canons number 1 through number 3, listed above. However, it must always be kept in mind that many manuscripts of the earliest centuries were produced by scribes who were inclined to brevity—often with a view to achieving better readability. Therefore, the “shorter reading” canon is not absolute.

Furthermore, it should be admitted that some of the other criteria are problematic when implemented. Two textual critics, using the same principle to examine the same variant unit, will not agree. For example, with respect to number 4, one critic will argue that one variant was produced by a copyist attempting to emulate the author’s style; the other will claim the same variant has to be original because it accords with the author’s style. And with respect to number 5, one will argue that one variant was produced by an orthodox scribe attempting to rid the text of a reading that could be used to promote heterodoxy or heresy; another will claim that the same variant has to be original because it is orthodox and accords with Christian doctrine (thus a heterodoxical or heretical scribe must have changed it). Thus, internal arguments—in and of themselves—often lead to opposite decisions about textual variants because each textual critic has his or her own subjective biases.

What Are Textual Variants [Errors] and How Many Are There?

Comfort thinks scribal gap-filling (in greater detail below) accounts for many textual variants (especially textual expansions and interpolations) in the New Testament—especially in the narrative books, the four Gospels, and Acts. Usually, textual critics examine textual variants as written deviations from the original text. However, there is another way to look at textual variants—namely, as individual “reader-receptions” of the text. By this, I mean that we can look at the origin of textual variants as having been created by various individual scribes as they interacted with the text in the process of reading it. In the centuries prior to the simultaneous-multiple production of copies via dictation (wherein many scribes in a scriptorium transcribed a text dictated to them by one reader), all manuscript copies were made singly—each scribe producing a copy from an exemplar. Many of the differences in the manuscripts came about as the result of transcriptional errors, but other variants were created during the process of reading, which is a clear indication that each scribe interpreted the text as he read it and did not merely copy it verbatim. Whether consciously or unconsciously, the scribe altered the text as he copied it and thereby left a written legacy of his individual readings.

As far as I know, no textual scholar has developed a theory for New Testament textual transmission that deals with this aspect of scribal reception. Perhaps it seems too obvious. The good scribe is expected not to have really processed the text but to have mechanically copied it word-by-word, even letter-by-letter. Jewish Masoretic scribes were trained to copy the Hebrew Scriptures in this manner. After every page, they would count the number of letters and compare them with the number of letters on the exemplar. If the copy differed, it was rejected. But even with this safeguard, some of these scribes became subjectively involved with the text and made changes as they copied. This phenomenon also applies to New Testament scribes. No matter how meticulous or professional, a scribe would still become subjectively involved with the text and—whether consciously or unconsciously—would produce a transcription that differed from his exemplar. Scribes internalized the text in the process of reading it and transcribing it, especially since many of them were emotionally involved with a work of literature that they deemed sacred and inspired.

The observations of certain literary theorists, who focused on reader-reception theory, help us understand the dynamic interaction between the scribe (functioning as a true reader) and the text he or she was copying. Textual critics must take into account the historical situation of the scribes who produced the manuscripts we rely on for doing textual criticism. Textual critics must also realize that scribes were interactive readers. Indeed, as many literary critics in recent years have shifted their focus from the text itself to the readers of the text in an attempt to comprehend plurality of interpretation, so textual critics could analyze variant readings in the textual tradition as being the products of different, personalized “readings” of the text created by the scribes who produced them.

This is where the work of Wolfgang Iser is helpful. To be specific, Iser’s work is useful for understanding how scribes read and processed a text as they transcribed it. Iser is concerned not just with the question of what a literary text makes its readers do but with how readers participate in creating meaning. In other words, the meaning of a text is not inherent in the text but must be actualized by the reader. A reader must act as cocreator of the text by supplying that portion of it which is not written but only implied. Each reader uses his or her imagination to fill in the unwritten portions of the text, its “gaps” or areas of “indeterminacy.” In other words, as the reader adopts the perspectives thrust on him or her by the text, experiences it sequentially, has expectations frustrated or modified, relates one part of the text to the other, imagines and fills in all that the text leaves blank, its meaning is gradually actualized. The reader’s reflection on the thwarting of his or her expectations, the negations of familiar values, the causes of their failure, and whatever potential solutions the text offers require the reader to take an active part in formulating the meaning of the narrative.

Glossary of Technical Terms for New Testament Textual Studies

Whereas readers do this gap-filling in their imaginations only, scribes sometimes took the liberty to fill the unwritten gaps with written words. In other words, some scribes went beyond just imagining how the gaps should be filled and actually filled them. The historical evidence shows that each scribe who made a text created a new written text. Although there are many factors that could have contributed to the making of this new text, one major factor is that the text constantly demands the reader to fill in the gaps. During the reading process, the reader must concretize the gaps by using his or her imagination to give substance to textual omission and/or indefiniteness. Since this substantiation is a subjective and creative act, the concretization will assume many variations for different readers. Iser calls the textual gaps “blanks”; each blank is a nothing that propels communication because the blank requires an act of ideation in order to be filled. Iser wrote, “Blanks suspend connectibility of textual patterns, the resultant break in good continuation intensifies the acts of ideation on the reader’s part, and in this respect the blank functions as an elementary function of communication.” According to Iser, the central factor in literary communication concerns the reader’s filling in of these textual blanks. His theory of textual gaps is useful for understanding scribal reader-reception. Of course, his perception of gaps or blanks is far bigger and more demanding on the reader’s imaginative powers than can usually be applied to New Testament scribes. Nonetheless, scribes were confronted with gaps or blanks that begged for imaginative filling. Many scribes, when confronted with such textual gaps, took the liberty to fill in those gaps by adding extra words or changing the wording for the sake of providing what they thought would be a more communicative text. Indeed, the entire history of New Testament textual transmission is one of the text getting bigger and bigger due to textual interpolations—i.e., the filling in of perceived gaps. We especially see the work of gap-filling in the substantial number of expansions in the D-text of the Gospels and Acts. Whoever edited this text had a propensity for filling in textual gaps, as he perceived them. Such gap-filling is especially pronounced in the book of Acts.

The Coherence-Based Genealogical Method (CBGM)

Modern Approaches to New Testament Textual Criticism

Not all scholars consider the above principles equally valid or applicable to New Testament textual criticism. Today, four approaches to textual criticism can be seen among New Testament scholars. Each of the four current approaches may be identified with individual scholars. For the sake of convenience, these approaches may be called Radical Eclecticism, Reasoned Eclecticism, Reasoned Conservatism, and Radical Conservatism. The term “eclectic” means that the scholar tends to view each textual variant on its own merits instead of blindly following one manuscript or group of manuscripts. The term “conservative” is used here to refer to a generally high view of the traditional Byzantine text type and/or the Textus Receptus.

Radical Eclecticism (G. D. Kilpatrick, J. K. Elliott)

Radical Eclecticism holds to what may be called a purely eclectic text. This approach prefers a text based solely on internal evidence. Adherents of this view argue that since the history of the New Testament text is untraceable, none of the text types carries any weight. Hence the reading of any manuscript may be original, since no manuscript or group of manuscripts is “best.” An eclectic scholar will thus choose the reading that commends itself as best fitting the context, whether in style or thought. This view, held primarily by a minority of British scholars, has been criticized for ignoring the value and importance of the external evidence, particularly the Greek manuscripts.

Reasoned Eclecticism (M. Holmes, B. M. Metzger, K. Aland)

David Alan Black writes, “Reasoned Eclecticism holds that the text of the New Testament is to be based on both internal and external evidence, without a preference for any particular manuscript or text type. This view of the text is represented in the Nestle-Aland and United Bible Societies’ Greek New Testaments. This approach often represents a predilection for manuscripts of the Alexandrian text type. This preference is based largely on Westcott and Hort’s theory that the Byzantine text is a conflation of the Alexandrian and Western texts, and that the superiority of the Alexandrian text over the Western text can be shown through internal evidence. This approach has occasionally been criticized for producing a new “Textus Receptus”—a canonized form of the New Testament text.”

Philip Comfort writes, “Most modern textual critics recognize that textual criticism cannot be myopic [biased, narrow-minded, lacking foresight]; it must be practiced with two eyes—one on the documents (external evidence) and the other on scribal tendencies (internal evidence). This method has been called “reasoned eclecticism.” According to Holmes, “Reasoned eclecticism applies a combination of internal and external considerations, evaluating the character of the variants in light of the manuscripts evidence and vice versa in order to obtain a balanced view of the matter and as a check upon purely subjective tendencies.”[1] Holmes further expands on this method in an article entitled “The Case for Reasoned Eclecticism,” wherein he makes a solid case for this method as being the most viable for the actual practice of New Testament textual criticism.[2] First, he urges that the critic must know and use the documents, citing the famous dictum of Hort, “Knowledge of documents should precede final judgment upon readings.”[3] But he then explains that this can take us only so far—in two respects. First, he argues that “documentary evidence can take us back to the earliest recoverable (or surviving) stage of the textual tradition, but it cannot take us any further. That is, on the basis of external evidence alone we cannot determine whether the earliest recoverable stage of the textual transmission is the autograph or a copy of it.”[4] Second, the extant documentary evidence often presents a situation where one cannot clearly determine which reading has the best documentary support. In the end, then, Holmes concurs with Zuntz who said that documentary evidence can “throw a very considerable weight into the scales of probability [but] will not by itself suffice to determine [a] choice between competing readings.”[5]

Those who practice textual criticism know this all too well. The situation then becomes one of emphasis. Does one give more weight to documentary evidence or to internal consideration? Scholars such as Tregelles, Hort, and Colwell (see comments below) place more emphasis on the documents. I tend to follow their lead. Other scholars, such as Kilpatrick, Boismard, and Elliott, place more emphasis on internal criticism, such that they advocate “thoroughgoing eclecticism.”[6] Other scholars practice reasoned eclecticism, as explained by Holmes. Among those are Aland and Metzger, though each has his own emphasis.”

Reasoned Conservatism (H. A. Sturz)

What might be called Reasoned Conservatism holds that each of the main text types is equally early and independent, going back separately into the second century. Like Reasoned Eclecticism, Reasoned Conservatism sees both internal and external evidence as useful. However, unlike Reasoned Eclecticism, which tends to follow the Alexandrian text, Reasoned Conservatism insists that no single text type can be preferred over all others, and instead emphasizes the geographical distribution of the text types. Scholars who hold to this view argue that the Byzantine text is older than the age of the earliest Byzantine manuscript (fifth century). For example, Byzantine readings once thought to be late have been found in early Egyptian papyri. Therefore, adherents of this view consider the Byzantine text type to be an early and independent witness to the text of the New Testament. They further believe that the reading that is the consensus of the majority of text types is most representative of the autographs. Reasoned Conservatism has been criticized for restoring the Byzantine text (which many feel to be “corrupt”) to a place of usefulness.

Radical Conservatism (Z. Hodges, A. Farstad)

Finally, the approach that may be called Radical Conservatism holds that the Byzantine text type most closely approximates the original text of the New Testament. Scholars who hold to this view prefer the reading of the majority of manuscripts, which are, of course, mainly Byzantine. Several of these scholars have produced the New King James Version, which is based on the Textus Receptus, thus perpetuating the tradition begun by William Tyndale in 1525 and continued in the King James Version of 1611. This approach has been criticized for being too mechanical and for ignoring the fact that manuscripts must be weighed and not just counted. For example, if ten manuscripts are copies of a single parent manuscript, then an error appearing in the parent will appear ten times in ten copies. But these ten copies are equal to a single authority, not to ten.

Aland’s Local-Genealogical Method

Kurt Aland (1979, 43) favors a type of textual criticism which he calls the local-genealogical method. He defines it as follows:

It is impossible to proceed from the assumption of a manuscript stemma, and on the basis of a full review and analysis of the relationships obtaining among the variety of interrelated branches in the manuscript tradition, to undertake a recensio of the data as one would do with other Greek texts. Decisions must be made one by one, instance by instance. This method has been characterized as eclecticism, but wrongly so. After carefully establishing the variety of readings offered in a passage and the possibilities of their interpretation, it must always then be determined afresh on the basis of external and internal criteria which of these readings (and frequently they are quite numerous) is the original, from which the others may be regarded as derivative. From the perspective of our present knowledge, this local-genealogical method (if it must be given a name) is the only one which meets the requirements of the New Testament textual tradition.

The “local-genealogical” method assumes that for any given variation unit, any manuscript (or manuscripts) may have preserved the original text. The problem with doing textual criticism on the local-genealogical basis is that the editors must decide what the authors most likely wrote on a variant-unit by variant-unit basis, which leads to extensive eclecticism (despite Aland’s protest to the contrary). The eclecticism is striking when we examine the selection process for variant readings within a single verse, such as Mark 6:51. In Mark 6:51, the expression και λιαν εκ περισσου εν εαυτοις εξισταντο (“and they were exceedingly, extremely amazed in themselves”) is found in A f¹³ Maj and was adopted as the text for the NU edition. Perhaps this longer reading was accepted over the shorter text (which omits εκ περισσου, “extremely”), found in א B (L), on the supposition that the Alexandrian scribes of א, B, and L were pruning excessive modifiers. However, in the next part of the verse, the shorter reading εξισταντο (“they were amazed”), found in א B L, was adopted by NU, as opposed to the longer reading εξισταντο και εθαυμαζον (“they were amazed and marveled”), found in A D W f¹³ Maj. This is a prime example of atomistic eclecticism (i.e., eclecticism on a variant-unit basis). Within one verse, the reading of א B L was first rejected and then subsequently accepted. It is more consistent to judge that א B L present the original text in both instances and that both longer readings are scribal expansions intended to accentuate the disciples’ amazement over the miracle they just witnessed. This understanding is also consistent with what we know of the overall character of these manuscripts.

This kind of inconsistency is not uncommon. In Matthew 8:21, NU rejected the witness of א B 33: ετερος δε των μαθητων ειπεν αυτω, κυριε, επιτρεψον μοι πρωτον απελθειν και θαψαι τον πατερα μου (“Another of the disciples said to him, ‘Lord, let me first return and bury my father’ ”). Instead, NU favored the reading found in C L W Θ 0250, which adds αυτου (“his”) after μαθητων (“disciples”). Metzger’s comments in TCGNT reveal that most of the committee thought that αυτου was deleted by the scribes of א B 33 to help readers understand that the scribe mentioned in 8:19 was not one of Jesus’ disciples. The excellent documentary testimony of א B 33 was thus rejected because of internal considerations. Four verses later (in 8:25), the testimony of the same manuscripts is accepted for the exclusion of οι μαθηται αυτου at the beginning of the verse.

Another occurrence of atomistic eclecticism occurs in the NU text of John 9:4. In the first part of the verse, NU reads ημας δει (“it is necessary for us”), following the testimony of P⁶⁶ P⁷⁵ א* B D L W 0124. In the second part of the verse, NU reads του πεμψαντος με (“the one having sent me”), following the testimony of א^c A B C D 0124 and rejecting the testimony of P⁶⁶ P⁷⁵ א* L W, which read του πεμψαντος ημας (“the one having sent us”). In the first part of this verse, the testimony of P⁶⁶ P⁷⁵ א B L W is accepted, but in the next part of the very same clause, the testimony of the P⁶⁶ P⁷⁵ א* L W was rejected. This is the result of eclecticism, wherein internal evidence is given more weight than documentary evidence (see TCGNT).

In another case, in Romans 8:11, the reading ο εγειρας χριστον εκ νεκρων (“the one having raised Christ from the dead”) is accepted into the NU text on the authority of B D² F G. The only merit the NU reading has is that it is the shortest one. However, in general, the NU editors were categorically suspicious of a reading supported by B with D F G (see TCGNT on Rom 8:11b), so it seems inconsistent that this reading would be accepted on the basis of B D² F G. But this is the result of the eclectic method.

These few examples show that many modern textual critics attempt to operate according to a syncretism of two conflicting theories: one that says the best readings are preserved in the best manuscripts and another that says the best readings are simply those that best fit the text, no matter what manuscripts they come from. As far as I am concerned, the best approach is to first establish which manuscripts (or groups of manuscripts) are the best authorities for each particular book or section (e.g., Paul’s Epistles, General Epistles) of the New Testament. Once these are reckoned, the burden of proof for any textual variation is to show that these manuscripts do not have the original wording. As always, the critic must first look for transcriptional causes of error or variation. If transcriptional errors cannot account for the variation, then the critic has to look to the criteria for internal evidence. But one needs very strong arguments on internal grounds to overthrow strong documentary attestation. Of course, this means that the critic must know each manuscript well and have adequate knowledge about the workmanship and tendencies of the scribe who produced it.

Metzger’s Judgment of Variant Readings according to Text-Types

Because there are so many individual manuscripts, textual critics are hard-pressed to know the individual characteristics of each manuscript. Consequently, many textual critics categorize the manuscripts into text-types, which they then use in their evaluation of textual variants. One of the foremost textual critics of our era, Bruce Metzger, exhibits this kind of evaluation. He placed the extant manuscripts into one of four text-types, usually called Alexandrian, Western, Caesarean, and Byzantine. Each of these requires some explanation. (More detailed explanations can be found in Metzger 1992, 211–219).

Alexandrian Manuscripts

The Alexandrian text is found in manuscripts produced by scribes trained in the Alexandrian scriptoral tradition, the best of its kind in Greco-Roman times. Such scribes were schooled in producing well-crafted, accurate copies. Among the New Testament manuscripts, it can be seen that there are several early Alexandrian manuscripts (sometimes called proto-Alexandrian) and later Alexandrian manuscripts. The earlier manuscripts are usually purer than the later ones in that the earlier are less polished and closer to the ruggedness of the original writings. In short, these manuscripts display the work of scribes who had the least creative interaction with the text; they were produced by scribes who stayed with their task of making faithful copies. Quite significantly, the text of several of the earlier or proto-Alexandrian manuscripts was transmitted quite faithfully. This is exemplified in the high percentage of textual agreement between P⁷⁵ and B, thereby affirming Hort’s theory that Codex Vaticanus traces back to an early, pure text. This textual relationship and others are detailed in my book, The Quest for the Original Text of the New Testament (1992, 101–118).

Metzger (1992, 216) lists the following Alexandrian witnesses, in the categories “Proto-Alexandrian” and “Later Alexandrian.”

Proto-Alexandrian:

^P45 (in Acts) P⁴⁶ P⁶⁶ P⁷⁵ א B Sahidic (in part), Clement of Alexandria, Origen (in part), and most of the papyrus fragments with Pauline text

Later Alexandrian:

Gospels: (C) L T W (in Luke 1:1–8:12 and John) (X) Z Δ (in Mark) Ξ Ψ (in Mark; partially in Luke and John) 33 579 892 1241 Bohairic

Acts: P⁵⁰ A (C) Ψ 33 81 104 326

Pauline Epistles: A (C) H I Ψ 33 81 104 326 1739

Catholic Epistles: P²⁰ P²³ A (C) Ψ 33 81 104 326 1739

Revelation: A (C) 1006 1611 1854 2053 2344; less good P⁴⁷ א

Western Manuscripts

The so-called “Western” text is a loose category. Actually, it is probably best to call it a kind of “popular” text inasmuch as most of the manuscripts that get put in this text-type share the common traits of scribal expansion, harmonization, and amelioration. Those who defend the cohesiveness of this text-type indicate that it seems to have developed at one point in history (mid-to late second century) and in a certain geographical region (Western Christendom). This form of the Gospels, Acts, and Paul’s Epistles circulated in North Africa, Italy, and Gaul (which are geographically Western), but so-called “Western” manuscripts have also come from Egypt and other locations in the East. It is represented in the Old Latin manuscripts, Syriac manuscripts, and in the D-text (a special brand of the Western text—see discussion at the beginning of Acts). The Western text also prevails in the writings of Marcion, Tatian, Irenaeus, and Tertullian.

The “Western” witnesses listed by Metzger (1992, 214) are as follows:

Gospels: D W (in Mark 1:1–5:30) 0171 it syr^s syr^c (in part), early Latin fathers, Tatian’s Diatessaron

Acts: 𝔓²⁹ P³⁸ P⁴⁸ D 383 614 syr^hmg, early Latin fathers

Paul’s Epistles: The Greek-Latin diglots D E F G; Greek Fathers to the end of the third century; it and early Latin Fathers; Syrian Fathers to about a.d. 450

The Western text is not apparent in the General Epistles and Revelation. The recently published papyrus, P¹¹² (fifth century), is Western. And I would put a question mark next to 𝔓²⁹ because its text is too small to determine its textual affinities.

Caesarean Manuscripts

Another small group of manuscripts constitute a group known as the Caesarean text. Various scholars such as Streeter and Lake demonstrated that Origen brought a text with him from Egypt to Caesarea, which was then transported to Jerusalem. This text, showing a mixture of Alexandrian and Western readings, is apparent in the following manuscripts—only in the Gospels: ��⁴⁵, W (in Mark 5:31–16:20), family 1 (f¹), family 13 (f¹³), Θ, 565, and 700.

Byzantine Manuscripts

The Byzantine manuscripts constitute the largest group and are the furthest removed from the original text in most sections of the New Testament. The one notable exception is the book of Revelation, where several Byzantine manuscripts preserve a purer form of the text.

The Byzantine manuscripts are as follows:

Gospels: A E F G H K P S V W (in Matthew and Luke 8:13–24:53) Π Ψ (partially in Luke and John) Ω and most minuscules

Acts: H L P 049 and most minuscules

Epistles: L 049 and most minuscules

Revelation: 046 051 052 and many minuscules

Metzger argues that usually a variant reading “which is supported by a combination of Alexandrian and Western witnesses is superior to any other reading” (1992, 218). The observant reader will see that this kind of statement appears repeatedly throughout Metzger’s textual commentary on the United Bible Societies’ Greek New Testament, in support of the committee’s decisions about certain readings. Metzger also made the following important observation:

In the evaluation of readings which are supported by only one class of witnesses, the student will probably find that true readings survive frequently in the Alexandrian text alone, less frequently in the Western group alone, and very rarely only in Caesarean witnesses. As a rule of thumb, the beginner may ordinarily follow the Alexandrian text except in the case of readings contrary to the criteria which are responsible for its being given preference in general. Such a procedure, however, must not be allowed to degenerate into merely looking for the reading which is supported by B and א (or even by B alone, as Hort was accused of doing); in every instance a full and careful evaluation is to be made of all the variant readings in the light of both transcriptional and intrinsic probabilities. The possibility must always be kept open that the original reading has been preserved alone in any one group of manuscripts, even, in extremely rare instances, in the Koine or Byzantine text. (1992, 218–219)

List of Major Textual Variants In the Greek New Testament In English Translation

Metzger’s observations are important, for they evolved from years of working with textual variants. But I would add one qualifier to the notion that a reading is likely original if it has support from several text-types. I would stipulate that the documentary support must be early and diverse. Diverse testimony among many later manuscripts (i.e., not the earliest ones), in my mind, signals only that the reading had been copied frequently in various sectors of the church; it does not necessarily validate a reading’s originality.

Documentary Approach (Samuel Prideaux Tregelles, F. J. A. Hort, B. F. Westcott, E. C. Colwell, Edward D. Andrews, Don Wilkins)

“Reasoned eclecticism” or the “local-genealogical” method in actual practice tend to give priority to internal evidence over external evidence, resulting in the atomistic eclecticism. I agree with Westcott and Hort that it has to be the other way around if we are going to recover the original text. In their compilation of The New Testament in the Original Greek, Hort wrote, “Documentary evidence has been in most cases allowed to confer the place of honour against internal evidence” (1881, 17).

Colwell was of the same mind when he wrote “Hort Redivivus: A Plea and a Program.” In this article, Colwell decried the “growing tendency to rely entirely on the internal evidence of readings, without serious consideration of documentary evidence” (1969a, 152). Colwell called upon scholars to attempt a reconstruction of the history of the manuscript tradition. But very few scholars have followed Colwell’s urgings because they believe (in agreement with Aland as quoted in appendix B) that it is impossible to reconstruct a stemma (a sort of manuscript “family tree”) for the Greek New Testament. Perhaps they hold this line because they fear that some will attempt to make a stemma leading back to the original, and that such a reconstruction will involve a subjective determination of the best line of manuscripts. Westcott and Hort have been criticized for doing this when they posited the “Neutral” text, leading from B back to the original.

However, a reconstruction of the early manuscript tradition does not necessarily mandate a genealogical lineage back to the original text—although that is the ultimate purpose of making a stemma. The reconstruction can help us understand the relationships between various manuscripts and provide insights into origin and associations. In the process, it might also be discovered that, out of all the extant manuscripts, some of the earliest ones are, in fact, the closest replications of the original text.

One of the most compelling reasons for returning to a documentary approach is the evidence that the second-century papyrus P⁷⁵ provides. This is the gospel manuscript (containing Luke and John) that has changed—or should have changed—nearly everyone’s mind about abandoning a historical-documentary approach. It is a well-known fact that the text produced by the scribe of P⁷⁵ is a very accurate manuscript. It is also well-known that a manuscript like P⁷⁵ was the exemplar for Codex Vaticanus; the texts of P⁷⁵ and B are remarkably similar, demonstrating 83-percent agreement (see Porter 1962, 363–376, a seminal article on this issue).

Prior to the discovery of P⁷⁵ (which was published in 1961), many textual scholars were convinced that the second- and third-century papyri displayed a text in flux, a text characterized only by individual independence. The Chester Beatty Papyrus, P⁴⁵, and the Bodmer Papyri, P⁶⁶ (uncorrected) and P⁷² (in 2 Peter and Jude), show this kind of independence. Scholars thought that scribes at Alexandria must have used several such manuscripts to produce a good recension—as is exhibited in Codex Vaticanus. Kenyon conjectured:

During the second and third centuries, a great variety of readings came into existence throughout the Christian world. In some quarters, considerable license was shown in dealing with the sacred text; in others, more respect was shown to the tradition. In Egypt this variety of texts existed, as elsewhere; but Egypt (and especially Alexandria) was a country of strong scholarship and with a knowledge of textual criticism. Here, therefore, a relatively faithful tradition was preserved. About the beginning of the fourth century, a scholar may well have set himself to compare the best accessible representatives of this tradition, and so have produced a text of which B is an early descendant. (1940, 250)

Much of what Kenyon said is accurate, especially about Alexandria preserving a relatively pure tradition. But Kenyon was wrong in thinking that Codex Vaticanus was the result of a “scholarly recension,” resulting from “editorial selection” across the various textual histories (1949, 208). Kenyon cannot be faulted for this opinion, because P⁷⁵ had not yet been discovered when he wrote. However, the discovery of P⁷⁵ and Vaticanus’s close textual relationship to it have caused textual critics to look at things differently, for it is now quite clear that Codex Vaticanus was a copy (with some modifications) of a manuscript much like the second-century papyrus P⁷⁵, not a copy of a fourth-century recension.

Zuntz held an opinion similar to Kenyon’s, positing an Alexandrian recension. After studying P⁴⁶, Zuntz imagined that the Alexandrian scribes selected the best manuscripts and gradually produced a text that reflected what they considered to be the original. In other words, they functioned as the most ancient of the New Testament textual critics. Zuntz believed that, from at least the middle of the second century to the fourth century, the Alexandrian scribes worked to purify the text from textual corruption. Speaking of their efforts, Zuntz wrote:

The Alexander correctors strove, in ever repeated efforts, to keep the text current in their sphere free from the many faults that had infected it in the previous period and which tended to crop up again even after they had been obelized [i.e., marked as spurious]. These labours must time and again have been checked by persecutions and the confiscation of Christian books, and counteracted by the continuing currency of manuscripts of the older type. Nonetheless they resulted in the emergence of a type of text (as distinct from a definite edition) which served as a norm for the correctors in provincial Egyptian scriptoria. The final result was the survival of a text far superior to that of the second century, even though the revisers, being fallible human beings, rejected some of its own correct readings and introduced some faults of their own. (1953, 271–272)

The point behind Zuntz’s conjecture of a gradual Alexandrian recension was to prove that the Alexandrian text was the result of a process beginning in the second century and culminating in the fourth century with Codex Vaticanus. In this regard, Zuntz was incorrect. This, again, has been proven by the close textual affinity between P⁷⁵ and B. The “Alexandrian” text already existed in the late second century; it was not the culmination of a recension. In this regard, Haenchen wrote:

In P⁷⁵, which may have been written around 200 a.d., the “neutral” readings are already practically all present, without any need for a long process of purification to bring them together miro quodam modo out of a multitude of manuscripts.… P⁷⁵ allows us rather to see the neutral text as already as good as finished, before that slow development could have started at all; it allows us the conclusion that such manuscripts as lay behind Vaticanus—even if not for all New Testament books—already existed for centuries. (1971, 59)

Kurt Aland’s thinking was also changed by P⁷⁵. He used to speak of the second- and third-century manuscripts as exhibiting a text in flux or even a “mixed” text, but not after the discovery of P⁷⁵. He wrote, “P⁷⁵ shows such a close affinity with the Codex Vaticanus that the supposition of a recension of the text at Alexandria, in the fourth century, can no longer be held” (1965, 336).

The discovery of P⁷⁵ shows that Hort was basically right in his assertion that Codex Vaticanus must trace back to a very early and accurate copy. Hort (1882, 250–251) had written that Codex Vaticanus preserves “not only a very ancient text, but a very pure line of a very ancient text.” But some scholars may point out that this does not automatically mean that P⁷⁵ and B preserve the original text. What it does mean, they say, is that we have a second-century manuscript showing great affinity with a fourth-century manuscript whose quality has been highly esteemed. However, Gordon Fee (1974, 19–43) has demonstrated that there was no Alexandrian recension before the time of P⁷⁵. In an article appropriately title “P⁷⁵, P⁶⁶, and Origen: The Myth of Early Textual Recension in Alexandria,” Fee posits that there was no Alexandrian recension before the time of P⁷⁵ (late second century) and Codex Vaticanus (early fourth) and that both these manuscripts “seem to represent a ‘relatively pure’ form of preservation of a ‘relatively pure’ line of descent from the original text.” In other words, the original text of Luke and John is virtually preserved in P⁷⁵. Of course, P⁷⁵ is not perfect, but it is closer to perfect than Codex Vaticanus, partially because it is 125–150 years closer to the original text.

Some textual critics, however, are not convinced that the P⁷⁵/B type of text is superior to another type of early text, which has been called the “Western” text. The “Western” form of the text was early in that it appears to have been used by Marcion, Irenaeus, Tertullian, and Cyprian—all of whom were alive in the second century. The name “Western” was given to this type of text because it circulated primarily in western regions like North Africa, Gaul, and Italy, but it was also present in Syria and even in Egypt. Thus, most scholars recognize that the “Western” text is not really a text-type; rather, it is a loose categorization of early texts that were not Alexandrian (which is why “Western” is often put in quotation marks in the literature). Some scholars see it as a complete misnomer. Colwell, for example, states, “The so-called Western text or Delta type text is the uncontrolled, popular edition of the second century. It has no unity and should not be referred to as the ‘Western text’ ” (1969b, 53). The Alands also see it to be nothing more than a loose association of manuscripts, arguing, “Wherever we look in the West, nowhere can we find a theological mind capable of developing and editing an independent ‘Western text.’ ” (1987, 54).

These observations aside, some scholars are still skeptical that the P⁷⁵/B type of text is at all superior to the Western text. They argue that the preference given to B and P⁷⁵ is based on a subjective appreciation of the kind of text they contain (generally terser than the “Western” text), rather than on any kind of theoretical reconstruction of the early transmission of the text (see Epp 1974, 390–394). It is argued that this same subjective estimation was at work when Westcott and Hort decided that B was intrinsically superior to D (Westcott and Hort 1882, 32–42). However, the notion that manuscripts like P⁷⁵ and B represent the best of textual purity is persistent, particularly among textual critics who have worked with many actual manuscripts—both of the proto-Alexandrian type and the so-called Western type. In the task of compiling transcriptions and/or doing textual analysis, these critics have seen firsthand the kind of errors, expansions, harmonizations, and interpolations that are far more present in Western manuscripts.

In conclusion, my preference for emphasizing the documentary method in making text-critical choices is revealed in the fact that I decide against many choices made by the editors of the NU text. The reader may see these decisions in the following notes:

Matthew 3:16; 4:24; 5:28; 8:21; 9:14, 26; 12:47; 13:35b; 14:16, 27, 30; 15:6b, 14; 17:9; 18:15; 19:22; 21:44; 25:6; 27:49
Mark 3:32; 6:51; 7:4; 15:12; 16:8 [ending to Mark]
Luke 3:22a; 8:43; 14:17; 17:24; 20:9; 22:43–44
John 1:34; 3:31–32; 5:44; 6:14; 7:9; 7:53–8:11; 9:4, 38–39a; 10:8, 16, 18; 11:45–46; 13:2a, 2c, 32; 16:23; 20:31; 21:18
Acts 3:6; 7:13, 38; 9:12; 16:12
Romans 3:4; 7:17; 8:11a, 23; 11:17; 12:14; 15:33 [placement of doxology]
1 Corinthians 1:14; 3:13; 4:2; 7:7, 15; 8:3a, 3b; 9:9b; 10:2; 12:10
2 Corinthians 4:5b; 5:3, 12
Galatians 1:3, 6, 15a; 2:12a, 12b; 3:21a
Ephesians 1:1b, 15, 18; 3:19; 4:24, 28; 5:2a, 20; 6:12a, 19
Philippians 3:3, 7, 10, 12a
Colossians 2:7a, 10, 13, 23; 3:6, 22b, 23; 4:8, 12
1 Thessalonians 3:2, 13; 5:4, 9
2 Thessalonians 2:13; 3:6
2 Timothy 3:15
Philemon 25
Hebrews 1:8; 3:2; 4:3a; 7:4, 28; 9:1, 19; 11:4; 12:1, 3, 4; 13:15, 21c, 24, 25a
James 1:17; 2:3; 4:14a; 5:4
1 Peter 1:12b; 2:21; 3:14, 18; 4:11; 5:8, 10b, 10c
2 Peter 1:3; 2:6a; 3:18b
1 John 3:23a; 5:20b
2 John 8
Jude 5
Revelation 1:6b; 9:12–13a, 13b; 11:8; 12:8a, 10; 13:18; 14:3a, 5; 15:3, 6; 16:5b; 18:2, 3; 19:11

Refining the Documentary Method

All textual critics—including those working with the classics—utilize both external criticism and internal criticism in the process of selecting the one variant reading that is most likely original. And all textual critics must do this on a variant-unit by variant-unit basis. Some give priority of place to internal criticism over external criticism; others do the opposite. The editors of the NU text demonstrate that they tried to do both; this can be seen in Metzger’s discussions in A Textual Commentary on the Greek New Testament (TCGNT). However, it is my observation that the resultant eclectic text exhibits too much dependence on the “local” aspect of the “local-genealogical” method. This means that the decision-making, on a variant unit-by-unit basis, produced a text with an uneven documentary presentation. Furthermore, the committee setting, with the committee voting on each significant textual variant, cannot help but produce a text with uneven documentation. In short, the NU Greek text, being eclectic, does not reconstruct a text that any ancient Christian actually read, even though it is probably a close replication of the original writings.

Those who adhere to the documentary approach should be aware of the best manuscripts for each book of the New Testament and/or section of the New Testament. Ideally, one should select the premier group of manuscripts as the primary witnesses for certain books and/or sections of the New Testament, not for the entire New Testament. Since each book of the New Testament was, in its earliest form, a separate publication, the manuscript selections need to be made on a book-by-book basis. Hort was too broad-reaching to have embraced Codex Vaticanus as the preeminent text for the entire New Testament, when we now know that there are superior witnesses for certain sections of the New Testament. The same can be said for Tischendorf, who was too enthusiastic about his prize find, Codex Sinaiticus. However, for several books of the New Testament, we can hardly do better than start with Codex Vaticanus and/or with Codex Sinaiticus—if only for the simple reason that they often contain more extant text than do the earlier papyri and that they usually provide witness to an early text.

Once the best manuscripts for each book and/or group of books in the New Testament are established, these manuscripts need to be pruned of obvious errors and singular variants. Then these will be the manuscripts used for determining the most likely original wording. The burden of proof on the praxis of textual criticism is to demonstrate that the best manuscripts, when challenged by the testimony of other witnesses, do not contain the original wording. The part of this process that corresponds to the Alands’ “localness” is that the text must be determined on a variant-unit basis. However, my view of the “genealogical” aspect is that it must be pre-established for an entire book and not re-created verse by verse, which accounts for an eclectic presenting a very uneven documentary presentation. Of course, internal criticism will have to come into play when documentary evidence is evenly divided. And, on occasion, it must be admitted that two (or more) readings are equally good candidates for being deemed the original wording.

In my book The Quest for the Original Text of the New Testament, published in 1990, I described various groupings of the papyrus manuscripts exhibiting textual affinities, on a book-by-book basis. These groupings covered the early papyri from P¹ to P⁹². In some significant articles written during the past few years, Eldon Epp has also explored grouping manuscripts into what he calls “textual clusters.”⁴⁸ He sees the papyri as belonging to one of four clusters, which he calls the “A” group for later Alexandrian papyri; the “B” group for early papyri that have affinities with Vaticanus (B); the “C” group for the papyri that are linked with what could be called “Caesarean,” and the “D” group for those papyri that have associations with Beza (D). My groupings are not as broad-based because I think groups need to be established for books or sections of the New Testament (such as the Gospels and Paul’s Epistles), as compared to the entire New Testament per se. Nonetheless, there is a great deal of overlap between my groupings and Epp’s. With the accession of more published papyri (P¹⁰⁰ to P¹¹⁵ in 1998–1999), we can expand the population of each broad group and then establish tighter textual communities—that is, manuscripts showing a high degree of textual agreement. Admittedly, this is somewhat of a tenuous procedure because of the fragmentary condition of several of the papyri. Nonetheless, it is a fruitful exercise to compare the smaller manuscripts with the larger papyri in an effort to establish textual relationships. The purpose of these efforts is to establish the premier textual group for each section of the New Testament.

The selections I list below largely follow Metzger’s identification of witnesses, but I would especially modify and expand Metzger’s list of Proto-Alexandrian manuscripts in two ways: (1) I would add more manuscripts, especially the more recently published papyri; (2) I would specify the proto-Alexandrian manuscripts by New Testament book or section. Thus, the Proto-Alexandrian manuscripts are as follows:

Gospels: P¹, P^4/64/67, P⁵, P²⁹, P³⁵, P³⁹, P⁵²?, P^66c, P⁷¹, P⁷⁵, P⁷⁷, P⁹⁰, P⁹⁵, P¹⁰¹, P¹⁰³, P¹⁰⁴, P¹⁰⁶, P¹⁰⁷, P¹⁰⁸

Acts: P⁴⁵, P⁵³, P⁹¹, 0189

Paul’s Epistles: P¹³ (Hebrews), P^15/16, P³⁰, P⁴⁰, P⁴⁶, P⁶⁵, P⁹², 0220

General Epistles: P²⁰, P²³, P⁷² (in part), P⁸¹, P¹⁰⁰

Revelation: P¹⁸, P²⁴, P⁴⁷, P⁹⁸, P¹¹⁵

In the main, I see the proto-Alexandrian manuscripts as being the best witnesses to the original text. Some may call this a subjective predetermination. But I honestly say that this favoritism has come from studying thousands of textual variants, as well as studying scribal tendencies, and usually coming to the conclusion that the proto-Alexandrian manuscripts have preserved the most primitive, if not the original, wording. Of course, these manuscripts are not perfect. P⁷⁵, one of the most pristine manuscripts, has flaws. Nonetheless, its fidelity and acuity far outweigh its imperfections.

Assessing Manuscripts through a Study of Singular Variants

In accord with Westcott and Hort’s mandate that “knowledge of documents must precede all judgments of readings,” Colwell devised a method wherein he could determine the peculiarities of each manuscript by studying the singular variants in that manuscript. Colwell believed that the singular readings of a manuscript were the textual creations of the scribe, and that an analysis of the patterns found within these singular readings would reveal the habits of the scribe.³⁷ James Royse did an extensive study of the major early papyri focusing on the singular readings which was based on the same rationale that Colwell proposed. In a lengthy, thorough dissertation, entitled “Scribal Habits in Early New Testament Papyri,” Royse characterized the scribal habits exhibited in several early manuscripts (P⁴⁵, P⁴⁶, P⁴⁷, P⁶⁶, P⁷², P⁷⁵) by studying each manuscript’s singular readings (i.e., readings found in that manuscript only, independent of all other extant documents).³⁸ Students should make good use of the work of Colwell and Royse in their description of the papyrus manuscripts. They should also apply the reader-reception methodology explained in the previous chapter to ascertain the response of the scribe to the text as a reader, as opposed to a mere copier. In other words, students should attempt to explain the creation of certain singular variants as being the result of scribal interaction with the text.

Singular readings provide the best—and perhaps only—means of studying a scribe’s reception of the text because they are individualized readings. Other variation units cannot be used for this study because it is always possible that the scribe was simply copying a reading from a previous exemplar. Since we are fairly certain that singular readings were not copied from other manuscripts, they must have been prompted by the text itself—or, should I say, by the scribe’s interaction with the text as a real reader. This is where we see a confluence of Iser’s theories and those of Jauss.

According to Iser, the implied reader is a textual prerequisite because it is regarded as a role of the reader that is written into the text. And it is a prerequisite for the production of meaning in that it is the composite of all the textual clues that are provided for the guidance of the actual reader in his interpretation of the text. The implied reader is therefore a sign-like, text-immanent to which actual readers could react in many different ways. The actual reader’s reactions depend upon what horizon of expectations the reader brings to the text. This is Jauss’s position.³⁹ When we combine these theories, it becomes clear that scribes who functioned as readers produced some very creative responses to the gaps (or lapses of meaning) they encountered in the text. These responses have been preserved for us in the form of singular variants.

Colwell and Tune defined a textual variation unit as that length of the text (1) where the Greek New Testament manuscripts present at least two variant forms and (2) where each variant form is supported by at least two Greek manuscripts.⁴⁰ When there is a variant reading supported by only one Greek New Testament manuscript, this is called a singular variant—as understood by many textual critics today.⁴¹ It is important to note that the definition of a singular variant does not include any mention of versional or patristic support, only of Greek manuscripts. Versions (as translations) have their own history of textual appropriation and transformation, which may have coincidentally matched what occurred in a Greek textual alteration without having been directly influenced by that Greek manuscript. Patristic citations are also problematic and cannot be counted toward excluding a Greek reading from being a singular variant if they happen to line up with the singular variant.

My criteria for a singular variant accords with Royse’s, who said that a singular reading is any variant reading which is found in only one of the continuous-text Greek manuscripts—that is, it is a reading found in one of the New Testament papyri, uncials, or minuscules.⁴² This categorization excludes lectionaries, patristic sources, and ancient versions because of the well-known difficulties of studying the evidence of such witnesses. Exclusion of this material not only facilitates the task of constructing a list of singulars easier; it also helps to enhance the objectivity of the list. My criteria for a singular variant also includes Colwell’s observation that there are such things as identical singular readings—that is, two scribes of two completely different eras and regions may have created the same reading coincidentally. Colwell said, “Since corruption was universal, identical singular readings with only minor scattered support elsewhere should be assumed to be coincidental in these agreements—unless other external evidence establishes relationship.”⁴³

Not all singular readings are significant. Some must be categorically eliminated from a study of scribal reception. These include obvious transcriptional errors, meaningless transpositions, itacisms, and nonsense readings. A few other kinds of singular readings may or may not be noteworthy; these are minor lexical substitutions and grammatical adjustments. Of course, both of these changes could have been prompted by some kind of perceived lack in the text, but not in the Iserian sense of a blank. A student needs to be judicial in dealing with such variants. Most of the other singular readings are worthy of analysis.

It is important to note that Colwell and Royse describe only the habits of particular scribes as copyists; they do not describe the receptions of scribes as readers. Thus, Colwell and Royse primarily analyze the results of their copying and attempt to explain all singular variants in the traditional terms of textual criticism. They both speak of spelling errors and grammatical emendations or flaws. They both speak of homoeoteleuton and homoeoarchton causing parablepsis or scribal leaps. They both speak of harmonization to the immediate context and harmonization to remote parallels. However, neither of them focus on the activity of the scribe as a reader, who brings his own horizon of expectations to the text and who is also impelled by various textual constructs to produce individualized interpolations or ingenious modifications. Such singular readings are not a display of aberrant copying as much as they are a reflection of how the scribe became involved with the reading process. True, many singular variants can be identified as having been created by the immediate context, which is a traditional canon in textual criticism. So, admittedly, there will be some overlap between internal criticism based on immediate context and an analysis of reader reception because both look to the context as providing the textual clues for reader reception. However, Colwell and Royse did not analyze what structured act in the text (in the Iserian sense) prompted the scribes as readers to make various changes. Nor did Colwell and Royse consider the scribe’s horizon of expectation as a motivating factor in stimulating some textual change. I think students analyzing singular variants should also attempt to see how the scribes, functioning as readers, reacted to the network of response-inviting structures in the text and filled in various blanks by drawing upon their repository of reading experience and life experience (Lebenswelt).

When Colwell asks the question, “why singular variants?” he furnishes the answer from a textual transmission perspective,⁴⁴ not necessarily from a reader-text interaction perspective. Thus, his characterizations of individual scribes is based on his observation of them as copyists, not as interactive readers. This is evident in the following comment: “One scribe is liable to dittography; another to omission of lines of text; one reads well; another remembers poorly.”⁴⁵ In context, Colwell’s definition of “reading” describes nothing more than the act of rote reading for the sake of copying. In Royse’s final analysis of the scribal tendencies of P⁴⁵, P⁶⁶, and P⁷⁵, he provides an illuminating profile of each of the scribes.⁴⁶ However, not one of these profiles describes the scribes as individualized, interactive readers. I do not say these things to criticize Colwell’s methodology or Royse’s analysis, for both scholars presented solid results that were consistent with what they set out to do. And the student is encouraged to follow their guidelines. But I would also urge students to analyze what Colwell and Royse did not analyze—namely, the interactive process of reading and how this was responsible for the creation of several significant variant readings.⁴⁷

David Alan Black writes, “Even though these various schools of thought can be identified, it is necessary to realize that textual critics might partially adopt the approaches of two or more schools, so that a synthesis often results. Likewise, schools tend to fluctuate over time due to the influx of new leaders and materials.”

Scribal Gap-Filling

It is Comfort’s opinion that scribal gap-filling accounts for many of the textual variants (especially textual expansions) in the New Testament—particularly in the narrative books (the Four Gospels and Acts). Usually, textual critics examine textual variants as accidental deviations from the original text. However, some variants may be accounted for more accurately as individual “reader-receptions” of the text. By this, I mean variants created by individual scribes as they interpreted the text in the process of reading it. In the centuries prior to the production of copies via dictation (wherein many scribes in a scriptorium transcribed a text as it was dictated to them by one reader), all manuscript copies were made singly—each scribe working alone to produce a copy from an exemplar. The good scribe was expected not to have really processed the text internally but to have mechanically copied it word by word, even letter by letter. But no matter how meticulous or professional, a scribe would become subjectively involved with the text and—whether consciously or unconsciously—at times produce a transcription that differed from his exemplar, thereby leaving a written legacy of his individual reading of the text.

Even a scribe as meticulous as the one who produced ��⁷⁵ could not refrain, on occasion, from filling in a perceived gap. This occurs in the parable in Luke 16:19–31 where the reader is told of an unnamed rich man and a beggar who has a name, Lazarus. Perceiving a gap in the story, the scribe gives the rich man a name: “Neues,” perhaps meaning “Nineveh” (see note on Luke 16:19). Other scribes gave names to the two revolutionaries crucified with Jesus: Zoatham and Camma (in some manuscripts), or Joathas and Maggatras (in other manuscripts; see note on Matt 27:38). Many other scribes filled in bigger gaps, especially in narratives. In the story of the salvation of the Ethiopian eunuch recorded in Acts 8:26–40, some scribes added an entire verse so as to fill in a perceived gap of what one must confess before being baptized. Thus, we are given these extra words in Acts 8:37, “And Philip said, ‘If you believe with all your heart, you may [be baptized].’ And he [the eunuch] replied, ‘I believe that Jesus Christ is the Son of God.’ ” (See note on Acts 8:37 for further discussion.)

The observations of certain literary theorists who focus on reader reception help us understand the dynamic interaction between the scribe (functioning as a true reader) and the text he or she was copying. Textual critics must take into account the historical situation of the scribes who produced the manuscripts we rely on for textual criticism. Textual critics must also realize that scribes were interactive readers. Indeed, as many literary critics in recent years have shifted their focus from the text itself to the readers of the text in an attempt to comprehend plurality of interpretation, so textual critics could analyze variant readings in the textual tradition as possibly being the products of different, personalized “readings” of the text created by the scribes who produced them.

The work of Wolfgang Iser is useful for understanding how scribes read and processed a text as they transcribed it. Iser is concerned not just with the question of what a literary text makes its readers do but with how readers participate in creating meaning. In other words, the meaning of a text is not inherent in the text but must be actualized by the reader. A reader must act as cocreator of the text by supplying that portion of it which is not written but only implied. Each reader uses his or her imagination to fill in the unwritten portions of the text, its “gaps” or areas of “indeterminacy.” In other words, the meaning of a text is gradually actualized as the reader adopts the perspectives thrust on him or her by the text, experiences it sequentially, has expectations frustrated or modified, relates one part of the text to the other, and imagines and fills in all that the text leaves blank. The reader’s reflection on the thwarting of his or her expectations, the negations of familiar values, the causes of their failure, and whatever potential solutions the text offers, require the reader to take an active part in formulating the meaning of the narrative.

While readers do this gap-filling in their imaginations only, scribes sometimes took the liberty to fill the unwritten gaps with written words. In other words, some scribes went beyond just imagining how the gaps should be filled and actually filled them. The historical evidence shows that each scribe who copied a text created a new written text. Although there are many factors that could have contributed to the making of this new text, one major factor is that the text constantly demands the reader to fill in the gaps.

A literary work is not autonomous but is an intentional object that depends on the cognition of the reader. As an intentional object, a literary work cannot fill in all the details; the reader is required to do this. During the reading process, the reader must concretize the gaps by using his or her imagination to give substance to textual omission and/or indefiniteness. Since this substantiation is a subjective and creative act, the concretization will assume many variations for different readers. For example, the Gospel of Luke says that the crowds who had watched Jesus’ crucifixion “returned home, beating their breasts” (Luke 23:48). Although it would seem that most readers are given enough text to visualize this scene, the imaginations of various scribes were sparked to consider how extensive their grief was or to re-create what they might have been saying to one another as they walked home. A few scribes, imagining a more intense reaction, added, “they returned home, beating their breasts and foreheads.” Other scribes provided some dialogue: “they returned home beating their breasts, and saying, ‘Woe to us for the sins we have committed this day, for the destruction of Jerusalem is imminent!’”

Iser calls the textual gaps “blanks;” each blank is a nothing that propels communication because the blank requires an act of ideation in order to be filled. “Blanks suspend connectibility of textual patterns, the resultant break in good continuation intensifies the acts of ideation on the reader’s part, and in this respect the blank functions as an elementary function of communication” (Iser 1978, 189). According to Iser, the central factor in literary communication concerns the reader’s filling in of these textual blanks. His theory of textual gaps is useful for understanding scribal reader-reception. Of course, his perception of gaps or blanks is far bigger and more demanding on the reader’s imaginative powers than was usually the case for New Testament scribes. Nonetheless, scribes were confronted with gaps or blanks that begged for imaginative filling. Many scribes, when confronted with such textual gaps, took the liberty to fill in those gaps by adding extra words or changing the wording to provide what they thought would be a more communicative text. Indeed, the entire history of New Testament textual transmission shows the text getting longer and longer due to textual interpolations—i.e., the filling in of perceived gaps. We especially see the work of gap-filling in the substantial number of expansions in the D-text of the Gospels and Acts. Whoever edited this text had a propensity for filling in textual gaps, as he perceived them. Such gap-filling is especially pronounced in the book of Acts, where the D-reviser made countless interpolations.

RESOURCES

Philip Comfort, Encountering the Manuscripts: An Introduction to New Testament Paleography & Textual Criticism (Nashville, TN: Broadman & Holman, 2005).

Philip W. Comfort, New Testament Text and Translation Commentary: Commentary on the Variant Readings of the Ancient New Testament Manuscripts and How They Relate to the Major English Translations (Carol Stream, IL: Tyndale House Publishers, Inc., 2008)

David Alan Black, New Testament Textual Criticism: A Concise Guide (Grand Rapids, MI: Baker Books, 1994).

Paul D. Wegner, A student’s guide to textual criticism of the Bible: its history, methods, and results, InterVarsity Press, 2006.

SCROLL THROUGH DIFFERENT CATEGORIES BELOW

BIBLE TRANSLATION AND TEXTUAL CRITICISM

BIBLICAL STUDIES / INTERPRETATION

EARLY CHRISTIANITY

CHRISTIAN APOLOGETIC EVANGELISM

TECHNOLOGY

CHRISTIAN THEOLOGY

CHRISTIAN FICTION

[1] Holmes, “New Testament Textual Criticism,” 55.

[2] Holmes, “The Case for Reasoned Eclecticism,” 77–100 in Rethinking New Testament Textual Criticism (editor, Black).

[3] Westcott and Hort, op. cit., 31.

[4] Holmes, “The Case for Reasoned Eclecticism,” 83.

[5] Zuntz, The Text of the Epistles, 283.^[5]

[6] See a good article on this by Elliott, “The Case for Thoroughgoing Eclecticism,” pp. 101–24 in Rethinking New Testament Textual Criticism. Grand Rapids: Baker.

(I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

 

A Uncials (Alexandrinus) most of NT; 5th c.

f Minuscules (a family of manuscripts including 13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709—known as the Ferrar group) Gospels; 11th–15th c.

¹³ Minuscules (a family of manuscripts including 13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709—known as the Ferrar group) Gospels; 11th–15th c.

Maj Minuscules The Majority Text; that is, a group consisting of thousands of minuscules which display a similar text. In the commentary, a few minuscules from this group are occasionally cited on their own: 1110, 1215, 1217, and 1221.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

L Uncials (Regius) Gospels; 8th c.

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

L Uncials (Regius) Gospels; 8th c.

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

L Uncials (Regius) Gospels; 8th c.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

A Uncials (Alexandrinus) most of NT; 5th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

f Minuscules (a family of manuscripts including 13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709—known as the Ferrar group) Gospels; 11th–15th c.

¹³ Minuscules (a family of manuscripts including 13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709—known as the Ferrar group) Gospels; 11th–15th c.

Maj Minuscules The Majority Text; that is, a group consisting of thousands of minuscules which display a similar text. In the commentary, a few minuscules from this group are occasionally cited on their own: 1110, 1215, 1217, and 1221.

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

L Uncials (Regius) Gospels; 8th c.

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

L Uncials (Regius) Gospels; 8th c.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

33 Minuscules All NT except Rev; 9th c.

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

NU the Nestle-Aland/United Bible Society Greek New Testament edition

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

L Uncials (Regius) Gospels; 8th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

Θ Uncials (038) Gospels; 9th c.

0250 Uncials Gospels; 8th c.

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

TCGNT Metzger, Bruce. A Textual Commentary on the Greek New Testament. 2nd ed. New York: United Bible Societies, 1994.

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

33 Minuscules All NT except Rev; 9th c.

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

33 Minuscules All NT except Rev; 9th c.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

NU the Nestle-Aland/United Bible Society Greek New Testament edition

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

א Uncials (Sinaiticus) most of NT; 4th c.

* the original, pre-corrected reading

B Uncials (Vaticanus) most of NT; 4th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

L Uncials (Regius) Gospels; 8th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

א Uncials (Sinaiticus) most of NT; 4th c.

^c corrections made in the manuscript.

A Uncials (Alexandrinus) most of NT; 5th c.

B Uncials (Vaticanus) most of NT; 4th c.

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

א Uncials (Sinaiticus) most of NT; 4th c.

* the original, pre-corrected reading

L Uncials (Regius) Gospels; 8th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

( I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

) I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

L Uncials (Regius) Gospels; 8th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

א Uncials (Sinaiticus) most of NT; 4th c.

* the original, pre-corrected reading

L Uncials (Regius) Gospels; 8th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

TCGNT Metzger, Bruce. A Textual Commentary on the Greek New Testament. 2nd ed. New York: United Bible Societies, 1994.

(I have used my own English translation of the Greek because it is often necessary to be very literal and/or to add material for clarity

 

NU the Nestle-Aland/United Bible Society Greek New Testament edition

B Uncials (Vaticanus) most of NT; 4th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

² corrections made in the manuscript.

F Uncials (Augensis) Paul’s Epistles; 9th c.

G Uncials (Boernerianus) Paul’s Epistles; 9th c.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

NU the Nestle-Aland/United Bible Society Greek New Testament edition

B Uncials (Vaticanus) most of NT; 4th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

F Uncials (Augensis) Paul’s Epistles; 9th c.

G Uncials (Boernerianus) Paul’s Epistles; 9th c.

TCGNT Metzger, Bruce. A Textual Commentary on the Greek New Testament. 2nd ed. New York: United Bible Societies, 1994.

B Uncials (Vaticanus) most of NT; 4th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

F Uncials (Augensis) Paul’s Epistles; 9th c.

G Uncials (Boernerianus) Paul’s Epistles; 9th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Gospels and Acts; early 3rd c.

⁴⁵ Papyri Gospels and Acts; early 3rd c.

𝔓 Papyri Paul’s Major Epistles (less the Pastorals); late 2nd c.

⁴⁶ Papyri Paul’s Major Epistles (less the Pastorals); late 2nd c.

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

L Uncials (Regius) Gospels; 8th c.

T Uncials (Borgianus) Luke, John; 5th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

X Uncials (Monacensis) Gospels; 10th c.

Z (Dublinensis) Matthew; 6th c.

Δ Uncials (037) Gospels; 9th c.

Ξ Uncials (040) Luke; 6th c.

Ψ Uncials (044) Gospels, Acts, Paul’s Epistles; 9th c.

33 Minuscules All NT except Rev; 9th c.

579 Minuscules Gospels; 13th c.

892 Minuscules Gospels; 9th c.

1241 Minuscules Gospels, Acts, Paul’s Epistles; 12th c.

𝔓 Papyri Acts 8; 10; late 4th c.

⁵⁰ Papyri Acts 8; 10; late 4th c.

A Uncials (Alexandrinus) most of NT; 5th c.

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

Ψ Uncials (044) Gospels, Acts, Paul’s Epistles; 9th c.

33 Minuscules All NT except Rev; 9th c.

81 Minuscules Acts, Paul’s Epistles 1044

A Uncials (Alexandrinus) most of NT; 5th c.

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

H Uncials (Coislinianus) Paul’s Epistles; 6th c.

I Uncials (Freerianus or Washington) Paul’s Epistles; 5th c.

Ψ Uncials (044) Gospels, Acts, Paul’s Epistles; 9th c.

33 Minuscules All NT except Rev; 9th c.

81 Minuscules Acts, Paul’s Epistles 1044

1739 Minuscules Acts, Paul’s Epistles; 10th c.

𝔓 Papyri Jas 2–3; 3rd c.

²⁰ Papyri Jas 2–3; 3rd c.

𝔓 Papyri Jas 1; ca. 200

²³ Papyri Jas 1; ca. 200

A Uncials (Alexandrinus) most of NT; 5th c.

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

Ψ Uncials (044) Gospels, Acts, Paul’s Epistles; 9th c.

33 Minuscules All NT except Rev; 9th c.

81 Minuscules Acts, Paul’s Epistles 1044

1739 Minuscules Acts, Paul’s Epistles; 10th c.

A Uncials (Alexandrinus) most of NT; 5th c.

C Uncials (Ephraemi Rescriptus) most of NT with many lacunae; 5th c.

2053 Minuscules Rev; 13th c.

2344 Minuscules Rev; 11th c.

𝔓 Papyri Rev 9–17; 3rd c.

⁴⁷ Papyri Rev 9–17; 3rd c.

א Uncials (Sinaiticus) most of NT; 4th c.

D-text Notes on the Western text

D Uncials (Bezae) Gospels, Acts; 5th c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

0171 Uncials Matt 10; Luke 22; ca. 300

it Old Latin

syr (Syriac Sinaiticus) Gospels; 4th c.

^s (Syriac Sinaiticus) Gospels; 4th c.

syr (Syriac Curetonianus) Gospels; 5th c.

^c (Syriac Curetonianus) Gospels; 5th c.

𝔓 Papyri Acts 18–19; ca. 300

³⁸ Papyri Acts 18–19; ca. 300

𝔓 Papyri Acts 23; 3rd c.

⁴⁸ Papyri Acts 23; 3rd c.

D Uncials (Bezae) Gospels, Acts; 5th c.

syr Syriac This siglum denotes a reading from the margin of syr^h.

^hmg Syriac This siglum denotes a reading from the margin of syr^h.

D Uncials (Bezae) Gospels, Acts; 5th c.

E Uncials (Laudianus 35) Acts; 6th c.

F Uncials (Augensis) Paul’s Epistles; 9th c.

G Uncials (Boernerianus) Paul’s Epistles; 9th c.

𝔓 Papyri Acts 26; 5th c.

¹¹² Papyri Acts 26; 5th c.

𝔓 Papyri Gospels and Acts; early 3rd c.

⁴⁵ Papyri Gospels and Acts; early 3rd c.

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

f Minuscules (a family of manuscripts including 1, 118, 131, 209) Gospels; 12th–14th c.

¹ Minuscules (a family of manuscripts including 1, 118, 131, 209) Gospels; 12th–14th c.

f Minuscules (a family of manuscripts including 13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709—known as the Ferrar group) Gospels; 11th–15th c.

¹³ Minuscules (a family of manuscripts including 13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709—known as the Ferrar group) Gospels; 11th–15th c.

Θ Uncials (038) Gospels; 9th c.

565 Minuscules Gospels; 9th c.

700 Minuscules Gospels; 11th c.

A Uncials (Alexandrinus) most of NT; 5th c.

E Uncials (Laudianus 35) Acts; 6th c.

F Uncials (Augensis) Paul’s Epistles; 9th c.

G Uncials (Boernerianus) Paul’s Epistles; 9th c.

H Uncials (Coislinianus) Paul’s Epistles; 6th c.

K (Cyprius) Gospels; 9th c.

P Uncials (Porphyrianus) Acts–Revelation; 9th c.

S Uncials (Vaticanus 354) Gospels; 949

W Uncials (Washingtonianus or the Freer Gospels) Gospels; 5th c.

Π Uncials (041) Gospels; 9th c.

Ψ Uncials (044) Gospels, Acts, Paul’s Epistles; 9th c.

H Uncials (Coislinianus) Paul’s Epistles; 6th c.

L Uncials (Regius) Gospels; 8th c.

P Uncials (Porphyrianus) Acts–Revelation; 9th c.

L Uncials (Regius) Gospels; 8th c.

B Uncials (Vaticanus) most of NT; 4th c.

א Uncials (Sinaiticus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Gospels and Acts; early 3rd c.

⁴⁵ Papyri Gospels and Acts; early 3rd c.

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri 1-2 Peter, Jude; ca. 300

⁷² Papyri 1-2 Peter, Jude; ca. 300

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Paul’s Major Epistles (less the Pastorals); late 2nd c.

⁴⁶ Papyri Paul’s Major Epistles (less the Pastorals); late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri John; late 2nd c.

⁶⁶ Papyri John; late 2nd c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

B Uncials (Vaticanus) most of NT; 4th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

D Uncials (Bezae) Gospels, Acts; 5th c.

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

B Uncials (Vaticanus) most of NT; 4th c.

NU the Nestle-Aland/United Bible Society Greek New Testament edition

NU text of Nestle-Aland 26th/27th [N] and the United Bible Societies 3rd/4th [U]

TCGNT A Textual Commentary on the Greek New Testament (2nd ed., Metzger) [1994]

NU text of Nestle-Aland 26th/27th [N] and the United Bible Societies 3rd/4th [U]

⁴⁸ Epp, “The Significance of the Papyri for Determining the Nature of the New Testament Text in the Second Century: A Dynamic View of Textual Transmission,” 71–103.

³⁷ Colwell, op. cit., 106–24.

³⁸ James Royse, “Scribal Habits in Early Greek New Testament Papyri.”

³⁹ Jauss, “Literary History as a Challenge to Literary Theory,” 7–37.

⁴⁰ Colwell and Tune, “Variant Readings: Classification and Use,” 259–61.

⁴¹ Epp and Fee, Studies in the Theory and Method of New Testament Textual Criticism, 50–57.

⁴² Royse, “Scribal Habits in Early Greek New Testament Papyri,” 45–46.

⁴³ Colwell, Studies in Methodology in Textual Criticism of the New Testament, 123.

⁴⁴ Ibid., 108.

⁴⁵ Ibid., 114.

⁴⁶ Royse, op. cit., 156–57, 423, 560.

⁴⁷ This was the goal of my dissertation, specifically for P⁴⁵, P⁶⁶, and P⁷⁵ (Gospel papyri). See Comfort, “The Scribe as Interpreter: A New Look at New Testament Textual Criticism according to Reader Response Theory” (1997).

𝔓 Papyri Luke and John; ca. 200

⁷⁵ Papyri Luke and John; ca. 200

D-text Notes on the Western text