The Textual Criticism of the New Testament

The Textual Criticism of the New Testament 

Gordon D. Fee
B.A., M.A., Seattle Pacific College; Ph.D., University of Southern California
Associate Professor of New Testament, Gordon-Conwell Theological Seminary

I.    Introduction

Textual criticism, commonly known in the past as “lower” criticism in contrast to the so-called “higher” (historical and literary) criticism, is the science that compares all known manuscripts of a given work in an effort to trace the history of variations within the text so as to discover its original form. Textual criticism is, therefore, of special significance to the interpreter in at least three ways: (1) It helps to determine the authentic words of an author. The first question the exegete asks is, What does the text say? before he asks, What does it mean? (2) The majority of Christians have access to the NT only in translation, and the basic consideration in choosing a translation is its accuracy in representing the original text of the author. A translator’s first concern must be that he is translating the actual words of the author before he decides what those words mean. (3) A knowledge of the history of textual variation will also help the interpreter to see how a passage was understood during the early history of the church. In many instances variant readings are a reflection of a scribe’s or a church’s theological interests, and sometimes such changes put one in direct contact with historical exegesis.

II.  The Need

The need for NT textual criticism results from a combination of three factors: (1) The originals, probably written on papyrus scrolls, have all perished. (2) For over 1,400 years the NT was copied by hand, and the copyists (scribes) made every conceivable error, as well as at times intentionally altering (probably with the idea of “correcting”) the text. Such errors and alterations survived in various ways, with a basic tendency to accumulate (scribes seldom left anything out, lest they omit something inspired). (3) There are now extant, in whole or in part, 5,338 Greek MSS, as well as hundreds of copies of ancient translations (not counting over 8,000 copies of the Latin Vulgate), plus the evidence from the citations of the NT in the writings of the early church fathers. Moreover, no two MSS anywhere in existence are exactly alike.

The task of the textual critic, therefore, is (1) to sift through all this material, carefully collating (comparing) each MS with all the others, in order (2) to detect the errors and changes in the text, and thus (3) to decide which variant reading at any given point is more likely to be the original.

III. The Sources

The sources for finding the original text are the Greek MSS, the ancient versions, and the citations by the early fathers. Although many of the extant MSS (both Greek and versional) are fragmentary and the majority do not contain the whole NT, there is such a quantity of material that even the most poorly attested NT book, the Book of Revelation, has been preserved in over three hundred Greek MSS, while the Gospels are extant in thousands of copies.

A.   The Greek Manuscripts

Primacy of position in the quest for the original text belongs to the Greek MSS, partly because they are copies of copies in the original language of the biblical authors, and partly because the oldest ones are generally earlier than the other evidence (though age is no guarantee of better quality). The MSS are of four kinds: papyri, uncials, minuscules, and lectionaries.

The original documents of the NT were probably written on papyrus scrolls. The scroll, however, was cumbersome both for reading and for finding specific passages. As a result, Christians very early began to use the codex, or leaf-form of book, to copy their sacred writings. All extant fragments and copies of the NT, therefore, are codices; no copies on scrolls have ever been discovered.¹

The book form also allowed Christians to include more than one document in a single codex, though it was not until the development of the canon and the emergence of large parchment codices (4th century a.d.) that copies of the entire NT were made.

1. The Papyri. The earliest codices were written on papyrus leaves in uncial (capital letter) script, with no separation of words and little or no punctuation. Because papyrus is naturally perishable, few of the early copies have survived except in the dry sands of Egypt. So far, fragments or larger sections of eighty-five different papyrus MSS have been discovered. These range in date from approximately a.d. 125 P⁵² a single small fragment of John 18:31–34, 37–38) to the eighth century (P⁴¹, P⁶¹), though the majority belong to the third and fourth centuries. Every NT book except 2 Timothy is represent ed in these MSS. Several of the papyri are well preserved and present the earliest significant witness to the NT text. For example, P⁴⁵ (c. a.d. 250) has substantial sections of the synoptic Gospels, P⁷⁵ (c. a.d. 200) contains more than half of Luke and John, P⁶⁶ (c. a.d. 200) about two-thirds of John, P⁴⁶ (c. a.d. 225) substantial portions of Paul’s letters, P⁷² (c. a.d. 275?) large sections of Jude and 1 and 2 Peter, and P⁴⁷ (c. a.d. 280) about one-half of Revelation.

2. The Uncials. About the beginning of the fourth century, vellum (or parchment) began to replace papyrus as the primary writing material. These prepared animal skins had the advantage both of greater durability and larger size, so that from the sixth century to the fourteenth almost all literary efforts of all kinds were written on parchment.

The scribes of the earlier of these codices (from the fourth to the ninth century) continued to use the uncial script. There are currently 268 known uncials, many of them preserved without blemish. Only one, however, Codex Sinaiticus (א, c. a.d. 350), preserves the entire NT. (It also contains the Epistle of Barnabas and the Shepherd of Hermas.) The great Codex Vaticanus (B, c. a.d. 325) includes everything except Hebrews 9:14–13:25 and Revelation, while the majority contain NT sections, such as the Gospels or the Pauline letters. These MSS are designated in two ways: by capital letter and by Arabic numeral with a zero prefixed. The earlier known MSS have two designations (D-05), while the later ones simply have the number (0268).

3. The Minuscules. At the beginning of the ninth century a script of small letters in a running hand (called “minuscule” or “cursive”), which stands in contrast with the uncial (capital letter) script, was created. The advantages of minuscule texts both in speed and economy were quickly recognized, so that by the end of the tenth century, uncial texts were no longer produced. The vast majority—2,792 to date—of extant MSS are these late minuscules. They are designated by Arabic numerals from 1 to 2,792.

4. Lectionaries. The second largest group of MSS of the NT are the lectionaries. These are texts written, not in regular sequence, but in accordance with the designated daily and weekly lessons from the Gospels and Epistles—lessons that had been developed in very early times.

There are presently 2,193 known lectionary MSS, the earliest fragments dating from the sixth century and complete MSS from the eighth. They are, therefore; both uncial and minuscule and contain either the Gospels or Epistles, or sometimes both. The lectionaries are designated by Arabic numerals prefixed with an italicized or cursive l(l2193).

B.   The Versions

Because of the broad missionary outreach of the early church, copies of most of the NT documents had been translated by the end of the second century into Latin, Syriac and Coptic. In the following centuries other translations followed: Gothic, Armenian, Georgian, Ethiopic, Slavonic, and Arabic.

Because the Old Latin, Old Syriac, and Coptic versions were made very early and because their geographical location is fairly well fixed, they are particularly important in the recovery of the original NT text. Their use, however, is complicated by several factors. In the first place, certain features of Greek syntax and vocabulary are difficult or impossible to convey in translation. One can never be certain, therefore, what their Greek text looked like. For example, Latin has no definite article and the Syriac cannot distinguish between the Greek aorist and perfect tenses. Furthermore, it is highly probable that more than one translation was made in each of these languages by different persons, in different places, using different Greek texts. Finally, the earliest extant MSS of these versions are copies nearly two hundred years later than the original translation. Consequently they have very likely suffered their own fate of textual corruption.

In spite of these complications, however, the ancient versions are a valuable source not only in the quest for the original text itself, but also in the attempt to trace the history of textual transmission and corruption. These older versions are variously designated: some are identified by small Latin letters (a, b, c, or it^a, it^b, etc.) for the OL, while the others are identified by a superscript designation after an abbreviated form of the version (syr^c syr^pal cop^bo).

The later versions and the “authorized revisions” of the older versions, viz., the Vul. and the Syriac Peshitta, are of more limited significance. Scholars, of course, make use of all evidence. But the bewilderingly complicated history of the Vul., which makes it a textual study in its own right, tends to give it a place of secondary importance even among the versions.

C.   Patristic Citations

The final source of data for the textual critic is from the citations and allusions to the NT found in the writings of the early church fathers. As with the versions, their usefulness is complicated by several mitigating factors.

Most often the fathers cited the NT from memory, so one can never be sure that their memory reflects the actual wording of their Greek text. Moreover, a father may have used several—and differing—copies of the NT. Finally, the available texts of the patristic writings also are copies, usually very late ones, and in some cases have suffered extensive corruption.

Yet when the painstaking work of reconstructing the NT text cited by one of the fathers is done, it is of great value. For it gives us a datable and geographically identifiable witness to the NT available to that particular father. Although such a witness is generally tertiary to the Greek MSS and the versions in the recovery of the original text, it is of primary importance in tracing the history of textual transmission.

D.   Manuscript Relationships

The immense amount of material available to the NT textual critic, exceeding all other ancient documents by hundreds of times, is both his good fortune and his problem. It is his good fortune because with such an abundance of material he can be reasonably certain that the original text is to be found somewhere in it. Quite in contrast to those searching for other original texts (including the OT), he scarcely ever needs to resort to textual emendation, though the possibility must always be kept open that the very first copy of the original MS, from which all others derived, had some uncorrected errors.

However, the abundance of material is likewise the textual critic’s problem, because no two copies are exactly alike, and the greater the number of copies, the greater the number of variants among them. Even in this day of computer technology, sifting through such an immense amount of material is a formidable task. This is especially so in light of the ideal that each piece of evidence must be used in order to identify the original by detecting possible corruption of the NT text.

The task, however, is not quite so formidable as it might at first appear. Although it is true that no two MSS are identical, it is equally true that many are so much alike that they tend to group themselves into three (some textual critics think four) major families of texts (text-types). Such text-types are identifiable on the basis of (1) the percentage of agreement certain MSS have with one another over a total area of variation and (2) the amount of agreement these MSS have in variant readings peculiar to them.

There is, first of all, a group of MSS that have all the appearances of being “local” texts, since they derive basically from Alexandria in Egypt. It is headed by P⁷⁵ and P⁶⁶ (c. A.D 200) in the Gospels, P⁴⁶ (c. 225) in Paul, P⁷² (c. 275?) in Peter and Jude, Codex B (c. 325), and the citations of Origen (225–250). It is also supported to a lesser degree by several other MSS (e.g., X C L W 33) and the later Alexandrian Fathers (Didymus, Athanasius, Cyril).

For many years textual critics have considered this text-type to be a carefully edited recension of the third century, created by the best Alexandrian scholarship on the basis of good ancient MSS. But the combined evidence of P⁷⁵, P⁷², P⁴⁶, and Origen has placed this text in all of its particulars squarely in the second century, or, so it seems, as early as Christianity was known in that city.

Although this text-type has occasional “sophisticated” variants, it commonly contains readings that are terse, somewhat rough, less harmonized, and generally “more difficult” than those of other text-types, though on closer study they regularly commend them selves as original. Furthermore, it is consistently so across all the NT books, with a minimal tendency to harmonize an author’s idiosyncrasies with more common Greek patterns. All these facts give the impression that this text-type is the product of a carefully preserved transmission.

A second group, equally as early as the Alexandrian, is commonly called “Western,” because variants peculiar to it are firmly established in texts found in North Africa (Tertullian, Cyprian, some OL), Italy (Novatian, some OL), and southern France (Irenaeus). “Western,” however, is something of a misnomer, for many of the peculiar variants of this text-type are also found in the East (Tatian and the Old Syriac) and occasionally in Alexandria (some quotations in Clement, in John 6–7 in P⁶⁶, in John 1–8 in א, and in Mark 1–5 in W).

In spite of this early and wide attestation to such a text, these various witnesses lack the homogeneity found in the Alexandrian and later Byzantine witnesses. The textual relationships are not consistently sustained over large portions of text. On the contrary, “Western” describes a group of MSS headed by Codex D, obviously related by hundreds of unusual readings, sometimes found in one or several, sometimes in others, but apparently reflecting an uncontrolled, sometimes “wild,” tradition of copying and translating. This text-type is particularly marked by some long paraphrases and long additions, as well as by harmonistic tendencies and substitutions of synonyms. In fact, the Western text of Acts is about 10 percent longer than other texts and almost certainly reflects an early revision.

One must be careful, however, not to dismiss a variant reading out of hand simply because it is Western. There are several instances, especially in some striking “omissions” but in other places as well, where scholars have cogently argued that the Western text preserves the original NT text. Moreover, the very antiquity of this text, and its wide distribution, should always gain for it a full hearing.

The third text-type, the “Byzantine” or “majority” text, is made up of over 80 percent of all the MSS. As a text-type it does not appear in history until about a.d. 350, but even then its origins are shrouded in mystery. Readings peculiar to this text first appear in a group of writers associated with the church of Antioch: the Cappadocians, Chrysostom, and Theodoret of Cyrus. These fathers had a NT about 90 percent along the way to the full Byzantine text of the Middle Ages. The earliest MS to reflect this text is from Alexandria (Codex A; c. 475—in the Gospels only), while the earliest full witnesses to it are MSS from the eighth century (E and Ω).

Does this text, therefore, represent a revision effected in Antioch in the fourth century? Most textual critics think so, but they do so on the basis of the secondary nature of its peculiar readings, not because of firm data. There are no early MSS from Asia Minor or Palestine. The earliest writers from these parts reflect a Western text, but there was no Origen or Tertullian in Antioch in the early third century to give us a large amount of data to study. Later in the century the scanty evidence from Methodius of Lycia and Tyre and, still later, from the text of Eusebius of Caesarea and Cyril of Jerusalem seldom reflects the peculiarities of this text-type. Thus the nature of the text in Antioch over many years is virtually unknown.

What is known is that such a text was available by a.d. 350, that it had partially begun to influence the text of Alexandria and Rome Uerome), that it was carried by Chrysostom from Antioch to Constantinople, and that probably through his influence it became the dominant text in the Eastern church.

Most of the readings peculiar to this text are generally recognized to be of a secondary nature. A great number of them smooth out grammar; remove ambiguity in word order; add nouns, pronouns, and prepositional phrases; and harmonize one passage with an other. Its many conflate readings (e.g., Mark 9:49), where the Byzantine text-type combines the alternative variants of the Alexandrian and Western texts, also reflect this secondary process.

Some scholars also find a “Caesarean” text-type in the Gospels, supported sometimes by P⁴⁵, W, Θ. family 1, family 13 and the citations of Origen (in Mark), Eusebius, and Cyril of Jerusalem. There is indeed some obvious textual relatedness among these witnesses (especially in Mark), but whether they constitute a separate text-type, rather than some unusual mixtures of the other three, remains doubtful.

Although there is general agreement that making such groupings is both a possible and a necessary task, the significance of such groupings remains contested. It is surely dubious procedure to accept or reject a reading solely because it is found in a certain text-type; on the other hand, such groupings, especially of the later (Byzantine) MSS, greatly reduce the work of sifting a multiplicity of MSS.

IV.      The Text in History

In order to understand the “how” of NT textual criticism, it is necessary to understand something of the history of the transmission of the text, as well as to have some knowledge of the history of textual criticism itself.

A.   Period of Confusion (to a.d. 400)

The vast majority of the errors in the NT MSS occurred during the period that is also the most difficult to reconstruct—the first four Christian centuries.

Much of the difficulty stems from the work of the earliest Christian copyists. In a time when the majority of people were illiterate and when Christianity periodically under went severe persecution, there were probably few professionally trained scribes in the service of the church. Moreover, seldom were the scribes possessed by the spirit of the scribes of later times who worked according to the instructions of the Lord given in Deuteronomy 12:32: “Thou shalt not add thereto, nor diminish therefrom.” In fact, the opposite seems to have been true of the scribes in the first two centuries. They introduced thousands of changes into the text. To be sure, the majority of their errors were unintentional and are easily discernible slips of the eye, ear, or mind. Hundreds of changes in the text were, however, made intentionally. Yet we should not think of these scribes as having acted from evil motives. If they often took many liberties in copying their texts, apparently they did so in most cases in an attempt to “help out.” They were more interested in making the message of the sacred text clear than in transmitting errorless MSS.

Thus, early scribes (and sometimes later ones) often “smoothed out” the Greek of the biblical writer by adding conjunctions, changing tenses of verbs, and changing word order. They also tended to clarify ambiguous passages by adding nouns or pronouns, by substituting common synonyms for uncommon words, and sometimes even by rewriting difficult phrases. One of the most common causes of error was the tendency to conform one passage to another. This harmonizing tendency is particularly frequent in the Gospels. It also occurs in parallel passages in Paul and Acts. There are also some instances—and these are usually very important ones—where scribes have added (or less often subtracted) whole sentences or narratives in the interest of doctrine or completeness.

During the second century in particular, when each NT book was being transmitted independently of the others and when there was wide geographical distribution of these documents with little or no “controls,” such scribal errors proliferated. Once an error was introduced into the text, it was then copied by the next scribe as his “received” text. Quite often a scribe “corrected” what he thought to be errors and in doing so created errors of his own. If, as did the scribe of P⁶⁶, he had a chance to check his copy against another, he may then have “corrected” his text by adding still other variants from that copy. So errors were created and compounded and so they tended to accumulate.

B.   Period of Transmission (400–1516)

Two significant events affected the history of the NT after a.d. 400. The Alexandrian text, which by 450 was already greatly influenced by the Byzantine, generally disappeared from use. The major causes for this was the demise of the patriarchate in Alexandria and the subsequent rise and spread of Islam.

On the other hand, Latin had meanwhile become the predominant language in the West, so that production of Greek texts ceased there. The great number of discrepancies found in the OL MSS had finally resulted in an “authorized” translation, the Latin Vulgate, made by Jerome c. 384. But it took about two hundred years before it superseded the more popular older translations. Meanwhile, as it was being copied and carried from one part of the West to another, the Vul. was variously conformed to the OL and developed local textual histories. Several attempts were made throughout the Middle Ages to purify Jerome’s text, but each of these recensions eventually resulted in further corruption. As a result, the over 8,000 extant Vul. MSS reflect an enormous cross contamination of text-types.

The result of these two factors was that the transmission of the Greek NT was generally limited to the Eastern church, where the majority of copies reflected the standardized text used at the capital, Constantinople. Thus the history of the Greek text during this period, with a few notable exceptions, is simply the history of a thousand years of copying MSS of the Byzantine text-type.

C.   Establishment of the Textus Receptus (1516–1633)

Johannes Gutenberg’s invention of printing by use of movable type was the next major factor in the history of the NT text. For now many copies of a book, all identical, could be produced. Although the first Greek NT actually to be printed was edited by Cardinal Ximenes in 1514, the first text to be published appeared in 1516 and was edited by the great Dutch humanist, Erasmus.

Unfortunately, these first editions, which were to serve as a base for all subsequent editions until 1831, were themselves based on late medieval MSS of inferior quality. In fact, Erasmus’s only MS of Revelation lacked the final leaf, which had contained the last six verses. For these verses Erasmus used the Vul., translating its text into Greek, with the result that his Greek text has readings that have never been found in any Greek MS.

Of the subsequent editions, three have special significance for the history of the NT text: (1) Robert Stephanus’s third edition (1550), which was based on Erasmus’s third edition, became the standard text in England and served as the base for the KJV of 1611. His fourth edition (1551) is also noteworthy in that it is the first text to be divided into numbered chapters and verses—the system still in use today.

(2) Theodore Beza, John Calvin’s successor in Geneva, published nine editions between 1565 and 1604, and this tended to stamp an imprimatur on the text of Erasmus. His editions of 1588–9 and 1598 were also used by the King James translators.

(3) A Greek text very much like those of Erasmus, Stephanus, and Beza, edited by Bonaventure and Abraham Elzevir (1633), became the standard text used on the continent. The term Textus Receptus (TR = “received text”) derives from the preface of this edition, in which the editors declared, “You therefore have the text which is now received by all, in which we give nothing altered or corrupted.” This boast was to hold good for over two hundred more years.

D.   Period of Discovery and Research (1633–1831)

The next period in the history of the NT text was one in which scholars made great efforts to amass new information from Greek MSS, the versions, and the fathers. Yet the texts published during this period continued to print the time-honored TR; the new evidence, especially that from much earlier MSS, was relegated to variant readings in the apparatus (i.e., the critical notes). Among the large number of scholars who made contributions during this period, especially noteworthy are J.A. Bengel (1734), who was the first to suggest a classification of MSS into text-types and to devise a system of evaluating variants according to merit; J.J. Wetstein (1751–2), who set forth extensive principles of textual criticism and began the device of designating MSS by symbols; and J.J. Griesbach, whose editions from 1774 to 1807 laid the foundation for all subsequent textual criticism. Griesbach modified Bengel’s classifications of textual groups into the basic three, which are still recognized. He elaborated and carefully defined the principles of textual criticism and showed great skill in evaluating the evidence for variant readings. Although his own text was not so divergent from the TR as those that would follow, his pioneer efforts paved the way for what was to come.

E.   Period of Constructive Criticism (1831–1881)

The period that followed Griesbach was to see the overthrow of the TR and the rise of new critical editions based on the more significant MS finds and the principles of criticism pioneered by Wetstein and Griesbach …

The first important break from the TR came in 1831 with the Greek text published by the German classicist Karl Lachmann. His was the first systematic attempt to produce a text using a scientific method rather than the mere reproduction of the text of the Middle Ages.

More significant still was the voluminous and monumental work of Constantine von Tischendorf. Besides bringing to light many hitherto unknown MSS, he published eight critical editions of the Greek NT, the last of which (1872) contained a critical apparatus giving all the variant readings of the known uncials as well as reading for many cursives, the versions, and the church fathers. This volume is still an indispensable tool for NT textual criticism.

Although many others made contributions during this period (especially S.P. Tregelles), the Greek text edited by B.F. Westcott and F.J.A. Hort (WH 1881) was to supersede all others in significance. So thoroughly and well did they do their work that almost all subsequent textual criticism is defined in relationship to it. Their forte was the refinement and rigorous application of a scientific methodology to the NT text. The result was issued in two volumes as The New Testament in the Original Greek. Volume 1 contained their resultant Greek text; volume 2 comprised a lengthy Introduction, written by Hort, and an Appendix, in which certain problem passages were discussed.

In the Introduction Hort set out in full detail what has become a classic statement of the methodology of textual criticism. Especially significant are his careful analyses and evaluations of the relative merits of the various text-types and their leading representatives. Above everything else, Hort forever laid to rest the TR. He offered three main arguments against the Byzantine text-type (he called it Syrian), which subsequent discoveries and researches have generally validated: (1) The Syrian text-type is filled with conflate readings, i.e., readings that combine the elements found in the earlier two text-types; (2) the readings peculiar to the Syrian text-type are never found in the ante-Nicene Fathers, neither East nor West; and (3) when the readings peculiar to this text-type are compared with rival readings on the principles of internal evidence, “their claim to be regarded as the original readings is found gradually to diminish, and at last to disappear” (Introduction, p. 116).

Westcott and Hort were thus left with a choice between the two earlier text-types. At this point internal considerations became the final arbiter, and they felt that a careful analysis of variants over many pages of text revealed the text of Egypt, or Alexandria, (which they presumed to call “Neutral”) to be far superior in almost every case. Thus their resultant text was an edition of the Neutral text-type, except in those instances where internal evidence was clearly against it.

F.    Since Westcott and Hort (1881 to present)

As one might expect, such a radical departure from the “received text” was not immediately accepted by all. This is particularly true of the English-speaking world, where the TR had long been in the hands of the majority of Christians through the KJV. The reaction to WH was led especially by J.W. Burgon, Edward Miller, and H.C. Hoskier. Unfortunately, much of the reaction, especially that of Burgon, took the form of rhetoric rather than argument; and what argument one does find is basically theological and speculative, but seldom supported by the actual textual data.

This is not to suggest that all subsequent scholarship has followed WH. Most scholars found their affirmation of the Alexandrian MSS as neutral to be too ambitious. In spite of such disavowals, however, all subsequent critical texts look far more like WH than like the TR or the Western MSS. Therefore, it is fair to say that, whether intentionally or not, the mainstream of NT textual criticism since WH has moved toward modifying and advancing their work. In this brief survey it is possible to sketch only some of the more important advances.

1. New Discoveries. Probably the most important advance since WH is the discovery of large quantities of new textual data of all kinds. Among these, the most significant are the papyri, because for the most part they represent evidence earlier than that available to Westcott and Hort.

Many of the first discoveries of earlier evidence showed such a textual mixture that Westcott and Hort’s theories of text-types were seriously called into question. But later discoveries, especially P⁴⁶, P⁷², and P⁷⁵, have tended to verify the basic positions of Westcott and Hort. Furthermore, the papyri have generally confirmed their opinion as to the late character of the Byzantine text-type. One does find an occasional variant in the early papyri that supports the later text-type, but none of the early papyri is even remotely related to the Byzantine MSS.

2. Other Researches. Besides the discovery of new MSS, other researches of various kinds have also greatly advanced the science of textual criticism since WH.

Especially noteworthy has been the work done that sheds more light on the versions and on Tatian’s Diatessaron (an arrangement of the four Gospels to form a single narrative) and the collecting and editing of the citations of the early fathers. The usefulness of this work is now far greater than in 1881.

In recent years, methodology in establishing textual relationships has also been greatly improved, not only for text-types in general but also for clearer definition of relationships within the great mass of Byzantine MSS. Such refinements of method have greatly increased the ability of textual critics to group MSS into their proper families and text-types.

Of particular interest to the exegete has been the work of such scholars as C.b.c. Williams and E.J. Epp, who have studied the theological tendencies of certain groups of variants. Such studies have made clear that not all textual variation is accidental or theologically unbiased. They further aid the exegete by throwing light on how certain passages were understood, or misunderstood, in the early church.

Two projects of large dimensions involving broad international cooperation are also of interest both to the scholar and to the interpreter: (1) The International Greek New Testament Project, composed of a team of American and British scholars, is preparing a critical apparatus of the Gospels that will include all known papyri and uncials, extensively representative cursives and lectionaries, all early versions, and citations of all church fathers to a.d. 500. (2) A team of German and French scholars, under the auspices of the Institut fur Neutestamentliche Textforschung in Munster, is at work on a new major critical edition, including a full critical apparatus. The general Epistles are the first scheduled for publication.

3. Critical editions. These discoveries and researches have resulted in a spate of critical texts since WH. A few should be noted because of their broad significance.

In 1913 H. von Soden published a long-awaited and massive work that included a critical text, a large and complicated apparatus, lengthy descriptions of MSS, and his own textual theory. This work, however, turned out to be a great disappointment. His textual theory never gained acceptance, his classifications of MSS have often proved to be wrong, and some of his collations are completely untrustworthy. Nevertheless, his ac cumulation of evidence goes beyond that of Tischendorf and is helpful to the expert when used with care.

More important to most exegetes are the smaller “pocket” editions. The most common of these is a series of editions begun by Eberhard Nestle in 1898. A twenty-fifth edition of this text was published in 1963, now under the supervision of Kurt Aland. This text was not a new critical text, but was rather based on the majority reading of the critical texts of Tischendorf, Westcott and Hort, and B. Weiss. The great usefulness of this edition has been its extensive, but abbreviated, textual apparatus.

In 1966 the United Bible Societies published a new “handbook” edition, edited by K. Aland, M. Black, B.M. Metzger, and A. Wikgren (C. Martini was added to the editorial board for the second edition [1968]). This text has been prepared especially for Bible translators and therefore has the following distinctives: (1) The critical apparatus is restricted primarily to meaningful variants, i.e., variants that may make a difference in the translation of the text; (2) each variant adopted in the text is given a notation as to the degree of certainty the editors felt it had; (3) each variant has a full citation of carefully selected representative evidence; and (4) there is a second apparatus giving meaningful alternatives in punctuation. A commentary on each variant, written by Metzger, was published in 1973.

A comparison of this text with WH and TR shows where a significant consensus of modern scholarship stands. For example, in Luke 10 the UBS edition varies from WH only eight times (plus six spelling differences), while it differs from the TR fifty-six times (plus twenty spelling differences). The reason for the differences between WH and the UBS, or among any of the modern critical texts, is fundamentally a matter of emphasis in methodology.

V.  The Method

For a full discussion of the method and practice of NT textual criticism one should consult the manuals by Greenlee or Metzger. Certain basic considerations may be noted here.

One criterion above all others superintends the scholar’s choice at any point of textual variation: the variant that best explains the origin of all the others is most likely to be original. In order to “best explain the origin of the others,” there are two factors that scholars must consider: external evidence (the MSS themselves) and internal evidence (having to do with the authors or scribes).

A.   External Evidence

The first thing one must do at any point of variation is to weigh the MS evidence supporting each variant. Thus one usually asks the following questions: How old are the witnesses supporting each variant or how old is their text? How good is the general quality of the MSS? How wide is the geographical distribution of the witnesses? This latter question is especially important, because early and widespread geographical distribution of a reading points to an original parent much further back before the document in question was widely scattered throughout the early church. With few exceptions, however, scholars are agreed that knowing the age or the geographical distribution of early witnesses in no way guarantees finding the original text.

B.   Internal Evidence

Internal evidence is of two kinds: transcriptional probability (what kind of error or change the scribe probably made) and intrinsic probability (what the author was most likely to have written).

1. Transcriptional probability has to do with scribal errors and is based on certain inductively derived criteria. For example, it is usually true that the more difficult reading is probably the original one, because it was the tendency of scribes to make the text easier to read. Again, the shorter reading is often the original one, because the scribes tended to add to the text. This criterion must, however, be used with great caution because scribes sometimes made omissions in the text either for smoothness or to remove what might be objectionable. Finally, a textual variant differing from quoted or parallel material is almost always original, since the tendency of scribes was to harmonize.

2. Intrinsic probability is the most subjective element in the methodology of textual criticism. It has to do with the style and vocabulary of the author, his ideas as they are elsewhere known, and the probabilities based on the immediate context.

Not all the criteria mentioned above are equally applicable in every case; in fact, in some instances they oppose one another. For example, the longer reading may be the more difficult one, or the reading most in accord with author’s style may be a harmonization with that style. In such stalemates the textual critic is usually forced back to the external evidence as final arbiter.

It is noteworthy that for most scholars over 90 percent of all the variations to the NT text are resolved, because in most instances the variant that best explains the origin of the others is also supported by the earliest and best witnesses.

C.   The Debate Over Method

With the rejection of Hort’s genealogical method, by which the reading of the Alexandrian witnesses was adopted except where internal evidence proved it secondary, there has emerged a method that may properly be called “eclectic.” Essentially, this means that the “original” text of the NT is to be chosen variant by variant, using all the principles of critical judgment without regarding one MS or text-type as necessarily preserving that “original.”

Despite a few notable exceptions, most of the differences that remain among critical texts result from a varying degree of weight given the external evidence.

On the one hand, there is a kind of eclecticism that, when all other criteria are equal, tends to follow Hort and to adopt the readings of the Alexandrian witnesses. This may be observed to a greater degree in the UBS edition and to a somewhat lesser degree in the Greek texts behind RSV and NEB, where early Western witnesses are given a little more consideration.

Another kind of textual theory was advocated by M.E. Boismard and was used in D. Mollat’s translation of John in the Jerusalem Bible. This is a kind of “eclectic Western” method in which great emphasis is placed on preference for the shorter readings as they are found in various Western witnesses, especially early versions and citations from certain fathers. The difficulty with this method seems to lie in the preference for the versions and fathers over against the whole Greek tradition, especially since many shorter readings may be shown to be translational paraphrase or untrustworthy citations apparently made from memory.

On the opposite side is the method of “rigorous eclecticism” practiced by G.D. Kilpatrick and his student J.K. Elliott. They advocate placing no weight on the MSS at all, but making every choice solely on the basis of internal principles. The difficulty with this method is that the results depend on the scholar’s preference of internal criteria, which in the case of Kilpatrick and Elliott seems to be for variants in an author’s style as over against the questions of transcriptional probability.

While, as has already been said, we may grant that not all of the principles of textual criticism are applicable to each variant, contemporary critics generally agree that questions of internal evidence should usually be asked first and that the weight of the MS evidence should be applied secondarily. What becomes obvious, however, is that on the grounds of internal evidence certain MSS tend to support the “original” text more often than others and that those MSS are the early Alexandrian. Therefore, when internal evidence cannot decide, the safest guide is to go with the “best” MSS.

VI.      The Significance

What difference does all of this make to the expositor? Much in every way. On the one hand, it provides him with confidence that for the most part the text he is interpreting, whether it be from a modern Greek text or a contemporary translation, truly represents what the biblical author actually wrote.

Nevertheless, and more significantly, there are places where the original text is not so certain. At such points textual criticism becomes an integral part of exegesis. In some instances, such as in John 7:1, whether the original text says that Jesus “did not wish” to go about in Galilee or “did not have the authority” to do so, or as in v.8, whether Jesus said he was not, or was not yet, going up to the feast, the textual choice will affect the interpretation of the passage.

In other instances, exegesis and textual choice go hand in hand. In John 1:34, did John the Baptist say, “This is the Son of God” (KJV, RSV) or “This is God’s Chosen One” (NEB, JB)? The MS evidence is divided, even among the early text-types. “Son” is found in the key Alexandrian witnesses (P⁶⁶ P⁷⁵ B C L cop^bo) as well as in several OL (aur c f i g) and the later Syriac witnesses, while “Chosen One” is supported by the Alexandrians P⁵ א Cop^sa as well as the OL MSS a b e ff² and the Old Syriac.

The question must finally be decided on internal grounds. As to transcriptional probability, one thing is clear: the variant is intentional, not accidental. But did a second century scribe alter the text to support a kind of adoptionist Christology, or did an orthodox scribe sense the possibility that the designation “Chosen One” might be used to support adoptionism, and so alter it for orthodox reasons? In terms of probabilities, the latter seems far more likely, especially since “the Son” is not changed elsewhere in the Gospel to fit adoptionist views.

But the final decision must involve exegesis. Since what John the Baptist said was almost certainly intended to be messianic and not a statement of Christian theology, the question is whether it reflects the messianism of such a passage as Psalm 2:7 or that of Isaiah 42:1. In light of the suffering, or paschal, lamb motif of John 1:29, it is surely arguable that “Chosen One” fits the context of the Gospel.

What finally points to “Chosen One” as original is the use the evangelist makes of the many confessions in the Gospel. All of them pick up different messianic motifs (1:29, 41, 49; 4:42; 6:14; 6:69; 11:27) and all of them “fit” their specific context (e.g., the “true Israelite” confesses him as “King of the Jews”; in the bread [manna] from heaven context he is called the Mosaic “prophet who is coming into the world”). Since “Chosen One” fits the context and gives the evangelist yet another Messianic confession of Jesus, it seems to be preferred as the original. But in either case, the interpreter must also do textual criticism.

Thus textual criticism, rather than being simply an exercise for the expert preceding exegesis, is also an integral part of the interpretation of the Word of God.

VII.     Bibliography

Books

Aland, K. Kurzegefasste Liste der Griechischen Handschriften des Neuen Testaments. Berlin: Walter de Gruyter, 1963.

Burgon, J.W. The Traditional Text of the Holy Gospels Vindicated and Established. Edited by E.F. Miller. London: 1896.

Colwell, E.C. Studies in Methodology in the Textual Criticism of the New Testament. Grand Rapids: Eerdmans, 1970.

______. What is the Best New Testament? Chicago: University of Chicago Press, 1952.

Elliott, J.K. The Greek Text of the Epistles to Timothy and Titus. Salt Lake City: University of Utah Press, 1968.

Epp, E.J. The Theological Tendency of Codex Bezae Cantabrigiensis. Cambridge: Cambridge University Press, 1966.

Fee, G.D. Papyrus Bodmer 11 (P⁶⁶) Its Textual Relationships and Scribal Characteristics. Salt Lake City: University of Utah Press, 1968.

Greenlee, J.H. Introduction to New Testament Textual Criticism. Grand Rapids: Eerdmans, 1964.

Hatch, W.H.P. Facsimiles and Descriptions of Minuscule Manuscripts of the New Testament. Cam bridge: Harvard University Press, 1951.

______. The Principal Uncial Manuscripts of the New Testament Chicago: University of Chicago Press, 1939.

Kenyon, F.G. The Text of the Greek Bible. New ed. London: Duckworth, 1949.

Lake, Kirsopp. The Text of the New Testament. Rev. ed. London: 1928.

Metzger, B.M. The Early Versions of the New Testament: Their Origin, Transmission and Limitations. Oxford: Clarendon Press, 1977.

______. The Text of the New Testament, Its Transmission, Corruption, and Restoration. Rev. ed. New York: Oxford University Press, 1968.

______. A Textual Commentary on the Greek New Testament: A Companion Volume to the United Bible Societies’ Greek New Testament (third edition). London/ New York: United Bible Societies, 1971.

Streeter, B.H. The Four Gospels: A Study of Origins. Rev. ed. London: Macmillan, 1936.

Westcott, B.F. and Hort, F.J.A. The New Testament in the Original Greek, with Introduction and Appendix. 2 vols. London: MacMillan, 1881–82.

Williams, C.S.C. Alterations to the Text of the Synoptic Gospels and Acts. Oxford: Basil Blackwell, 1951.

Zuntz, G. The Text Of the Epistles. London: The British Academy, 1953.

Articles

Birdsall, J.N. “The New Testament Text.” The Cambridge History Of the Bible, vol. 1. Edited by P.R. Ackroyd and C.F. Evans. Cambridge: Cambridge University Press, 1970, pp. 308–77.

Fee, G.D. “P⁷⁵, P⁶⁶, and Origen: The Myth of Early Textual Recension in Alexandria,” New Dimensions in New Testament Study. Edited by R.N. Longenecker and M.C. Tenney. Grand Rapids: Zondervan, 1974, pp. 19–45.

______. “Rigorous or Reasoned Eclecticism—Which?” Studies in New Testament Language and Text. Edited by J.K. Elliot. Leiden: Brill, 1976 pp. 174–197.

______. “The Text of John in Origen and Cyril of Alexandria: A Contribution to Methodology in the Recovery and Analysis of Patristic Citations.” Biblica 52 (1971): 357–94.

Hodges, Z.C. “The Greek Text of the King James Version.” BS 125 (1968): 334–45.

Kilpatrick, G.D. “An Eclectic Study of the Text of Acts.” Biblical and Patristic Studies in Memory of R.P. Casey. Edited by J.N. Birdsall and R.W. Thomson. Freiburg: 1963, pp. 64–77.

______. “The Greek New Testament Text of Today and the Textus Receptus.” The New Testament in Historical and Contemporary Perspective: Essays in Memory Of G.H.C Macgregor. Edited by H. Anderson and W. Barclay. Oxford: 1965. pp. 189–208.

1 Father Jose O’Callaghan recently suggested that some Greek fragments of scrolls in Qumran Cave 7 should be identified as parts of the NT (Biblica, 53 [1972]: 91–100; trans. by W.L. Holladay and published as a supplement to the JBL, 91 [June, 1972]); however, his “find” has not held up under careful scrutiny (see, e.g., C.D. Fee, “Some Dissenting Notes on 7Q5 = Mark 6:52–53,” JBL, 92 [1973]: 109–112).

MSS Manuscript(s)
MS Manuscript(s)
c. circa, about
א Codex Sinaiticus
B Codex Vaticanus
OL Old Latin
viz. videlicet, namely
Vul. Vulgate
C Codex Ephraemi Syri
D Codex Bezae
A Codex Alexandrinus
KJV King James Version
TR Textus Receptus
WH Westcott and Hort, The New Testament in Greek
Nestle Nestle (ed.) Novum Testamention Graece
UBS The United Bible Societies’ Greek Text
RSV Revised Standard Version
NEB The New English Bible
JB The Jerusalem Bible
b Babylonian Gemara
BS Bibliotheca Sacra

Gaebelein, F. E. (1979). The Expositor's Bible Commentary, Volume 1: Introductory Articles. Grand Rapids, MI: Zondervan Publishing House.

Page .  Exported from Logos Bible Software 4, 9:20 PM May 11, 2013.