I am reproducing the introduction so you can see my quotes 
in their original context. If you wish to read Metzger's entire 
commentary you must buy the book.

A Textual Commentary on the Greek New Testament, 1971

Bruce M. Metzger

PREFACE

The present volume is designed to serve as a companion to the third edition of the United Bible Societies' Greek New Testament, edited by Kurt Aland, Matthew Black, Carlo M. Martini, Bruce M. Metzger, and Allen Wikgren.

One of the chief purposes of the commentary is to set forth the reasons that led the Committee, or a Majority of the members of the Committee, to adopt certain variant readings for inclusion in the text and to relegate certain other readings to the apparatus. On the basis of a record of the voting of the Committee, as well as, for most sessions, more or less full notes of the discussions that preceded the voting, the present writer has sought to frame and express concisely (a) the main problem or problems involved in each set of variants and (b) the Com­mittee's evaluation and resolution of those problems. In writing the commentary it was necessary not only to review what the Committee had done, but also to consult once again the several commentaries, concordances, synopses, lexicons, grammars, and similar reference works that had been utilized by members of the Committee during their discussions. More than once the record of the discussion proved to be incomplete because, amid the lively exchange of opinions, the Committee had come to a decision without the formal enunciation of those reasons that appeared at the time to be obvious or self-evident. In such cases it was necessary for the present writer to supplement, or even to reconstruct, the tenor of the Committee's discussions.

The general Introduction to the commentary includes an outline of the chief kinds of considerations that the Com­mittee took into account in choosing among variant readings. By becoming acquainted with these criteria (pp. xxv-xxviii) the reader will be able to understand more readily the presuppositions that underlie the Committee's evaluations of the di­vergent readings.

In addition to the 1440 sets of variant readings supplied in the apparatus of the Bible Societies' edition, the selection of which was made chiefly on the basis of their exegetical impor­tance to the translator and student, the Committee suggested that certain other readings also deserved discussion in the supplementary volume. The author has therefore included comments on about 600 additional sets of variant readings, scattered throughout the New Testament; the majority of them, it will be noted, occur in the book of Acts, which, because of its peculiar textual problems, seemed to demand special attention (see the Introduction to the book of Acts).

In the comments on the variant readings for which the text-volume supplies an apparatus, it was considered sufficient to cite merely the more important manuscript witnesses; the reader of the commentary will be able, if he wishes, to supple­ment the partial citation of evidence by consulting the fuller apparatus in the text-volume. On the other hand, occasionally the discussion in the commentary supplements the apparatus in the text-volume by the citation of additional witnesses, a few of which were not known at the time of the Committee's work, and others of which had been deemed unimportant for citation in the apparatus. Since the present volume is designed to assist translators and students who may not have available an extensive library, the comments on the 600 additional sets of variant readings are accompanied by a more or less full citation of evidence, drawn from such standard apparatus critici as those of Tischendorf, von Soden, Nestle, Merk, Bover, Souter, Hoskier (for Revelation), and Wordsworth and White, as well as from editions of individual manuscripts (for information con­cerning the Greek manuscripts cited in the commentary but not in the text-volume, see below, pp. 771 ff.).

The writing of the commentary was begun during 1964, when the author, on sabbatical leave from his usual academic duties, was a member of the Institute for Advanced Study in Princeton. During the following years, as the first draft of each major section was completed, it was circulated among the other members of the Committee to make certain that the comments reflected adequately the Committee's deliberations. Frequently it had happened that the members of the Committee differed in their evaluation of the textual evidence, and thus many readings were adopted on the basis of a majority vote. In special cases, when a member holding a minority opinion had strong feelings that the majority had seriously gone astray, opportunity was given for him to express his own point of view. Such occasional comments, identified by the writer's initials and enclosed within square brackets, are appended to the main discussion of the textual problem in question.

The author is grateful to Professors Black, Martini, and Wikgren who, having read the typescript of the commentary, made several suggestions, corrections, and additions which have been incorporated into the volume; for the errors that remain he alone, of course, is responsible. Appreciation must also be expressed to Dr. Robert P. Markham for his capable and courteous assistance given at all stages of the work. The formidable task of typing the handwritten copy of the manu­script was executed with exceptional accuracy by Mrs. Richard E. Munson. Similarly the craftsmen of the firm of Maurice Jacobs, Inc., deserve commendation for the high quality of their work, which included the preparation of a special font of Greek type to represent the script used in uncial manuscripts. Assistance in the onerous task of proofreading was given by Dr. Markham, Mr. Stanley L. Morris, Mrs. Munson, Dr. Erroll Rhodes, and Professor Wikgren. Finally, I wish to express my sincere thanks to Dr. Eugene A. Nida of the American Bible Society for haying invited me to prepare this companion to our Greek text. Although the writing of the volume proved to be a far greater and much more exacting task than it appeared when I accepted the invitation, now that it is completed I am grateful to him for having given me the opportunity of en­larging, as one may hope, the usefulness of the United Bible Societies' edition of the Greek New Testament.

BRUCE M. METZGER

Princeton Theological Seminary

September 30, 1970

INTRODUCTION

Most commentaries on the Bible seek to explain the meaning of words, phrases, and ideas of the scriptural text in their nearer and wider context; a textual commentary, however, is concerned with the prior question, What is the original text of the passage? That such a question must be asked-and an­swered!-before one explains the meaning of the text arises from two circumstances: (a) none of the original documents of the Bible is extant today, and (b) the existing copies differ from one another.

Despite the large number of general and specialized com­mentaries on the books of the New Testament, very few deal adequately with textual problems. In fact, there is none that deals comprehensively with the entire New Testament, and those that supply the fullest discussions were written during the past century and are, of course, seriously out of date today. Among nineteenth century works devoted exclusively to textual problems are Rinck's commentary on the Acts of the Apostles and the Epistles, and Reiche's three volumes on the Pauline and Catholic Epistles. Not nearly so extensive but much more widely known are the "Notes on Select Readings" which are included in the second volume, entitled Introduction [and] Appendix, of B. F. Westcott and F. J. A. Hort's The New Testament in the Original Greek (Cambridge and London, 1881). Approximately 425 passages are considered in these Notes, some of which involve 1engthy discussions that remain per­manently valuable, while others provide merely the citation of evidence without comment. The second edition of the volume (1896) contains nearly 50 additional Notes, prepared by F. C. Burkitt and dealing with the newly discovered Sinaitic Syriac manuscript of the Gospels. At the close of the century Edward Miller, a disciple of Dean J. W. Burgon, issued Part I of his Textual Commentary upon the Holy Gospels (London, 1899), covering the first fourteen chapters of the Gospel according to Matthew. This work, however, is misnamed, for instead of being a commentary in the usual sense of the word, it comprises nothing more than a critical apparatus of variant readings.

The twentieth century saw the publication of de Zwaan's doctoral dissertation devoted to the textual problems of 2 Peter and Jude, and Turner's elaborate analyses of Markan usage, culminating in "A Textual Commentary on Mark 1." More recently R. V. G. Tasker has provided about 270 brief "Notes on Variant Readings" in the Appendix to his addition of The Greek New Testament (Oxford and Cambridge, 1964), the text of which is to be regarded as lying behind The New English Bible (1961).

In the following pages the reader will find a succinct state­ment of (1) the history of the transmission of the New Testa­ment text, (2) the principal criteria used in choosing among conflicting witnesses to the text, and (3) the chief witnesses to the New Testament listed according to types of text.

I. HISTORY OF THE TRANSMISSION OF THE NEW TESTAMENT TEXT

In the earliest days of the Christian church, after an-apostolic letter was sent to a congregation or an individual, or after a gospel was written to meet the needs of a particular reading public, copies would be made in order to extend its influence and to enable others to profit from it as well. It was inevitable that such handwritten copies would contain a greater or lesser number of differences in wording from the original. Most of the divergences arose from quite accidental causes, such as mistaking a letter or a word for another that looked like it. If two neighboring lines of a manuscript began or ended with the same group of letters or if two similar words stood near each other in the same line, it was easy for the eye of the copyist to jump from the first group of letters to the second, and so for a portion of the text to be omitted (called homoeoarcton or homoeoteleuton, depending upon whether the simi­larity of letters occurred at the beginning or the ending of the words). Conversely the scribe might go back from the second to the first group and unwittingly copy one or more words twice (called dittography). Letters that were pronounced alike were sometimes confused (called itacism). Such accidental errors are almost unavoidable whenever lengthy passages are copied by hand, and would be especially likely to occur if the scribe had defective eyesight, or was interrupted while copying or, because of fatigue, was less attentive to his task than he should have been.

Other divergences in wording arose from deliberate attempts to smooth out grammatical or stylistic harshness, or to eliminate real or imagined obscurities of meaning in the text. Sometimes a copyist would substitute or would add what seemed to him to be a more appropriate word or form, perhaps derived from a parallel passage (called harmonization or assimilation). Thus, during the years immediately following the composition of the several documents that eventually were collected to form the New Testament, hundreds if not thousands of variant read­ings arose.

Still other kinds of divergencies originated when the New Testament documents were translated from Greek into other languages. During the second and third centuries after Chris­tianity had been introduced into Syria, into North Africa and Italy, into central and southern Egypt, both congregations and individual believers would naturally desire copies of the Scrip­tures in their own languages. And so versions in Syriac, in Latin, and in the several dialects of Coptic used in Egypt were produced. They were followed in the fourth and succeeding centuries by other versions in Armenian, Georgian, Ethiopic, Arabic and Nubian in the East, and in Gothic and (much later) Anglo-Saxon in the West.­

The accuracy of such translations was directly related to two factors: (a) the degree of familiarity possessed by the translator of both Greek and the language into which the translation was made, and (b) the amount of care he devoted to the task of making the translation. It is not surprising that very considerable divergences in early versions developed, first, when different persons made different translations from what may have been slightly different forms of Greek text; and, second, when these renderings in one or another language were transmitted in handwritten copies by scribes who, familiar with a slightly different form of text (either a divergent Greek text or a divergent versional rendering), adjusted the new copies so as to accord with what they considered the preferable wording.

During the early centuries of the expansion of the Christian church, what are called "local texts" of the New Testament gradually developed. Newly established congregations in and near a large city, such as Alexandria, Antioch, Constantinople, Carthage, or Rome, were provided with copies of the Scriptures in the form which was current in that area. As additional copies were made, the number of special readings and renderings would be both conserved and, to some extent, increased so that eventually a type of text gew up which was more or less peculiar to that locality. Today it is possible to identify the type of text preserved in New Testament manuscripts by com­paring their characteristic readings with the quotations of those passages in the writings of Church Fathers who lived in or near the chief ecclesiastical centers.

At the same time the distinctiveness of a local text tended to become diluted and mixed with other types of text. A manu­script of the Gospel of Mark copied in Alexandria, for example, and taken later to Rome would doubtless influence to some extent copyists transcribing the form of the text of Mark heretofore current at Rome. On the whole, however, during the earliest centuries the tendencies to develop and preserve a particular type of text prevailed over the tendencies leading to a mixture of texts. Thus there grew up several distinctive kinds of New Testament text, the most important of which are the following.

The Alexandrian text, which Westcott and Hort called the Neutral text (a question-begging title), is usually considered to be the best text and the most faithful in preserving the original. Characteristics of the Alexandrian text are brevity and austerity. That is, it is generally shorter than the text of other forms, and it does not exhibit the degree of grammatical and stylistic polishing that is characteristic of the Byzantine and, to a lesser extent, of the Caesarean type of text. Until recently the two chief witnesses to the Alexandrian text were codex Vaticanus (B) and codex Siniaticus (א), parchment manuscripts dating from about the middle of the fourth century. With the acquisition, however, of the Bodmer Papyri, par­ticularly p66 and p75, both copied about the end of the second or the beginning of the third century, evidence is now available that the Alexandrian type of text goes back to an archetype that must be dated early in the second century. The Sahidic and Bohairic versions frequently contain typically Alexandrian readings.

The Western text, which was widely current in Italy and Gaul as well as in North Africa and elsewhere (including Egypt), can also be traced back to the second century. It was used by Marcion, Tatian, Irenaeus, Tertullian, and Cyprian. Its presence in Egypt is shown by the papyri p38 (about A.D. 300) and p48 (about the end of the third century). The most important Greek manuscripts that present a Western type of text are codex Bazae (D) of the fifth or sixth century (containing the Gospels and Acts), codex Claromontanus (D) of the sixth century (containing the Pauline epistles), and, for Mark 1.1 to 5.30, codex Washingtonianus (W) of the late fourth or early fifth century. Likewise the Old Latin versions are noteworthy witnesses to a Western type of text; these fall into three main groups, the African, Italian, and Hispanic forms of Old Latin texts.

The chief characteristic of Western readings is fondness for paraphrase. Words, clauses, and even whole sentences are freely changed, omitted, or inserted. Sometimes the motive appears to have been harmonization, while at other times it was the enrichment of the narrative by the inclusion of tradi­tional or apocryphal material. Some readings involve quite trivial alterations for which no special reason can be assigned. One of the puzzling features of the Western text (which generally is longer than the other forms of text) is that at the end of Luke and in a few other places in the New Testament certain Western witnesses omit words and passages that are present in other forms of text, including the Alexandrian. Although at the close of the last century certain scholars were disposed to regard these shorter readings as original (Westcott and Hort called them "Western non-interpolations"), since the acquisition of the Podmer Papyri many scholars today are inclined to regard them as aberrant readings (see the Note on Western Non-Interpolations, pp. 191-193).

In the book of Acts the problems raised by the Western text become most acute, for the Western text of Acts is nearly ten percent longer than the form which is commonly regarded to be the original text of that book. For this reason the present volume devotes proportionately more space to variant readings in Acts than to those in any other New Testament book, and a special Introduction to the textual phenomena in Acts is provided (see pp. 259-272).

The Caesarean text, which seems to have originated in Egypt (it is attested by the Chester Beatty Papyrus p45), was brought, perhaps by Origen, to Caesarea, where it was used by Eusebius and others. From Caesarea it was carried to Jerusalem, where it was used by Cyril and by Armenians who, at an early date, had a colony at Jerusalem. Armenian missionaries carried the Caesarean text into Georgia, where it influenced the Georgian version as well as an uncial Greek manuscript of about the ninth century ((-), codex Koridethi). Furthermore, perhaps Euthalius's scholarly edition of the Pauline Epistles was made at Caesarea (so Zuntz).

Thus it appears that the Caesarean type of text has had a long and checkered career.  According to the view of most scholars, it is an Eastern text, dating from the early part of the third century, and is characterized by a distinctive mixture of Western readings and Alexandrian readings. One may also observe a certain striving after elegance of expression, a feature that is especially typical of the Byzantine type of text.

Another Eastern type of text, current in and near Antioch, is preserved today chiefly in Old Syriac witnesses, namely the Sinaitic and the Curetonian manuscripts of the Gospels and in the quotations of Scripture contained in the works of Aphraates and Ephraem.

The Byzantine text, otherwise called the Syrian text (so Westcott and Hort), the Koine Text (so von Soden), the Ecclesiastical text (so Lake), and the Antiochian text (so Ropes), is, on the whole, the latest of the several distinctive types of text of the New Testament, It is characterized chiefly by lucidity and completeness. The framers of this text sought to smooth away any harshness of language, to combine two or more divergent readings into one expanded reading (called conflation), and to harmonize divergent parallel passages. This conflated text, produced perhaps at Antioch in Syria, was taken to Constantinople, whence it was distributed widely throughout the Byzantine Empire. It is best represented today by codex Alexandrinus (in the Gospels; not in Acts, the Epistles, or Revelation), the later uncial manuscripts, and the great mass of minuscule manuscripts. Thus, except for an occasional manu­script that happened to preserve an earlier form of text, during the period from about the sixth or seventh century down to the invention of printing with moveable type (A. D. 1450-56), the Byzantine form of text was generally regarded as the au­thoritative form of text and was the one most widely circulated and accepted.

After Gutenberg's press made the production of books more rapid and therefore cheaper than was possible through copying by hand, it was the debased Byzantine text that became the standard form of the New Testament in printed editions. This unfortunate situation was not altogether unexpected, for the Greek manuscripts of the New Testament which were most readily available to early editors and printers were those that contained the corrupt Byzantine text.

The first published edition of the printed Greek Testament, issued at Basel in 1516, was prepared by Desiderius Erasmus, the Dutch humanist scholar. Since Erasmus could find no manuscript that contained the entire Greek Testament, he utilized several for the various divisions of the New Testament. For the greater part of his text he relied on two rather inferior manuscripts now in the university library at Basel, one of the Gospels and one of the Acts and Epistles, both dating from about the twelfth century. Erasmus compared them with two or three others, and entered occasional corrections in the margins or between the lines of the copy given to the printer. For the book of Revelation he had but one manuscript, dating from the twelfth century, which he had borrowed from his friend Reuchlin. As it happened, this copy lacked the final leaf, which had contained the last six verses of the book. For these verses Erasmus depended upon Jerome's Latin Vulgate, translating this version into Greek. As would be expected from such a procedure, here and there in Erasmus's reconstruction of these verses are several readings which have never been found in any Greek manuscript-but which are still per­petuated today in printings of the so-called Textus Receptus of the Greek New Testament. In other parts of the New Testa­ment Erasmus also occasionally introduced into his Greek text material derived from the current form of the Latin Vulgate.

So much in demand was Erasmus's Greek Testament that the first edition was soon exhausted and a second was called for. It was this second edition of 1519, in which some (but not nearly all) of the many typographical blunders of the first edition had been corrected, that Martin Luther and William Tyndale used as the basis of their translations of the New Testament into German (1522) and into English (1525).

In the years following many other editors and printers issued a variety of editions of the Greek Testament, all of which reproduced more or less the same type of text, namely that preserved in the later Byzantine manuscripts. Even when it happened that an editor had access to older manuscripts-­as when Theodore Beza, the friend and successor of Calvin at Geneva, acquired the fifth or sixth century manuscript that goes under his name today as well as the sixth century codex Claromontanus he made relatively little use of them, for they deviated too far from the form of text that had become standard in the later copies.

Noteworthy early editions of the Greek New Testament include two issued by Robert Etienne (commonly known under the Latin form of his name, Stephanus) , the famous Parisian printer who later moved to Geneva and threw in his lot with the Protestants of that city. In 1550 Stephanus published at Paris his third edition, the editio Regia, a magnificent folio edition. It is the first printed Greek Testament to contain a critical apparatus; on the inner margins of its Rages Stephanus entered variant readings from fourteen Greek manuscripts as well as readings from another printed edition, the Com­plutensian Polyglot. Stephanus's fourth edition (Geneva, 1551), which contains two Latin versions (the Vulgate and that of Erasmus), is noteworthy because in it for the first time the text of the New Testament was divided into numbered verses.

Theodor Beza published no fewer than nine editions of the Greek Testament between 1565 and 1604, and a tenth edition appeared posthumously in 1611. The importance of Beza's work lies in the extent to ,which his editions tended to popularize and stereotype what came to be called the Textus Receptus. The translators of the Authorized or King James Bible of 1611 made large use of Beza's editions of 1588-89 and 1598.

The term Textus Receptus, as applied to the text of the New Testament, originated in an expression used by the Elzevir brothers, who were printers in Leiden and later in Amsterdam. The preface to their second edition of the Greek Testament (1633) contains the sentence: Textum ergo habes nunc ab omnibus receptum in quo nihilimmutatum aut corruptum damus ("There­fore you [dear reader] now have the text received by all, in which we give nothing changed or corrupted"). In one sense this proud claim of the Elzevirs in behalf of their edition seemed to be justified, for their edition in most respects, not different from the approximately l60 other editions of the printed Greek Testament that had been issued since Erasmus's first published edition of 1516. In a more precise sense, however, the Byzantine form of the Greek text, repro­duced in all early printed editions, was disfigured, as was mentioned above, by the accumulation over the centuries of myriads of scribal alterations, many of minor significance but some of considerable consequence.

It was the corrupt Byzantine form of text that provided the basis for almost all translations of the New Testament into modern languages down to the nineteenth century. During the eighteenth century scholars assembled a great amount of information from many Greek manuscripts, as well as from versional and patristic witnesses. But, except for three or four editors who timidly corrected some of the more blatant errors of the Textus Receptus, this debased form of the New Testament text was reprinted in edition after edition. It was only in the first part of the nineteenth century (1831) that a German classical scholar Karl Lachmann, ventured to apply to the New Testament the criteria that he had used in editing texts of the classics. Subsequently other critical editions appeared, including those prepared by Constantin von Tischendorf, whose eighth edition (1869-72) remains a Monumental thesaurus of variant readings, and the influential edition prepared by two Cambridge scholars, B. F. Westcott and F. J. A. Hort (1881). It is the latter edition that was taken as the basis for the  present United Bible Societies' edition. During the twentieth century, with the discovery of several New Testament manuscripts much older than any that had hitherto been available, it has become possible to produce editions of the New Testa­ment that approximate ever more closely to what is regarded as the wording of the original documents.

II. CRITERIA USED IN CHOOSING AMONG CONFLICTING READINGS IN NEW TESTAMENT WITNESSES

In the preceding section the reader will have seen how, during about fourteen centuries when the New Testament was transmitted in handwritten copies, numerous changes and accretions came into the text. Of the approximately five thousand Greek manuscripts of all or part of the New Testament that are known today, no two agree exactly in all particulars. Confronted by a mass of conflicting readings, editors must decide which variants deserve to be included in the text and which should be relegated to the apparatus. Although at first it may seem to be a hopeless task amid so many thousands of variant readings to sort out those that should be regarded as original, textual scholars have developed certain generally acknowledged criteria of evaluation. These considerations depend, it will be seen, upon probabilities, and sometimes the textual critic must weigh one set of probabilities against another. Furthermore, the reader should be advised at the outset that, although the following criteria have been drawn up in a more or less tidy outline form, their application can never be undertaken in a merely mechanical or stereotyped manner. The range and complexity of textual data are so great that no neatly arranged or mechanically contrive set of rules can be applied with mathematical precision. Each and every variant reading needs to be considered in itself, and not judged merely according to a rule of thumb. With these cau­tionary comments in mind, the reader will appreciate that the following outline of criteria is meant only as a convenient description of the more important considerations that the Committee took into account when choosing among variant readings.

The chief categories or kinds of criteria and considerations that assist one in evaluating the relative worth of variant readings are those which involve (I) External Evidence, having to do with the manuscripts themselves, and (II) Internal Evidence, having to do with two kinds of considerations, (A) those concerned with Transcriptional Probabilities (i. e. relating to the habits of scribes) and (B) those concerned with Intrinsic Probabilities (i. e. relating to the style of the author).

OUTLINE OF CRITERIA

I. EXTERNAL EVIDENCE, involving considerations bearing upon:

A. The date and character of the witnesses. In general, earlier manuscripts are more likely to be free from those errors that arise from repeated copying. Of even greater importance, however, than the age of the document itself are the date and character of the type of text that it embodies, as well as the degree of care taken by the copyist while producing the manuscript.

B. The geographical distribution of the witnesses that support a variant. The concurrence of witnesses, for example, from Antioch, Alexandria, and Gaul in support of a given variant is, other things being equal, more significant than the testimony of witnesses representing but one locality or one ecclesiastical see. On the other hand, however, one must be certain that geographically remote witnesses are really in­dependent of one another. Agreements, for example, between Old Latin and Old Syriac witnesses may sometimes be due to common influence from Tatian's Diatessaron.

C. The genealogical relationship of texts and families of witnesses. Mere numbers of witnesses supporting a given variant reading do not necessarily prove the superiority of that reading. For example, if in a given sentence reading x is supported by twenty manuscripts and reading y by only one manuscript, the relative numerical support favoring x counts for nothing if all twenty manuscripts should be discovered to be copies made from a single manuscript, no longer extant, whose scribe first introduced that particular variant reading. The comparison, in that case, ought to be made between the one manuscript containing reading y and the single ancestor of the twenty manuscripts containing reading x.

D. Witnesses are to be weighed rather than counted. That is, the principle enunciated in the previous paragraph needs to be elaborated: those witnesses that are found to be generally trustworthy in clear-cut cases deserve to be accorded pre­dominant weight in cases when the textual problems are am­biguous and their resolution is uncertain. At the same time, however, since the relative weight of the several kinds of evidence differs for different kinds of variants, there should be no merely mechanical evaluation of the evidence.

II. INTERNAL EVIDENCE, involving two kinds of probabilities:

A. Transcriptional Probabilities depend upon considerations of the habits of scribes and upon palaeographical features in the manuscripts.

1. In general, the more difficult reading is to be preferred, particularly when the sense appears on the surface to be erroneous but on more mature consideration proves itself to be correct. (Here "more difficult" means "more difficult to the scribe," who would be tempted to make an emendation. The characteristic of most scribal emendations is their superficiality, often combining "the appearance of improvement with the absence of its reality." Obviously the category "more difficult reading" is relative, and sometimes a point is reached when a reading must be judged to be so difficult that it can have arisen only by accident in transcription.)

2. In general the shorter reading is to be preferred, except where

(a) Parablepsis arising from homoeoarcton or homoeo­teleuton may have occurred (i. e., where the eye of the copyist may have inadvertently passed from one word to another having a similar sequence of letters); or where

(b) The scribe may have omitted material which he deemed to be (i) superfluous, (ii) harsh, or (iii) contrary to pious belief, liturgical usage, or as ascetical practice.

3. Since scribes would frequently bring divergent passages into harmony with one another, in parallel passages (whether quotations from the Old Testament or different accounts in the Gospels of the same event or narrative) that reading which involves verbal dissidence is usually to be preferred to one which is verbally concordant.

4. Scribes would sometimes

(a)     Replace an unfamiliar word with a more familiar synonym;

(b)     Alter a less refined grammatical form or less elegant lexical expression in accord with contemporary Atticizing preferences; or

(c)     Add pronouns, conjunctions, and expletives to make a smoother text.

B. Intrinsic Probabilities depend upon considerations of what the author was more likely to have written. The textual critic takes into account

1. In general:

(a)     The style and vocabulary of the author throughout the book:

(b)     The immediate context; and

(c)     Harmony with the usage of the author elsewhere; and,

2. In the Gospels:

(a)     The Aramaic background of the teaching of Jesus;

(b)     The priority of the Gospel according to Mark; and

(c)     The influence of the Christian community upon the formulation and transmission of the passage in question.

It is obvious that not all of these criteria are applicable in every case. The textual critic must know when it is appropriate to give greater consideration to one kind of evidence and less to another. Since textual criticism is an art as well as a science, it is inevitable that in some cases different scholars will come to different evaluations of the significance of the evidence. This divergence is almost inevitable when, as sometimes happens, the evidence is so divided that, for example, the .more difficult reading is found only in the later witnesses, or the longer reading is found only in the earlier witnesses.

In order to indicate the relative degree of certainty in the mind of the Committee for the reading adopted as the text, an identifying letter is included within braces at the beginning of each set of textual variants. The letter {A} signifies that the text is virtually certain, while {B} indicates that there is some degree of doubt concerning the reading selected for the text. The letter, {C} means that there is a considerable degree of doubt whether the text or the apparatus contains the superior reading, while {D} shows that there is a very high degree of doubt concerning the reading selected for the text. In fact, among the {D} decisions sometimes none of the variant readings commended itself as original, and therefore the only recourse was to print the least unsatisfactory reading.

III. LISTS OF WITNESSES ACCORDING TO TYPE OF TEXT

The following are some of the more important witnesses to the text of the New Testament arranged in lists according to the predominant type of text exhibited by each witness. It will be observed that in some cases different sections of the New Testament within the same witness belong to different text­-types.

[Under each of the following headings is a list of manuscripts.]  

Alexandrian Witnesses

Western Witnesses

Caesarean Witnesses

Byzantine Witnesses

In assessing the preceding lists of witnesses two comments are appropriate. (a) The tables include only those witnesses that are more or less generally acknowledged to be the chief representatives of the several textual types. Additional wit­nesses have at times been assigned to one or another category; for example, among the weaker representatives of the Caesarean text B. H. Streeter was inclined to include the deluxe, purple manuscripts N, 0, and Z, as well as U A. ~ 157 544 fam. 1424, 1071 and 1604, and other scholars have identified still other witnesses as Caesarean.

(b) While the reader is encouraged to refer from time to time from the commentary to the above lists of witnesses, it must never be supposed that identity of external support for two separate sets of variant readings requires identical judg­ments concerning the original text. Although the external evidence for two sets of variant readings may be exactly the same, considerations of transcriptional and/or intrinsic prob­abilities of readings may lead to quite diverse judgments con­cerning the original text. This is, of course, only another way of saying that textual criticism is an art as well as a science, and demands that each set of variants be evaluated in the light of the fullest consideration of both external evidence and internal probabilities.

Home    Received Text or Critical Text