NT textual criticism is the science of discerning the reading of the original Greek text of the NT. NT textual criticism is relatively different from the textual criticism of the Hebrew OT, since the two Testaments were copied in substantially different ways, resulting in quite different issues and types of copying problems.

Textual criticism is necessary for two reasons: (1) none of the original texts for any of the books of the NT (the autographs) have survived; (2) all the surviving copies that we do have differ from one another in at least minor ways. NT textual criticism is the discipline of examining all the readings found in the surviving copies (including other early translations and the writings of the church fathers) in order to discern the most-likely original text of the NT.

In many ways there is nothing surprising or unusual about this activity. This same discipline is used in the case of all ancient documents where the original no longer exists and there are multiple, but different, surviving copies. What sets the textual criticism of the Bible apart from the textual criticism of Plato, Aristotle, or any other extrabiblical author is the importance of Scripture. It is absolutely crucial that scholars be as accurate as possible in discerning the words of the Bible. Present-day doctrinal statements about the inspiration and inerrancy of Scripture usually focus on how it is the actual documents written by the original authors of the Bible (the autographs) that were inspired and inerrant, and not necessarily each and every copy or translation that was ever made. At the same time, NT textual criticism provides strong and ample evidence that the standard translations of today are reasonable copies of the original texts and therefore are inspired and authoritative.

Another reality that sets the textual criticism of the Bible apart from the textual criticism of any other ancient document involves the large number of manuscripts containing all or parts of the NT. Aristotle’s writings (384–322 BC) have survived in only five ancient manuscripts, the earliest of which was last copied around AD 1100. Thus, the manuscript support for discerning the proper wording of Aristotle’s writings is five manuscripts, the earliest of which was copied some fourteen hundred years after its original composition. The NT, by contrast, has been preserved in almost 5,500 handwritten Greek manuscripts, the earliest of which may be only decades removed from the actual composition of the NT. In addition, there are tens of thousands of other manuscripts of the NT translated into other languages of the early church (especially Latin, Syriac, and Coptic) and perhaps something like a million quotations and allusions to the NT in the writings of the church fathers. The amount of manuscript support for the Bible is without parallel when compared with any other ancient writing, thus providing a firm foundation for the trustworthiness of the Bible.

New Testament Manuscripts

If all the surviving Greek manuscripts of the NT from all the museums and ancient-book rooms around the world could be gathered together and examined, there would be a number of obvious differences. For one thing, these manuscripts are written on different materials. The oldest manuscripts include some 116 papyri written on papyrus sheets (made from the stems of a papyrus plant pressed together to make a flat writing surface) and date from as early as AD 125 until the eighth century. Virtually all the rest of the hand-copied manuscripts were written on parchment (leather from animal skins stretched thin) and range in date from the second century to as late as the sixteenth century.

Another significant difference involves the style of handwriting. Here there are two primary categories: uncials (majuscules) and minuscules. The three hundred or so uncial manuscripts were written using an early style of “capital letters” (comparable to hand-printing a text in all capital letters) and range in date from the second century to as late as the tenth or eleventh century, with the majority between the fourth and tenth centuries. The 2,800 or so minuscule manuscripts were written in a later style of semicursive using primarily “lowercase letters” (more comparable to cursive penmanship today) and range in date from the ninth to the sixteenth centuries.

A third basic difference involves the arrangement of the material. In addition to standard biblical texts arranged in regular canonical order, almost half (about 2,200) of the manuscripts of the Greek NT are lectionaries composed of series of short passages designed to be read on a fixed regular schedule, almost like the responsive readings found in many modern hymnals. These range in date from the ninth to the sixteenth centuries and cover the vast majority of the passages in the NT, and thus they constitute another significant source of information about the proper reading of the original text.

Other obvious differences in outward appearance involve the scope of the original work and how well it has been preserved. Only a small number of manuscripts were written to encompass the entire Bible, including the OT (usually the Greek translation called the “Septuagint”), the NT, and frequently various books of the Apocrypha as well. Other manuscripts were designed to include only the NT. However, the vast majority of the surviving Greek manuscripts of the NT originally included only a portion of the NT (due primarily to the cost in materials and labor), most often a collection of the four Gospels, or the Pauline Epistles, or some other smaller subset of the NT. Still another reality is that many ancient manuscripts may have lost pages or the binding may have broken, leaving only incomplete sections of the original manuscript. Some NT manuscripts have survived only in the form of torn fragments, sometimes only a portion of a page in size.

Different Readings

Different readings among these manuscripts fit into two broad categories: unintentional changes and deliberate corrections. Unintentional changes took place when a scribe was genuinely trying to make a good copy and due to simple human error failed to read a letter clearly, did not hear a word clearly (assuming that the text was being dictated aloud), inadvertently wrote a homonym instead of the right word, lost the proper place on the page, or mistakenly assumed that some marginal note on the page belonged in the text. Deliberate corrections were also a recognized part of the work of a scribe whereby a scribe tried to make a text read better or more clearly. Although some of the early church fathers warn of heretics making deliberate changes in the text for theological reasons (actual instances of this are hard to document), sincere Christians sometimes sought to improve the theological clarity or style of the text.

There are various estimates of the number of places where different manuscripts have different readings. If one is taking stock of any and all differences regardless of how minor, then there are more than ten thousand different places in the text that have textual problems. The Textus Receptus, or Majority Text (underlying the KJV translation), has some five to six thousand differences from the Greek texts underlying most other, more recent English translations, although most of these are minor differences in word order, spelling, and so forth and make no practical difference in content. The standard scholarly United Bible Societies Greek NT includes a discussion of something like 1,440 sets of variant readings that were deemed sufficiently significant for inclusion in the critical apparatus. A contemporary translation such as the NIV lists over a hundred specific instances of textual uncertainty in the NT of sufficient significance to provide a possible alternative translation.

Still another question involves the theological significance of these differences among the surviving manuscripts. Although this can be a potentially threatening topic, it should be noted at the beginning that no area of Christian doctrine stands or falls because of possible textual problems. The entire foundation of theology is firm and certain. Nevertheless, there are perhaps several dozen places where textual differences have definite implications for either doctrine or practice. For example, does 1 Tim. 3:16 help establish the deity of Christ by referring to Christ as “God” (KJV), or is “he” (most modern translations) the correct reading? Or when we quote the Lord’s Prayer, should we include the final words “For thine is the kingdom, and the power, and the glory forever. Amen” (from the KJV)? Or, for still another issue, is it proper to preach and teach from the resurrection stories in Mark 16:9–20 or the story of the woman caught in adultery in John 7:53–8:11 (both of which are absent from early and reliable manuscripts)? English versions answer these questions differently.

However, a large number of textual problems are more minor in nature and involve little if any practical differences in application. For example, in 1 John 1:4 was John focusing on “your” joy or “our” joy? But by far the largest category involves textual differences that make no discernible differences at all in our English translations. Examples include differences in Greek word order, a choice between certain Greek tenses, similar grammatical constructions, the usage of synonyms, and some spelling differences. Differences in this final category are so minor that often they cannot be detected in English translations.

Different Families or Text Types

Another difference involves the “text type,” or family of manuscripts. Certain manuscripts with similar readings have a family resemblance with other similar manuscripts. The issue here involves patterns of copying errors among these manuscripts. In the late nineteenth century, two British scholars, B. F. Westcott and F. J. A. Hort, developed what they identified as the genealogical method for NT textual criticism. One component of this approach involves the recognition of at least three different “families” of related manuscripts presumably related to three different geographical areas of the early church: Byzantine or Syrian manuscripts (based in Syria or the later Byzantine Empire in Turkey), Alexandrian manuscripts (based in Egypt), and Western manuscripts (based more in Italy and North Africa). This classification system involves generalizations comparable to discerning a person’s nationality and ethnic background by various physical features. Some manuscripts are relatively easy to classify using this system, others are more difficult, and still others turn out to be quite mixed in nature. Yet there is a certain helpfulness to this system.

Certain conclusions can be drawn from this classification system. The vast majority of manuscripts (some 80 to 90 percent) are the Byzantine text type. These include virtually all the minuscules and lectionaries and tend to be relatively later in date. The remaining 10 to 20 percent of the manuscripts tend to be earlier and include the Alexandrian manuscripts, the handful of Western manuscripts, and a good number of early manuscripts that cannot be easily placed in either of these two categories. The Alexandrian readings tend to be briefer and more succinct in style, while the Western readings tend to be more unusual and eccentric, in contrast to the Byzantine ones, which tend to be more polished and harmonizing.

Applying Textual Criticism

There are different approaches to NT textual criticism. The two primary ones are (1) the King James/Majority Text approach associated with the KJV, and (2) the contemporary scholarly approach followed by virtually all other standard translations beginning with Westcott and Hort’s English Revised Version of 1881. Both approaches are built around an understanding of the history of the text. The Majority Text argument is built on the assumption that the original reading will have been preserved in the largest number of surviving manuscripts, in this case the relatively later Byzantine manuscripts. The contemporary scholarly approach assumes a so-called genealogical method whereby the Majority Text of the Byzantine manuscripts is believed to be the later “offspring” of the earlier Alexandrian and Western manuscripts and therefore more or less irrelevant for discovering the original text. Thus, this method essentially rejects the Byzantine manuscripts in favor of the earlier ones, especially the Alexandrian ones. Manuscripts are still counted and dated, but priority is given to the earlier manuscripts and to readings found in several different text types.

Textual Criticism, Old Testament–Centuries separate the earliest attested Hebrew manuscripts (MSS) and the canonical form of the biblical text. The transmission of the Hebrew Bible in many ancient and medieval sources reveals differences between the texts; consequently, textual criticism involves the comparison and analysis of those textual differences in both ancient sources and modern printed editions of the OT. Text critics collect from MSS and other textual witnesses all the details in which these texts differ from one another and then evaluate those differences to arrive at the most accurate original reading. Some of these differences were created in the course of textual transmission, while others are the result of scribal additions and the processes that created readings and texts over the centuries. In many cases, these differences are minimal and exert no significant theological or interpretational influence on the text itself.

Goals and Assumptions

The text critic seeks to recover the processes of the text’s written transmission so as to restore it not to the most ancient form or earliest literary strand of the biblical corpus, nor to the earliest attested textual form, but rather to the copy that contained the finished literary product and that stood at the beginning of the textual transmission process. Consequently, the text critic accepts the notion of a single, authentic text (Urtext) from which all extant MSS of a specific book have evolved rather than the existence of ancient parallel texts. Text critics identify and resolve textual errors in the text by revealing the history of their emergence, explaining how those errors or inconsistencies occurred. They consider and evaluate deviant or differing readings from the textual apparatus (notes written at the bottom of the printed critical versions) of the Masoretic Text (MT), such as Biblia Hebraica Stuttgartensia and R. Kittel’s Biblia Hebraica. Textual scholars remove or correct readings from extant or existing final forms of MSS and sometimes, in the absence of manuscript evidence, conjecturally emend or “suggest translational alternatives” for the text when the extant evidence defies a reasonable or clear reading. The most recent work by the Hebrew University Bible Project (HUBP) eliminates all scholarly conjectures proposed without manuscript evidence from its consideration in the text-critical process.

Discrepancies between textual witnesses to the OT include differences in book, chapter, and verse divisions, as well as layout and chapter sequence. The nature of scribal transmission assumes that although the scribes carefully copied the sacred texts, errors occasionally occurred. These textual corruptions may be unintentional, such as the reduplication of a letter or a word, confusion between letters that resemble one another, wrong word divisions, incorrect vocalization, or the accidental reversal of two letters. Occasionally a word may have been omitted, or words from marginal or interlinear comments (called a “gloss”) may have been inserted in the main text. In the case of intentional corruptions or alterations, scribes attempted to harmonize the morphology or grammar of a text, give explanations (also called “conflation”), incorporate material from parallel passages, substitute euphemistic words or phrases in place of offensive material, and, rarely, alter the texts theologically when the original wording seemed disrespectful to God.

Text Families and Translations

Some scholars theorize that the Hebrew MSS evolved from three local textual families arranged geographically, though more-recent scholarship seems to refute this view as simplistic. According to this theory, during the fifth century BC two local texts developed independently of each other in Babylon and Palestine. A third, the Alexandrian family, broke off from the Old Palestinian text and eventually served as the “source” or “model” for the translation of the Septuagint (LXX) between 250 BC and AD 150. The Samaritan Pentateuch resulted from the Palestinian family and, combined with the Babylonian tradition, formed the official corpus of the Pentateuch and the Former Prophets. The corpus of the Latter Prophets derived solely from the Palestinian text. This combination of texts formed the authoritative basis for the MT. The Leningrad Codex (AD 1009) is the oldest preserved complete Hebrew Bible, providing one of the best examples of the MT. The 1947 discovery of the Qumran scrolls gave text-critical scholars access to customary scribal activity and affirmed the existence of pluriform, or multiple, text types. These scrolls, transmitted during the third through first centuries BC, represented all the OT books except Esther and provided a Hebrew text witness that reinforced the relative veracity of the MT.

Ancient translations, such as the LXX, provided the source text from which the Old Greek, Lucian, Eusebius, and other Greek versions developed. In addition, the LXX formed the basis of the Old Latin, Vulgate, and Syriac versions as well as the Targumim. The importance of the elementary text-critical principle “Texts do not count, they weigh,” which forms the basic assumption when evaluating textual witnesses and comparing potential readings, becomes clearer in light of these developments. For example, a text critic would consider a reading in a Qumran witness as the equivalent or superior to readings attested in the LXX and its offspring texts.

Weighing Variant Readings

A consideration of variant readings involves all Hebrew and reconstructed details that differ from an accepted form of the MT, including additions, subtractions, letter and word differentiation, differences in word division, vocalization (or pronunciation), and word sequence. Scholars denote the MT as the textus receptus, or “received text,” which forms the translation base then of the Hebrew Scriptures. In some cases, critics of the OT believe that the MSS and versions reflect varying stages or versions in the editing of a final text. For example, 2 Sam. 22 and Ps. 18 are parallel texts in slightly different forms. Text critics acknowledge the existence of alternate forms of identical texts that were equally acceptable to the ancient poet by assuming that during the time of the OT, authors modernized or contemporized texts in an appropriate manner. One text does not have to be superior to the other; instead, both are legitimate, as in the case of Isa. 2:2–4 and Mic. 4:1–5.

OT text critics explore explanations for contradictory or different readings among text traditions, evaluate the weight of the textual evidence, and analyze a host of readings to determine which reading best explains the others or harmonizes with the literary evidence. In addition, text scholars ask whether a particular reading is grammatically acceptable and whether the word or phrase fits into the syntactic context of the sentence and the scope of the wider passage. In the absence of evidence to the contrary, text critics generally prefer the shorter reading because throughout history texts have a tendency to expand. In addition, when the MT and all other witnesses offer a text that is unobjectionable, makes sense, and has been preserved without a variant, scholars often assume that it is the “original text” preserved by the tradition and should be accepted. Furthermore, in the case of two alternative readings, if an explanation is available as to the manner in which one of these may have arisen from the other, then scholars consider the explicable reading as the weaker of the two.

The complexity of the text-critical enterprise, particularly when applied to the Hebrew Scriptures, requires a combination of intuition, skill, experience, careful attention to detail, and facility in cognate languages. The absence of extrabiblical Hebrew texts makes the OT textual analysis more difficult, since the scholar has a limited corpus of material from which to draw conclusions. OT text criticism is an ongoing process that will engage biblical scholars for years to come.