Can textual criticism actually help us figure out what the original reading was? How does this work, on a practical level? In this short video, Dan Wallace explains why he believes it does work: 1
Now, Dan Wallace concludes his discussion about the reasoned eclectic approach to New Testament textual criticism:
There are three pieces of external evidence that textual critics use to determine which variant is more likely to reflect the original wording: date and character, genealogical solidarity, and geographical distribution.
Date and Character
The reading found in the earliest manuscripts is usually to be preferred. This is because the fewer intermediary copies between a given manuscript and the autographs, the more likely it is to reflect the wording of the autograph. Also, the manuscripts which have a good pedigree overall are more likely to be correct in a given instance. A careless scribe in the second century may produce a less accurate document than a careful copyist working in the fifth century. Consequently, date is not an automatic preference, but must be combined with the known character of a given manuscript.
The majority of our manuscripts were written in areas where certain traditional wordings were duplicated repeatedly. That is to say, most manuscripts were copies (or copies of copies) of a regional archetype. Thus, a variety of patterns of readings emerged in each area, giving that locale a distinctive text form. A reading is considered genealogically solid when most of the manuscripts—especially most of the better ones—that belong to a certain text-type agree. This is evidence that the local ancestor of that text form in all probability contained that wording. For a discussion of which manuscripts belong to which text types, see Metzger, A Textual Commentary on the Greek New Testament, 14–16; Holmes, “New Testament Textual Criticism,” 59–60.
As was mentioned above, the Alexandrian, Western, and Byzantine are the three major text forms. The Alexandrian was produced especially in Egypt, the Western in Rome and the west (though also elsewhere), and the Byzantine mostly in the east. Most scholars believe that the Alexandrian and Western texts have roots early in the second century. The Byzantine text, however, was a later development, based largely on Western and Alexandrian manuscripts.
The best manuscripts—for genealogical solidarity—of a given text-type are those that do not have mixture from the other text forms. When the better Alexandrian manuscripts, for example, have the same wording, there is a high degree of confidence that the Alexandrian’s regional ancestor had that wording. This is true even though that archetype no longer exists. It is a simple deduction from the available evidence.
Thus, by using genealogical solidarity, scholars can posit the date of a reading within a text-type back to its regional archetype. This is similar to the deduction one would make upon meeting an extended family of 30 brown-eyed persons: the ancestors also most likely had brown eyes. Since the Alexandrian and Western texts have roots from early in the second century, when each of these text forms has genealogical solidarity, one can have confidence that such a reading is the reading of their regional archetypes and that this reading most likely originated in the second century.
A variant found in geographically widespread locations in the first three or four centuries of the Christian era is more likely to be original than one that is found in only one location. The geographical spread of witnesses that have the same wording is a significant factor in determining the wording of the autographs. It is less likely that witnesses agree because of agreed-upon changes when such witnesses are found throughout the Mediterranean world than if they are clustered in just Rome or Antioch.
Thus, if a third-century papyrus from Egypt, a third-century Latin text from Rome, and a third-century Father from Gaul all have the same wording in a given passage, they are most likely reflecting the wording of an earlier source. That is, geographical distribution shows that a reading was not produced by collaboration. Also, like genealogical solidarity, it is evidence that the reading is much earlier than any of the extant sources that have the reading. Geographical distribution shows—with a good deal of certainty—that a reading dates to before the time of the witnesses that have it.
What about readings found in geographically diverse witnesses from the fifth century and later? Christianity became not only a legal religion but the official religion of Constantine’s empire in the fourth century. After this date geographical distribution is no longer nearly as useful because by this time there would have been extensive mixture among the manuscripts as information was freely exchanged all over the Roman Empire.
Again we consider Mark 16:9–20. The great majority of Greek manuscripts have these verses, but the great majority of manuscripts were produced after the fourth century. The best Greek manuscripts in this passage are א (‘) and B (no papyri are extant for Mark 16). But they are only from one region, the Alexandrian. Yet Eusebius, writing in the early fourth century, and Jerome, writing almost a century later, both speak explicitly about the majority of manuscripts with which they were acquainted: such manuscripts, they say, almost always lacked these verses. In later centuries, the reverse situation was becoming the norm, as can be seen in the comments of Victor of Antioch (fifth/sixth century). And the earliest and best manuscripts of the three most important versions—the Latin, Coptic, and Syriac—give evidence of ending the Gospel at 16:8.
Thus, date and character, genealogical solidarity, and geographical distribution all contribute to the view that Mark’s Gospel ended at 16:8. In combination with the internal evidence, this becomes a compelling argument for most scholars that these verses were not part of the original Gospel of Mark.
These three divisions of external evidence—date and character, genealogical solidarity, and geographical distribution—are three keys to unlocking the external data and deciding which reading is most likely the basis for the other(s). These aspects of external evidence need to be rigorously examined and compared.
But in places where the early manuscripts disagree, where there is minimal geographical distribution as well as fracturing within an early text form, or where one of the readings is predictable (i.e. the kind of wording many scribes in different areas could create independently of one another), internal evidence correspondingly increases in significance.
This is the sense in which reasoned eclecticism offers a balanced approach to the two compartments of evidence: if internal evidence, for example, is inconclusive, while external evidence clearly favors one reading, then the external evidence outweighs the internal, and vice versa.
The textual variant that has the greater claim to authenticity will be found in the earliest, best, and most geographically widespread witnesses. It will fit the context and author’s style, and will be the obvious source of the rival reading(s)—the different readings can be clearly traced to changes made in copying the more authentic text. For the great majority of textual problems, the evidence is clear as to which reading is authentic and which is not.
However, often the external evidence seems to favor one reading while the internal evidence favors another. How do scholars decide in such cases? If a particular variant is found only in non-Greek manuscripts, in a few late manuscripts, or only in those that have a very poor pedigree—even if its internal credentials are excellent—it is rejected. In such instances, external evidence is deciding factor. With the many thousands of manuscripts extant today, unpredictable accidents and unknowable motives may be the cause of a stray reading here or there.
Sometimes the external evidence is solidly on the side of one variant, but a few significant manuscripts and compelling internal evidence support a different variant. In such instances, the second reading most likely reflects the autographic wording. Some problems cannot yet be solved with our present state of knowledge. Others can only be solved by persistence, patience, and a determination to uncover the true text.
1 The video is an excerpt from an edition of the John Ankerberg Show, entitled, “Which English Translation of the Bible is Best for Christians to Use Today?” The entire show is available for purchase here. This excerpt is publicly available on YouTube.
2 Daniel B. Wallace, “Textual Criticism of the New Testament,” ed. John D. Barry et al., The Lexham Bible Dictionary (Bellingham, WA: Lexham Press, 2016).