BASIC PRINCIPLES OF TEXTUAL CRITICISM

According to Carl Holladay, text criticism is the attempt to "establish the original wording or form of the biblical text insofar as this is possible" (HarperCollins Dictionary of the Bible, 141). This is a formidable task for students of the NT, who encounter evidence from literally thousands of ancient manuscripts, versions, citations, lectionaries, and the like. How, given multiple texts that read differently, does one arrive at a most likely reading?

Over the decades, New Testament scholars have developed some canons for textual analysis. Most of them are grounded in simple common sense, but two rules stand apart as the most basic.

See Bruce M. Metzger, The Text of the New Testament, ch. 8, for a fuller treatment of many of the topics discussed here. See also Bart D. Ehrman, The New Testament: A Historical Introduction to the Early Christian Writings (2d ed.), 442-50, for an exceptionally lucid summary.

1. Choose the reading that best explains the origin of the others. In theory, at least, there originally was a single text for each NT document. As copies multiplied, so did the textual discrepancies. But (in theory) each discrepancy may be traced to specific moments in the transmission of the text. A scribe may intentionally have altered the text for any of a number of reasons, or perhaps a transcriptional error has come into play. Ultimately (in theory), a single manuscript is the origin of all other readings.
2. In order to judge a particular reading, one must first construct its history. In other words, one should be able to explain how an hypothetical original text gave rise to other readings (Metzger 207).

These two principles provide the larger framework, but other principles enable us to make reasonable judgments in specific cases.

First, there is external evidence. External evidence has to do with the various manuscripts themselves.

1. The date of a particular reading is important. In this case, it is not primarily the age of the manuscript that matters, but our ability to trace particular readings to types of texts with strong and ancient pedigrees. Some later mss. tend to agree strongly with very ancient ones; by association, those later mss. may carry more authority than some mss. that actually predate them.
2. The geographical distribution of a reading is significant, though often difficult to discern for beginning readers.
3. The text-type of a manuscript source can be significant. Witnesses are weighed rather than counted. Readings supported by both Alexandrian and Western witnesses are usually superior. (Except B D G in Paul, where B is somewhat Western.) Often the Alexandrian text alone will provide the most probable reading. But every case must be judged on its own merits; text-type alone rarely settles the issue.
   • Most manuscripts belong to the Koine or Byzantine type. While the Koine witnesses are the most numerous -- and provide the basis for the Textus Receptus from which the KJV derives -- they generally represent an inferior text-type. According to Metzger, the Koine text reflects the combination of elements from other text-types by Lucian of Antioch in the late third century (Text, 212).
     • The most important witnesses include A (Alexandrinus, except for Paul) E F G H and most minuscules (largely medieval mss.).
   • The Western text-type provides the clearest example of scribal innovation (e.g., the Western Text of Acts is about ten percent longer than other text-types.) It is also very ancient.
     • The most important witnesses are D (Bezae) and the Old Latin versions. For Acts, p²⁹, p³⁸, and p⁴⁸.
   • The Caesarean text-type shares common features with the texts used by Origen while he was in Caesarea. Its readings are quite diverse in their affinities with other text-types.
     • The most important witnesses include Q (the Greek letter theta represents Codex Koridethi) W fam. 1 fam. 13.
   • The Alexandrian text-type derives from scholarly circles in ancient Alexandria and can be traced to the second century.
     • Its primary witnesses include p⁴⁵ p⁴⁶ p⁷⁵ ) (the Hebrew letter Aleph represents Codex Sinaiticus) B (Vaticanus) Clement of Alexandria, and some of Origen's writings. For Paul A (Alexandrinus) shows Alexandrian affinities.

Second, there is internal evidence. Internal evidence has to do with the intrinsic likelihood of the various readings.

1. The more difficult reading is to be preferred. Scribes are more likely to smooth things out than to rough them up.
2. However, "the reading deemed original should be in harmony with the author's style and usage elsewhere" (Metzger 217).
3. The shorter reading is to be preferred. Scribes tend to add rather than subtract material. Exceptions include parablepsis arising from homoeoteluton or the omission of material for stylistic or theological readings (see rule 1).
4. When some readings parallel other biblical texts, the divergent reading is to be preferred. Again, scribes are more likely to harmonize sacred texts than to create discord among them.