Excerpted for review purposes from: Leland M. Haines, Authority of Scripture,
Chapter 6: Greek Text, (Indiana, 2008)
Chapter 6: - Translations and the Greek Text:
Introduction: Should Shorter Readings be Preferred?
Omissions: Errors of Omission abound
Haplography: Jumps in text from 'like to like'
'Prefer the Shorter Reading': A Dead Principle
Byzantine Text-type: wrongly rejected
Summary: Old Canons of Textual Criticism unsound
Excerpted for review purposes from: Leland M. Haines,
Authority of Scripture,
Chapter 6: Greek Text, (Indiana, 2008)
Headings have been added for clarity and navigation purposes.
Translations and the Greek Text
Should Shorter and Harder Readings
Are the shorter and harder readings really the preferred method to establish the text? Are these sound reasons for the selection of "good readings"? Wide use of the shorter readings would not allow scribes to make accidental omissions and use of harder readings could result in many instances of unintelligible texts. Therefore it is obvious these rules cannot be applied without the use of common sense.
Would not one expect the apostles and their associates to write in a clear, easy-to-read style? Surely they were not clumsy and unrefined writers who used short and hard-to-understand language. This raises the question, Were deliberate changes made to improve their text? Does the Byzantine text show evidence of added words?
Errors of Omission Abound
Clark was one of the first to challenge the "shorter readings" concept after studying Cicero's oration. "The evidence yielded by such researches is not favourable to the hypothesis of extensive interpolation." [165 ] This study showed accidental omissions were much more common than scribal interpolation, and sometimes whole lines were omitted.
Clark then examined the shortest texts found in several B and א passages. "Number of character" for the Matthew 20:28; Luke 5:14; John 5:4; 7:53-8:11; Mark 16:9-20 passages showed "that the passages defend each other, and that the theory of interpolation is less likely than that of accidental omission." In these cases, he believes whole column(s) or page(s) in an archetype were omitted. [166 ] He also gives data for a suspected Luke 23:34 and 38 passage to show that 42 letters or two lines were omitted. [167 ]
Clark's second study again showed scribes often omitted one or more lines. [168 ] When he applied his "Longer Reading" to the Gospels and Acts, he concluded the Western text was better than Westcott & Hort's Alexandrian text. [169 ]
Pickering wrote that
Actually, a look at a good apparatus or at collations of MSS reveals that the "Byzantine" text-type is frequently shorter than its rivals. Sturz offers charts which show that where the "Byzantine" text with early papyrus support stands against both the "Western" and "Alexandrian" it adds 42 words and omits 36 words in comparison to them. The "Byzantine" comes out somewhat longer but the picture is not lopsided. Among the added words are 9 conjunctions and 5 articles but among the omitted are 11 conjunctions and 6 articles, which would make the "Byzantine" less smooth than its rivals. 
The 'shorter reading' principle has been questioned more recently by Royse. [171 ] He found
...[the textual] "critic is free to choose that reading which 'seems' best, and thus to introduce what often appear arbitrary choices of one reading instead of another." 
"Such findings raise serious questions about the truth of the principle that 'the shorter reading is to be preferred', and at the very least suggest that the simple statement of principle (even with the noting of certain exceptions) is an inadequate guide to the earliest period of the transmission of the New Testament text." 
Colwell wrote in "Method in Evaluating Scribal Habits: A Study of P45, P66, P75" [174 ] that the habits of each scribe in a singular reading was to make certain types of errors. "Colwell found that all three scribes of the early papyri had a marked tendency to omit words."  He wrote that
as an editor the scribe of P45 wielded a sharp axe. The most striking aspect of his style is its conciseness. The dispensable word is dispensed with. He omits adverbs, adjectives, nouns, participles, verbs, personal pronouns--without any compensating habit of addition. He frequently omits phrases and clauses. He prefers the simple to the compound word. In short, he favors brevity. He shortens the text in at least fifty places in singular readings alone. But he does not drop syllables or letters. His shortened text is readable. 
Colwell wrote that "a half dozen times",
the scribe of P75 chooses brevity. . . . The best example is John 12:38 where instead of "the word which he said" the redundant "which is said" is omitted. . . . One of his habits is to omit personal pronouns; he drops more than a dozen and adds one. 
In general Colwell found the scribe of P66 wild in copying. This often led to shortening the text. Colwell summarized the P66 scribe's habits:
P66 has an inclination towards omissions, it is not 'according to knowledge,' but is whimsical and careless, often leading to nothing but nonsense." 
Haplography: Jumps from 'like to like' in text
What mechanism would account for a shortening of the text? Colwell answers this:
The leap from the same to the same is a familiar phenomenon to all students of manuscripts. It is really the case of the misplaced scribe. The scribe loses his place, looks around and finds the same word, or at least the same syllable or letter, and starts from there. If he looks ahead to find his place, the result is a gap in the text. . . .
P66 has 54 leaps forward, and 22 backwards. . . .
P75 has 27 leaps forward, and 10 backwards,
P46 has 16 leaps forward, and 2 backwards.
From this it is clear that the scribe looking for his lost place looked ahead three times as often as he looked back. In other words, the loss of position usually resulted in a loss of text, an omission. 
Colwell and Tune commented about the work of correctors in a study of John 11:
The largest single cause of the singulars in our set of readings is the omission or the contraction of words (about 65 per cent of the instances). In order, the other causes are: spelling or inflectional differences, substitution of other words, and addition of other words (seldom). [180 ]
The corrector's work shows the scribes often made omissions; that is, they made the text shorter but seldom longer.
The 'Prefer the Shorter Reading' Principle is Dead
When we consider the statement, "the shorter reading is preferable," can we see any reason, apart from repetition and tradition, why it should be right or wrong? We can produce reasons for thinking sometimes that the longer text is right and sometimes that the shorter text is right, but that will not demonstate our maxim. 
The most that can be said about the "shorter reading" theory is that scribes often deleted words.
As Royse wrote:
Such findings raise serious questions about the truth of the principle that the shorter reading is to be preferred, and at the very least suggest that the simple statement of the principle (even with noting of certain exceptions) is an inadequate guide to the earliest period of the transmission of the New Testament text. What one has are shorter texts and longer texts, and one must have some justification for making the claim that scribes have lengthened the shorter one rather than shortened the longer one. 
Royse gives examples of the text being shortened with "no good reason."
The Byzantine Text wrongly Rejected
The implication of scribes' shortening a text has major implications on acceptance of the generally shorter Alexandrian text. We write "generally" shorter because it would be unfair to call it the shorter text. Wallace notes that out of the 6,577 differences between the Majority Text [183 ] (the Byzantine text) and the Alexandrian text (UBS3, NA26), in places the Alexandrian text is longer and 1,589 where it is shorter, [184 ] or 29% of the time it is shorter.
"One might, with some justification, wonder why the textual critics responsible for UBS3 seem to suspend this canon [shorter text is the best ] almost exactly as many times as the Byzantine text had shorter readings (i.e., if the Western readings are not in the purview of the discussion)." [185 ]
This observation shows their prejudices against the Byzantine text.
The question may be asked, were many of the Alexandrian shorter readings due to scribal errors? Also, it may indicate the middle of the road Byzantine text is better than the shorter Alexandrian and the longer Western texts.
There may be another explanation for the differences in text length. Did the authors make different editions of the manuscripts? We know that modern authors often revise their books. The New Testament authors may likewise have made revisions.
It is highly unlikely there would be so few text-types if they were caused by scribal correction. Scribal correction surely would have caused hundreds and hundreds of text-types. Blass has suggested that Luke brought out two editions of his Gospel in order to account for the Western text to be 8.5 percent longer than the Alexandrian.
Hort wrote, as mentioned earlier, the Byzantine texts "run smoothly and easily in form." [186 ] Instead of this being a mark against the Byzantine text, isn't this what would be expected of the apostles' writings? Few would think that their writings would be rough and hard-to-read.
Salmon wrote that Hort's harder readings "as a note of genuineness is a reading implies error on the part of a sacred writer." [187 ] It would be more natural to expect harder reading to come about though scribes dropping out words. Thus the two chief "shorter" and "harder" criteria go together as signs of scribal errors instead of as a sign of originality.
In summary, these rules at best must be applied very carefully to individual texts and are not sound principles to accept or reject certain text-types. Westcott & Hort's use of them to choose the best text-type is inappropriate.
Original Footnotes:165. Albert C. Clark, Recent Developments in Textual Criticism, Oxford: Clarendon Press, 1914, p. 10. See also pp. 20, 21.
166. Ibid., pp. 25-26.
167. Ibid., p. 27.
168. Albert C. Clark, The Descent of Manuscripts, Oxford: Clarendon Press, 1914.
169. Albert C. Clark, The Primitive Text of the Gospels and Acts, Oxford, 1914.
170. Pickering, op. cit., p. 222.
171. James R. Royse, "Scribal Habits in the Transmission of New Testament Texts," pp. 139-61, in Wendy Doniger O'Flaherty, ed., The Critical Study of Sacred Texts, Berkeley, Cal.: Berkeley Religious Studies Series, Graduate Theological Union, 1979.
172. Ibid., p. 145,
173. Ibid., p. 155.
174. Ernest C. Colwell, Studies in Methodolgy in Textual Criticism of the New Testament. Grand Rapids: Eerdmans, 1969, p. 112. Referred to by Royse in O'Flaherty, op. cit., p. 154.
175. Royse, op. cit., p. 154
176. Studies, op. cit., pp. 118, 119.
177. Ibid., p. 121.
178. Ibid., p. 123.
179. Ibid., p. 112.
180. Ibid., p. 62.
181. G. D. Kilpatrick, "Griesbach and the Development of Text Criticism," a paper given at Griesbach Bicentennial Colloquium, Munster, July 26-31, 1976, p. 6.
182. Royse, op. cit., 1979, p. 155.
183. Zane C. Hodge and Arthur L. Farstad, editors, The Greek New Testament According to the Majority Text, Nashville: Thomas Nelson, 1985.
184. Daniel B. Wallace, Some Second Thoughts on the Majority Text, Dallas: Dallas Theological Seminary and Galaxie Software, 1997, p. 277.
185. Ibid., footnote 33.
186. Westcott & Hort, Introduction, pp. 115-16.
187. Salmon, op. cit., p. 26.