Even more to the point are the implications of the Donatist movement which developed right after the Diocletian campaign. It was predicated in part on punishing those who had betrayed their manuscripts to destruction during the recent persecution. Obviously, some did not betray their MSS or there would have been no one to judge the others. Moreover those whose commitment to Christ and His Word was such that they who withstood the torture would be the most careful about the pedigree of their MSS. Hence, the purest specimens would have been the most likely to have survived. The main stream of transmission would have this fountain as its origin.

Since the Byzantine (majority) textform dominates over 90% of the extant MSS, those who wish to reject it cannot concede the possibility that the transmission of the text was in any sense normal. If it had been, then the majority must reflect the original wording, especially since the consensus is so massive. Thus, it is argued that the "ballot box" was "stuffed" - that the Byzantine text was imposed by ecclesiastical authority, but only after it was systematically concocted from other older texts in the early 4th century. Yet, as we have already stated, there exists absolutely no historical evidence to support this conjecture.

Also, numerous studies have demonstrated that the mass of Byzantine MSS are not monolithic; there are many distinct strands or strains of transmission, seemingly independent. Some of these go back to the 3rd century (if not earlier). This is demonstrated by Codex A in that it conflates some of those strands in Revelation. Asterius, a pupil of Lucian of Antioch (d. 341) used MSS that were clearly Byzantine. His Syntagmation is generally believed to have been published before the Council of Nicea (325); accordingly, his MSS would have been 3rd century.

But why is "the shorter reading to be preferred?" Because, we are told, scribes had a propensity to add bits and pieces to the text. But that would have to be a deliberate act, for it has been demonstrated that accidental loss of place (a parablepsis) results in omission far more often than addition. For the most part, the only way to add accidentally is to copy part of the text twice, however the copyist would have to be really drowsy not to catch himself at it. So, any time a shorter reading could be the result of parablepsis, it should be viewed with suspicion.

Even when deliberate, omission should still be more frequent than addition. If there is something in the text that someone doesn't like, it attracts his attention and he may be tempted to do something about it. Correspondingly, it requires more imagination and effort to create new material to add than to delete that which is already there. Material suggested by a parallel passage would be an exception. Further, it has been demonstrated that most scribes were careful and conscientious, avoiding even unintentional mistakes. Those who engaged in deliberate editorial activity were really rather few, but some were flagrant offenders (like Aleph in Revelation).

Why is "the harder reading to be preferred?" The assumption is that a perceived difficulty would motivate an officious copyist to attempt a "remedy". But in the case of a such a presumed deliberate alteration, how can degrees of "hardness" actually be ascribed? We don't know who did it or why. Due allowance must be made for possible ignorance, officiousness, prejudice, and malice. Moreover, this canon is unreasonable for the more absurd a reading is, whether by accident or design, the better is its claim to be "original" since it will certainly be the "hardest".

It does not take a prophet or an apostle to see that this canon is wide open to satanic manipulation, both in the original creation of variants and in their present day evaluation. Nevertheless, since it is demonstrable that most copyists did not make deliberate changes, where there is massive agreement among the extant MSS this canon should not even be considered. Indeed, where there is massive agreement among the MSS none of the subjective canons should be used - they are unnecessary and out of place. Of the more than 6,000 differences between UBS3 and the Textus