A PRIMER ON NEW TESTAMENT TEXTUAL CRITICISM (IN MANAGEABLE, BITE-SIZED CHUNKS)
© 2015, by John Tors. All Rights Reserved.
Understand each bite before moving on to the next one!
The First Bite
A) The New Testament (NT) was written in Koine Greek in the 1st-century AD, but the printing press was not invented (in the West) until the mid-15th century AD. This means that for about 1,400 years, books – including the NT – could only be reproduced through copying by hand. (A “manuscript” is a handwritten document.)
B) Through these centuries, very many copies were made of the NT, and a great number of them survive to this day (though only a small minority of all that were made). There are currently some 5,795 full or partial NT manuscripts known to exist. Most were copied carefully by professional scribes, but some were done by amateurs (some of whom didn’t even know Greek!) and are of low quality and error-ridden.C) The “manuscript attestation” for the NT (i.e. the number of manuscripts and how close they are to the date of composition of the original writing) is far better than for any other ancient writing. Our earliest NT manuscript, a small fragment designated P52, is considered to date from about 35 years after the date of the original composition.
D) The following chart shows the distribution of the manuscripts by century. The further back you go in time, the fewer the manuscripts from that century.
A) Scribes copying by hand inevitably made mistakes, so no extant NT manuscript is perfect. Each mistake creates a “variant reading” at that point in the text compared to what is at that point in other manuscripts. For example:
- At 1 Timothy 3:16b most manuscripts have “God was manifested in the flesh” but some have “Who was manifested in the flesh”
- At John 1:18b most manuscripts have “the only begotten Son” but some have “the only begotten God.”
- Most manuscripts include John 5:4 but a few omit it.
Textual criticism is the practice of comparing the readings at each point of variation among the manuscripts to determine which one was the original reading and which are errors.
B) As the total number of manuscripts increases, so does the total number of variant readings.
C) However, it is essential to note that each scribe made his own mistakes at his own random places, so where his copy had mistakes, all (or almost all) of the others did not have mistakes. So the original reading could always be found by determining what reading is found in the majority of the manuscripts at each point of variation. Errors could thus be cross-corrected so that the original text would be preserved.
A) The Enlightenment Era began in the 17th century and was fully realized by the 18th. The ethos of this era among the intellectuals and leaders was the view that “Man is the measure of all things,” that reason and rationality are the means to attain to truth, and faith and revelation are to be discarded. Thus the goal became to drive God out and that required destroying the credibility of the Bible.
B) To do this, the intellectuals developed what was variously called Biblical scholarship, Biblical criticism, Higher criticism, or Scientific scholarship. It started with the assumption that all things supernatural were impossible, and sought to explain the Bible as a purely human product about a man who was nothing more than a human being. It was “science falsely so called,” because the actual evidence was trumped by fiat proclamations – and, like “the emperor’s new clothes,” all “intelligent” people had to agree with the assertions of these scholars.
C) In particular, the Bible was attacked on three fronts (what I call the “Three-Headed Monster”): Historical Criticism (who wrote the Gospel books and epistles and when and how), Textual Criticism, and Darwinism. Each of these was fundamentally built on unproven liberal paradigm assumptions as needed to destroy Biblical credibility.
The Fourth Bite
Textual criticism may actually be the most dangerous of the three heads, for two reasons:
- The conclusions of historical criticism and the claims of Darwinism are overt attacks against the claims of Christianity, and so they are seen as the threats that they are and are opposed (to a greater or lesser extent) by evangelical scholars. Textual criticism, however, is a much more subtle attack and so is not recognized as a threat by evangelicals. They mistakenly believe that it is a neutral scholarly endeavour and therefore they embrace it wholeheartedly.
- Historical criticism and Darwinism are external attacks that say the Bible is wrong. Their claims can be rejected and the inerrant Bible held on to. But the textual criticism promoted by liberals and accepted wholesale by evangelical scholars inserts errors into the text of the Bible itself. For example:
The beginning of the gospel of Jesus Christ, the Son of God. As it is written in Isaiah the prophet, “Behold, I send my messenger before your face, who will prepare your way.” (Mark 1:1-2 ESV. Also NIV, NASB, RSV, HCSB et al.)
As written thus, it is a mistake, as the first part of the quote is from Malachi 3:1, and Isaiah, and it is not “written in Isaiah the prophet.” If this reading is in the Bible, then there is an undeniable error in the Bible. This reading is found in only sixteen manuscripts.
The beginning of the gospel of Jesus Christ, the Son of God. As it is written in the Prophets: “Behold, I send My messenger before Your face, Who will prepare Your way before You.” (Mark 1:1-2 NKJV)
Since the first part of the quote is from Malachi 3:1 and the second from Isaiah 40:3, the reading “in the prophets” is completely correct. This reading is found in about 1,700 manuscripts and another 2,000 lectionaries. So why is the minority reading that introduces an error into the Bible the one that is put into every major modern translation except the NKJV? Because putting errors into the Bible is exactly what textual criticism has been developed to do, as we shall see.
The Fifth Bite
A) As with historical criticism, the foundations of text-critical theory were laid by German rationalist scholars. The first step was to separate the extant manuscripts into “text types” or “families:” the Byzantine, which included the vast majority of manuscripts, mostly dating from the 6th century and later, were carefully done and had little variation in their text; the Alexandrian, which consisted of a small number of manuscripts from Egypt that were older but had texts that differed dramatically from one another; and the Western, which included only one Greek manuscript.
- The shorter reading is to be preferred, “for scribes were much more prone to add than to omit.”
- The more difficult reading is to be preferred e.g. “That reading is rightly considered suspect that manifestly gives the dogmas of the orthodox better than the others.”
- The reading that best explains the origin of the others is to be preferred.
- The reading that creates discrepancies with other quoted or parallel material is to be preferred.
C) Notice that (1) the rules were all designed to make sure that wherever there is a choice between variants one of which puts an error or a problem into the text and one of which doesn’t the textual critic must always choose the variant that puts an error or problem into the text (which is always, by definition, the “more difficult reading” and the one “that creates discrepancies”); and (2) the rules presuppose that scribes who believed the Bible was the word of God nevertheless took it upon themselves to freely change it (to fix the supposed original errors and problems).
D) It is crucial to note that these rules did not come from any actual study of scribal habits. No proof of any sort was offered; they were simply proclaimed to be true – and yet they are followed to this day even by evangelical textual critics.
E) Because of these rules, variants that introduce an error or make the Bible look bad are taken to be the original readings, regardless of how few MSS they are in.
The Sixth Bite
A) Around the middle of the 19th century, two old manuscripts of the Bible came to light. Codex Sinaiticus (designated “א”) and Codex Vaticanus (designated “B”). They dated to around the middle of the 4th century AD, and so were about one hundred years older than the earliest one previously known at that time.
B) What liberal scholars loved most about א and B is that both were missing the last twelve verses of the Gospel According to Mark (Mark 16:9-20). By insisting that these were the most reliable NT manuscripts and insisting on “literary dependence” (i.e., that the authors of the Gospel books were not writing independently but were copying one from another) and on “Markan priority” (the claim that the Gospel According to Mark was the first one written – an idea that no one believed until the discovery of א and B, after which Markan priority became scholarly orthodoxy in very short order), liberal scholars were able to advance the following argument: (1) Mark’s Gospel book was the first one written; (2) Matthew and Luke copied from Mark, so they offer no independent testimony, and (3) only Mark’s counts, and Mark’s doesn’t actually have a resurrection story (based on א and B), which means that the idea of Jesus’ resurrection was a legend added later, and not an historical fact!
C) To advance this idea, it was necessary to insist that א and B are good, reliable manuscripts. The problem is that we knew very soon that this was not even remotely true. As Dean John Burgon pointed out,
Codex Vaticanus omits words or whole clauses 1,491 times in the Gospel accounts alone and is “disfigured throughout with repetitions.” Codex Sinaiticus “abounds with errors of the eye and pen … On many occasions ten, twenty, thirty, forty words are dropped through very carelessness. Letters and words, even whole sentences, are frequently written twice over, or begun and immediately cancelled; while that gross blunder …. whereby a clause is omitted because it happens to end in the same words as the clause preceding, occurs no less times than one hundred and fifteen times in the New Testament.”
Herman Hoskier did a careful comparison of these two codices and found more than 3,000 significant disagreements between them in the Gospel books alone, which actually exceeds the number of times they agree with each other (and which, by the way, means they cannot correctly even be classified as belonging to the same so-called “text type.”) Thus, Codex Sinaiticus and Codex Vaticanus should be afforded no credibility by anyone serious about textual criticism.
Liberal scholars, of course, blithely ignored this. Not only did א and B gave them what they wanted – a means to discredit the resurrection – but the corrupt nature of these manuscripts gave them plenty of opportunities to find “shorter readings,” “more difficult readings,” and “readings that create discrepancies with other quoted or parallel material” that they could claim were the original readings and thus insert more errors into the actual text of the Bible. Evangelicals, strangely, continue to follow along on this to this very day.
The Seventh Bite
A) To enhance the credibility of א and B, which were described as being members of the Alexandrian text-type – and, indeed, to promote the Alexandrian text type, with all its variations and discrepancies, as the best text-type (i.e., closest to the original NT text) – it was necessary to destroy its rival, the Byzantine text type.
B) This was done in 1881 by two British scholars, B.F. Westcott and F.J.A. Hort. Their main argument was the “genealogical method.” They argued that we cannot assume that the majority of manuscripts are right, because many manuscripts may all be copied from the same original, and so are not independent witnesses to the original text. Instead of counting manuscripts, they said, we must study their genealogical tree (i.e., which manuscript was copied from which), to determine which were earlier and so which had the original readings. This approach, they said, proves that the Byzantine text is a later, corrupted form of the New Testament text. Metzger speaks of
the force of the genealogical method, by which the later, conflated text is demonstrated to be secondary and corrupt.
C) This was a complete scam. Westcott and Hort never actually built a genealogical tree of New Testament manuscripts, and this cannot be done as there are simply not enough manuscripts, particularly from the early centuries, to construct such a tree. Westcott and Hort merely raised the theoretical possibility that the majority might be wrong, and then tried to argue that “conflate readings” in the Byzantine manuscripts show that the entire text type is indeed secondary and corrupt. (On the contrary, they show the exact opposite.) This scam gave the liberal scholars what they wanted – an excuse to throw out the vast majority of New Testament manuscripts, which were generally good, in favour of a handful of corrupt, error-ridden ones – and so the Westcott-Hort theory was accepted as scholarly orthodoxy.
The Eighth Bite
A) During the 20th century, and well over a century after Griesbach’s canons became the scholarly orthodoxy for textual criticism, some scholars finally got around to studying actual scribal habits by examining the manuscripts themselves to see what scribes had done in real life (as opposed to what Griesbach proclaimed that they did). Some of these studies are:
Tarelli, C.C. “Omissions, Additions, and Conflations in the Chester Beatty Papyri,” JTS, 1938
Colwell, Earnest Cadman. “Scribal Habits in Early Papyri: A Study in the Corruption of the Text.” in Hyatt, J. Philip. ed. The Bible in Modern Scholarship: Papers read at the 100th meeting of the Society of Biblical Literature. London: Carey Kingsgate Press, 1965.
P.M. Head, “The Habits of New Testament Copyists. Singular Readings in the Early Fragmentary Papyri of John,” Biblica 85, 2004
Royse, James R. “Scribal Habits in Early Greek New Testament Papyri.” Atlanta: Society of Biblical Literature, 2008. [This one is a book almost 1,000 pages in length.]
B) What all of these studies found is that Griesbach’s rules are wrong. In fact, they are backwards. Scribes did not take it upon themselves to alter the text, and intentional changes are rare. By far the most common mistake is accidental omission, a random error.
C) This means that the text-critical theory upon which our evangelical scholars are basing their textual decisions is wrong at its very core.
The Ninth Bite
A) Beginning in 1889, archaeologists found numerous old papyrus manuscripts of the New Testament, some of which were much older than א and B; some went back to the 2nd and 3rd. (They were all only partial and usually fragmentary.) Scholars hailed such early papyri as P66 and P75, though in fact these are also very corrupt manuscripts. P66, in fact, is far worse than א or B.
B) The texts found in these early papyri did not neatly align with the Alexandrian, Byzantine, or Western text type, but were described as “mixed texts.” However, as Kurt Aland, the doyen of textual criticism in the 20th century and editor of the Nestle-Aland Greek New Testament, pointed out, text types cannot be mixed before they exist. He explains the significance of this:
It is impossible to speak of mixed texts before recensions have been made (they can only follow them), whereas, the NT manuscripts of the second and third centuries which have a ‘mixed text’ clearly existed before recensions were made … The simple fact that all these papyri, with their various distinctive characteristics, did exist side by side, in the same ecclesiastical province … is the best argument against the existence of any text-types, including the Alexandrian and the [Byzantine]. We still live in the world of Westcott and Hort with our conception of different recensions and text-types … We can no longer take their conception as valid.
C) This means that the whole concept of text-types is invalid and should be abandoned. We have known this since 1965, but it has been blithely ignored by liberal scholars, and evangelicals have uncritically followed in their wake.
The Tenth Bite
B) As we have seen, the vast majority of scribal errors are random.
C) The combination of (A) a large, stratified data base and (B) random event variants means that we can and must apply statistical analysis to the issue. True, this is beyond the skills set of the artsy textual critics, both liberal and evangelical, but that does not mean we cannot do it; indeed, we must.
D) To do this, we need actual, real-life numbers. Textual critics make statements like the following:
A theoretical presumption remains that a majority of extant documents is more likely to represent a majority of ancestral documents at each stage of transmission than vice versa … [but this] presumption is too minute to weight against the smallest tangible evidence of other kinds.
This statement may impress an artsy Bible scholar, but not someone who understands mathematics. Really, Mssrs. Westcott and Hort? The presumption is “too minute”? What is the magnitude of the presumption? How did you calculate it? How do you weigh the “tangible evidence” to see if it exceeds the weight of the presumption? (Don’t expect an answer from the textual critic; you may as well be asking your dog to recite Shakespeare. In Hungarian.)
E) Let us work with some actual numbers now. P66, with its 400 itacisms and 482 other singular errors, has about one error for every seventeen words. This is about as careless as the manuscripts get. To give the benefit of the doubt to the other side, let us assume all scribes were this sloppy, and use 1 error per 17 words as the error rate. Further, let us assume that only five copies were made of each original New Testament book.
For an error to infect the majority of manuscripts, it would have to enter at least three of these five “first-generation” copies. What is the chance that scribes would make a mistake at the same point in three copies? It is 1/17 x 1/17 x 1/17 = (1/17)3 = 1 in 4,913. Nor would it be enough to make a mistake the same point; it would have to be the same mistake.
Next, if five copies are made of each first-generation copy, there will be 25 second-generation copies, and for an error to infect the majority of manuscripts at this point, it would have to enter at least thirteen of them. What is the chance that scribes would make a mistake at the same point in thirteen copies? It is 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 x 1/17 = (1/17)13 = 1 in 9.9 million billion. And, again, it would have to be the same mistake in all thirteen corrupted copies.
The upshot of all this is that the chance of any erroneous reading ever being found in a majority of manuscripts is vanishingly small. And that’s not a fiat proclamation; it is conclusion based on actual mathematics.
F) What this means is that in textual criticism, the focus should be on the readings, not on the manuscripts in which they are found. Find the reading that is in the majority of the manuscripts and you have the original reading.
The Eleventh Bite
A) Despite the facts outlined above, liberal scholars continue as always, using the canons and championing an errant text based on corrupted manuscripts while studiously ignoring the Byzantine manuscripts.
B) Evangelical scholars also continue in these old (and incorrect) ways. The problem is that the paradigm assumptions of textual criticism have been set by liberal scholars, and evangelical scholars were taught these in seminary as axioms and have accepted them without any serious question.
C) Having absorbed the scholarly orthodoxy, evangelical scholars are not willing to re-examine them, regardless of the evidence against it.
D) The result of all this is that long-dead German rationalists and their liberal allies tell us that we should consider the original readings of the word of God to be those that are more unorthodox, more difficult, lacking, internally contradictory, and erroneous, and virtually all of our modern scholars and modern Bible translations – evangelical or otherwise – slavishly do that very thing.
E) Undeniably, these claims of textual criticism end all belief in inerrancy, because they ensure that errors are considered the original readings. It is not all surprising, then, that almost all evangelical scholars have quietly abandoned the doctrine of inerrancy (at least in any meaningful sense i.e., inerrancy that means “no errors”). And that, in turn, fundamentally affects how seminary-trained pastors view the Bible. If we are looking for reasons to explain the drastic decline and enervation of the church, we do not have to look far beyond this.
NOTE: See our companion article “GIGO: Unearthing a Decisive New Tipping Point for Textual Criticism” for more fascinating facts about textual criticism
APPENDIX 1: THE CASE OF JOHN 5:4 AS AN EXAMPLE
John 5:4 doesn’t exist in your Bible if you are using an NIV, ESV, NASB, HCSB, RSV, NRSV, or any major modern translation other than the NKJV. The text jumps straight from 5:3 to 5:5; v. 4 has been omitted. In this case, the omission of this verse doesn’t introduce an actual error into the text, but it is the “more difficult reading,” because it makes the Bible look silly. Let us examine this textual issue, beginning with the reading in the ESV:
Now there is in Jerusalem by the Sheep Gate a pool, in Aramaic called Bethesda, which has five roofed colonnades. 3 In these lay a multitude of invalids—blind, lame, and paralyzed. 5 One man was there who had been an invalid for thirty-eight years. 6 When Jesus saw him lying there and knew that he had already been there a long time, he said to him, “Do you want to be healed?” 7 The sick man answered him, “Sir, I have no one to put me into the pool when the water is stirred up, and while I am going another steps down before me.” (John 5:2-7, ESV)
The story here makes no sense at all. How is “Sir, I have no one to put me into the pool when the water is stirred up” an answer to the question, “Do you want to be healed?” What does being healed have to do with being put into a pool? It’s just babble.
On the other hand, here is the account in the NKJV:
Now there is in Jerusalem by the Sheep Gate a pool, which is called in Hebrew, Bethesda, having five porches. 3 In these lay a great multitude of sick people, blind, lame, paralyzed, waiting for the moving of the water. 4 For an angel went down at a certain time into the pool and stirred up the water; then whoever stepped in first, after the stirring of the water, was made well of whatever disease he had. 5 Now a certain man was there who had an infirmity thirty-eight years. 6 When Jesus saw him lying there, and knew that he already had been in that condition a long time, He said to him, “Do you want to be made well?” 7 The sick man answered Him, “Sir, I have no man to put me into the pool when the water is stirred up; but while I am coming, another steps down before me.”
Here the man’s answer makes perfect sense, as the connection between being put into the pool and being healed is stated. The difference, of course, is the inclusion of v. 4, which is missing from the ESV, NIV, NASB, HCSB, RSV, NRSV et al.
Does the verse belong there? Let us look at the evidence:
The verse is missing in eleven manuscripts, including five that are extremely corrupted (P66, p75, א, B, and D = Codex Bezae Cantabrigiensis) and three that are late (one each from the 9th, 10th, and 12th centuries). It is present in about 1,375 manuscripts of the Gospel According to John and therefore, as we have seen, is certainly authentic.
The critic who wants to claim that v. 4 is not authentic must assert that a scribe took it upon himself to alter the Bible by adding this passage, but, as we have seen, the evidence indicates that scribes did not do such a thing.
If v.4 is authentic, its absence in a few manuscripts is accounted for by accidental omission by a scribe, and, as we have seen, the evidence indicates that this was by far the most common source of variants in the New Testament.
Every line of evidence, therefore, makes it clear that v. 4 is indeed authentic. The scholars who say that v. 4 is not part of the original text have not all individually studied the matter; they simply checked a standard book on textual criticism – which passes on the scholarly orthodoxy based on liberal paradigm assumptions – and in turn pass on what they found there. Their pronouncements do not override the actual evidence that has herein been presented.
Westcott and Hort offered four lines of argument against the Byzantine text, and these were seen as sufficiently convincing that it became axiomatic among scholars that the Byzantine text was secondary and corrupt. But this assessment of the Byzantine text was achieved under false pretences, as every argument offered by Westcott and Hort can now be seen to be bogus.
As we have seen, Westcott and Hort’s trump card was the genealogical method, “by which the later, conflated [Byzantine] text is demonstrated to be secondary and corrupt.” As we have seen, however, the genealogical method was never applied to actual manuscripts by Westcott and Hort, nor can it be, so their argument against the testimony of the majority of the manuscripts was never actually made in real life and should never have been accepted – particularly since rigorous mathematical analysis unquestionably disproves it. So Westcott and Hort’s theory should rightly be seen as dead in the water from the very beginning. Nevertheless, let us examine their other arguments.
Having failed to apply the genealogical method to the actual manuscripts, Westcott and Hort did a bait-and-switch, arguing that if the manuscripts of a text type are characterized by secondary readings, then all of the manuscripts of that text type are to be considered secondary and corrupt: “The evidence obtained in this manner is Internal Evidence, not Genealogical,” they argued, “but the validity of the inferences depends on the genealogical principle that community of reading implies community of origin.”
What is this “internal evidence” that Westcott and Hort proferred? They made three claims:
The Byzantine text “contained combined or conflate readings which are clearly composed of elements current in earlier forms of text.”
There are no distinctively Byzantine quotes anywhere in the writings of the ante-Nicene Church Fathers.
The Byzantine readings are obviously late since the Byzantine text type is “conspicuously a full text. It delights in pronouns, conjunctions, and expletives and supplied links of all kinds, as well as in more considerable additions … it presents the New Testament in a form smooth and attractive, but appreciably impoverished in sense and force.”
Let us examine each of these claims in turn. First, conflate readings are readings of the sort
… and were continually in the temple praising and blessing God. Amen.” (Luke 24:53)
where in some manuscripts you find “and were continually in the temple blessing God”, and in others “and were continually in the temple praising God”.
According to Westcott and Hort, conflate readings were “clearly composed of elements current in earlier forms of text.” Let us consider Luke 24:53, which was one of their prime examples of this.
The reading “and were continually in the temple praising and blessing God” appears in about 1,250 manuscripts of Luke and is in the NKJV.
The reading “and were continually in the temple blessing God,” is found in only four corrupt manuscripts and is found in the ESV, NIV, NASB et al.
One (and only one) corrupt manuscript, Codex Bezae Cantabrigiensis, reads “and were continually in the temple praising God.”
Westcott and Hort insisted that the compilers of the Byzantine text, seeing “blessing” in some manuscripts and “praising” in others (though it’s only in one), couldn’t decide which was the original and so joined (conflated) the two and created a new reading and put that into the Byzantine text (although they knew it could not be the original one, since they themselves had just created it).
However, if “blessing” was the original reading, as Westcott and Hort and their acolytes claim, they must first explain where the reading “praising” came from before they can claim that the Byzantine compilers conflated the two. Yet Metzger himself cannot come up with even a suggestion as to why a scribe would have changed “blessing” to “praising.” (See Metzger, Bruce M. A Textual Commentary on the Greek New Testament, Second Edition. Stuttgart: Deutsche Bibelgesellschaft, United Bible Societies, 1994, pp. 163-164.) But without someone creating the reading “praising,” there are not two readings available to be conflated.
On the other hand, if “praising and blessing” is the original reading, then it is clear why it is found in the overwhelming majority of the manuscripts, and the alternate readings “praising” and “blessing” are readily explained as due to accidental omission – which is by far and away the most common error scribes committed – by a few scribes, which would also explain why these readings are in so few manuscripts.
Clearly, readings of the sort “praising and blessing” are not conflated readings at all; on the contrary, “blessing” and “praising” are truncations of the original reading. Therefore, such readings, far from proving that the Byzantine text is “secondary and corrupt,” do the exact opposite. Although this is undeniable and obvious, liberal scholars in their lust to introduce errors into the NT text and undermine the evidence for the resurrection of Jesus simply ignore all this so that they can discard the Byzantine text. The fact that so many evangelical scholars uncritically follow the liberal lead is truly disturbing.
Second, Westcott and Hort claimed that
There are no distinctively Byzantine quotes anywhere in the writings of the ante-Nicene Church Fathers.
However, Edward Miller presented a list of NT quotations from the writings of the Church Fathers predating AD 400 and showed that they agreed with the Byzantine text against the Westcott-Hort text in about 60% of the cases.
Furthermore, with the discovery of the New Testament papyri, we have more direct evidence of the form of the text than we can obtain from the Church Fathers, and Harry Sturz has shown that there are over 150 distinctively Byzantine readings found in these papyri, and this certainly puts paid to Westcott and Hort’s second argument.
We should note that the acolytes of Westcott and Hort try desperately to override this by arguing that, while Byzantine readings may be early, the Byzantine text type is not. However, since, as we have shown, the focus needs to be on the readings and not on text types (which constitute an unsustainable concept), this argument is worthless.
This topic does, however, illustrate the curious double standards which textual critics sometimes employ to maintain their theory. Textual critics maintain that the putative Western text type is very early but there is far less evidence for it than for the Byzantine text type. Yet Fee tries to bolster the case for the Western text type by asserting that “variants peculiar to it are firmly established in texts found in North Africa, … Italy, … and southern France,” while Metzger appeals to the versions, saying
Old Latin and Old Syriac readings often agree with the Greek text of Codex Bezae.
So both Fee and Metzger claim that Byzantine readings, even though they are found in the earliest Greek manuscripts, the papyri, do not prove the early existence of the Byzantine text, but early Western readings that are not even found in Greek manuscripts but only in much later version manuscripts, do prove the early existence of the Western text type! Is there any conceivable justification for this double standard? The correct answer is: No.
Westcott and Hort’s final argument is that the Byzantine readings are obviously late since the Byzantine text type is “conspicuously a full text. It delights in pronouns, conjunctions, and expletives and supplied links of all kinds, as well as in more considerable additions … it presents the New Testament in a form smooth and attractive, but appreciably impoverished in sense and force.” This obviously cannot even be called an argument. It is based on nothing more than the implication that the original Bible must have been written in a rough and unattractive way, which is ridiculous. We also note that pronouns, conjunctions, and supplied links are exactly the sorts of words that can be easily lost through accidental omission, which, again, is the most common scribal error by far. So if the Alexandrian text has fewer of these, they were most likely lost by accidental omission. This, too, is therefore an indication that Westcott and Hort’s arguments actually support the primacy of the Byzantine text, rather than showing it to be secondary. And we need not waste time on the asinine statement that the Byzantine text is “appreciably impoverished in sense and force.” At best this is but a personal opinion (and a rather silly one); at worst, it is a cynical attempt at emotional manipulation.
And so we reach the end of our investigation of Westcott and Hort’s case against the Byzantine text, the case that convinced the scholarly world – including evangelical scholars – to replace the inerrant word of God with a text that is replete with error and disharmony. We have seen that every one of their arguments is an utter failure when it comes to demonstrating the primacy of this errant text by denigrating the Byzantine text. On the contrary, as we have seen, the arguments are consistent with all that we have previously seen in showing the primacy of the Byzantine text. While it is not difficult to see why liberal scholars would nevertheless accept the errant text promoted by Westcott and Hort, it is exceedingly difficult to see why evangelicals would allow themselves to be led by the nose to embrace, and, indeed, to champion an error-ridden and demonstrably inferior text that destroys the concept of inerrancy.
After insisting that the Byzantine text was secondary and corrupt, Westcott and Hort tried to account for its creation and dominance by asserting that it had been put together in the 4th-century AD, probably by a certain Lucian of Antioch, as an official recension of the NT that was “executed and imposed upon the churches by ecclesiastical authority.” Since the Byzantine text is not, in fact, a late recension at all, there is not much point to considering this suggestion. Nevertheless, we note Fee’s understatement that this suggestion is not based on “firm data.” John Wenham points out that
The supposed official recensions which are said to have given the [Byzantine] text its form … are unknown to history.”
And Sturz also affirms that there is no proof whatsoever that the Byzantine text was created as a Lucianic recension. As he pointedly says,
There are abundant and varied witnesses to the editing process which resulted in the Latin Vulgate. But in the case of the Byzantine text – silence.
This silence, he says, is “inexplicable; it is not what one would expect. It seems logical that there should be as great or greater reaction to the replacement of a people’s whole Greek New Testament (the original language) than there was to Jerome’s revision of the Old Latin (a translation.)”
The suggestion, then, that the Byzantine text was created by Lucian of Antioch in the 4th century seems to be yet another example of text-critical ideology ruling over actual evidence. But that is par for the course in text-critical scholarship – evangelical or otherwise.
 In addition, the NT was translated into other languages in ancient times and then transmitted in those languages, and many of these manuscripts have also survived: 10,000+ in Latin; 2,587 in Armenian; 4,101 in Old Slavonic; 2,000+ in Ethiopic; and others.
 By way of comparison, the second best attested ancient writing is Homer’s The Iliad, with 1,797 extant copies, the earliest coming from 400 years after the date of the original composition. Caesar’s Gallic Wars exists in 251 copies, the earliest from 950 years after the date of composition.
 NT manuscripts are divided into papyri (designed P with a superscript), uncials, minuscules, and lectionaries.
 From Aland, Kurt and Barbara Aland. The Text of the New Testament. Revised and Enlarged. Erroll F. Rhodes, trans. Grand Rapids, MI: William B. Eerdmans Publishing Co., p. 81. This information is somewhat out of date.
 Wilbur Pickering claims to have perfect archetype manuscripts for 22 of the 27 NT books (Pickering, Wilbur N. The Greek New Testament According to Family 35. Lexington, KY, 2014, p. iv).
 Unfortunately, many of the liberal paradigm assumptions of historical criticism (such as late dating of the Gospel books – i.e., “decades after the resurrection” – and Markan priority) are accepted uncritically by many or most evangelical scholars, and a great many of them compromise with Darwinism, too.
 And, of course, they should be. The evidence is decidedly against them.
 An example of this: The beginning of Jesus’ response to the rich young ruler is “Why do you call Me good? No one is good but One, that is, God” in the parallel accounts in Matthew 19:17, Mark 10:18, and Luke 18:19. It is the same in all three Gospel books. However, eight manuscripts of Matthew read, ““Why do you ask me about what is good? There is only one who is good.” Although the parallel reading, “Why do you call Me good? No one is good but One, that is, God” is found in at least 900 manuscripts of Matthew, Griesbach’s canon demands that the reading that creates a discrepancy with parallel material must be chosen, so that is what ends up in the NIV, ESV, NASB, HCSB, RSV, NRSV et al. Then Muslim apologists such as Shabir Ally use this reading to “prove” that Matthew was elevating the portrait of Jesus by altering His original words, which might be (mis)construed as a denial of His deity.
 Actually, their testimony in this matter is questionable. The scribe of Codex Vaticanus left an entire blank column after Mark 16:8 , and this is the only place in the manuscript where such a thing is done; this suggests he knew that there was more to the Gospel According to Mark, and he left space for it to be included, but for some reason it was never done. In the case of Codex Sinaiticus, the original pages containing the end of the Gospel According to Mark were removed and replaced at a later date by pages that seem to have been written in such a way as to hide the fact that the ending of the Gospel book had now been removed. In fact, 16:9-20 are found in about 1,800 manuscripts of the Gospel According to Mark, and are quoted by 2nd-century Church Fathers more than 150 years before the writing of Codex Sinaiticus and Codex Vaticanus.
 “During the latter half of the nineteenth century, however, a new hypothesis took center stage. Renewed textual examinations … reached the conclusion of Marcan priority … text-critical arguments … establish that Matthew has secondary readings and Mark has original readings.” (Williams, Matthew C. Two Gospels From One: A Comprehensive Text-Critical Analysis of the Synoptic Gospels. Grand Rapids: Kregel Publications, 2006, pp. 23, 215)
 Burgon, John W. The Last Twelve Verses of the Gospel According to S. Mark Vindicated Against Recent Critical Objectors and Established. Oxford and London: James Parker and Co., 1871.
 Hoskier, Herman C. Codex B and Its Allies: A Study and an Indictment. 2 vols. London: Bernard Quaritch, 1914.
 Pickering, Wilbur N. The Identity of the New Testament Text. Revised edition. Nashville: Thomas Nelson Publishers, 1980, p. 218 f.n. 68
 But see ibid.
 Metzger, Bruce M. The Text of the New Testament: Its Transmission, Corruption, and Restoration. Third, enlarged edition. New York and Oxford: Oxford University Press, 1992, p. 136
 In fact, among the 5,795 Greek manuscripts, the only genealogical relationships that have been established are between P75 and B, among the seven manuscripts that constitute “family 1” (f1), and among the thirteen manuscripts that make up “family 13” (f13). That is all.
 P66, in its extant length of 808 verses, has 400 hundred itacisms and another 482 “singular readings,” 40% of which are nonsense readings. That is one singular error for every 17 words – and that doesn’t include errors in it that may also be found in other manuscripts. P75, meanwhile, has 145 itacisms and 257 other singular readings, 25% of which are nonsensical. (Colwell, Earnest Cadman, op. cit., pp. 374-376. Itacisms are vowel substitutions. “Singular readings” are textual variants that are not found in any other manuscript. “Nonsense Readings include words unknown to grammar or lexicon, words that cannot be construed syntactically, or words that do not make sense in the context.”)
 The early papyri predated the supposed date of formation of these text types.
 Aland, Kurt. “The Significance of the Papyri for Progress in New Testament Research,” The Bible in Modern Scholarship. Ed. J.Philip Hyatt. New York: Abingdon Press, 1965, pp. 325-346. Recensions are revised editions of a text, in this case the original of each of the text types.
 Or 5,7776 if we do not consider the members of P75/B, f1, and f13 independent.
 Of course, this should already have been obvious from the bogus nature of the arguments against the Byzantine text type and from the fact that we have only been able to find a handful of genealogical relationships among the extant manuscripts.
 Westcott, B.F. and F.J.A. Hort. Introduction to the New Testament in the Original Greek, With Notes on Selected Readings. Peabody, MA: Hendrickson Publishes, 1988, p. 45 [Original Edition: Westcott, B.F and F.J.A. Hort. The New Testament in the Original Greek. 2 vols. London: Macmillan and Co., 1881]
 If it entered only two, the correct reading would be found in three, so the correct reading would be in the majority of manuscripts going forward.
 First-generation copies are copies made of the original document. Second-generation copies are copies made of first-generation copies, third-generation copies are copies made of second-generation copies, and so forth.
 If errors entered three of the five copies but they were all different errors, then each would infect only 20% of the manuscripts. The correct reading would still be found in a plurality (40%) of the manuscripts.
 Another factor that should be taken into account (but isn’t) is the quality of each individual manuscript. Demonstrably corrupt manuscripts such as א and B and P66 and P75 should never have been taken seriously for textual criticism.
 What M.J. Harper says about history students is equally true about seminarians: “Historians … aren’t asked to consider the evidence for the paradigm assumptions of their subject; they are taught the paradigms in the first five minutes – or better still are assumed to know them when they arrive.” (Harper, M.J. The History of Britain: The Shocking Truth About the English Language. Cambridge: Icon Books Ltd., 2006, p. 34. Bolding added.)
 John Wenham was a notable exception to this.
 e.g., reading “Isaiah the prophet” in Mark 1:2 in place of “the prophets”
 The English Majority Text Version (EMTV) and Wilbur Pickering’s WPNT are both based on the Byzantine text, but they cannot yet be considered “major” translations, as few people have even heard of them at this point. Let us hope that that will change.
 “Scholars today generally agree that one of the chief contributions made by Westcott and Hort was their clear demonstration that the Syrian (or Byzantine) text is later than the other types.” (Metzger, op. cit., p. 135).
 ibid., p.1 36
 i.e. a variant that is not the original reading
 Westcott and Hort, op. cit., pp. 60-61
 Metzger, Text, p. 135
 Westcott, B.F and F.J.A. Hort. The New Testament in the Original Greek. 2 vols. London: Macmillan and Co., 1881, p. 133; cited in ibid., p. 132
 Burgon, John William. The Traditional Text of the Holy Gospels. Arranged, completed, and edited by Edward Miller. London: George Bell & Sons and Cambridge: Deighton, Bell and Co., 1896, pp. 94-101. The acolytes of Westcott and Hort tried to avoid this evidence by claiming that Burgon had used inferior manuscripts of the Church Fathers, but since their only basis for saying those manuscripts were inferior was that the quotes agreed with the Byzantine text, this claim be safely ignored.
 Sturz, Harry A. The Byzantine Text-Type and New Testament Textual Criticism. Nashville: Thomas Nelson Publishers, 1984.
 Metzger, Text, p. 293; Holmes, Michael W. “The ‘Majority text debate’: new form of an old issue.” Themelios, New Series 8:2 (January 1983), p. 16; Fee, Gordon D. “The Majority Text and the Original Text of the New Testament.” The Bible Translator 31:1 (January 1980), pp. 115-116
 Holmes, ibid.; Fee, Gordon D. “Modern Textual Criticism and the Revival of the Textus Receptus.” JETS 21:1 (March 1978), p. 28
 Epp, Elton Jay. “The Twentieth Century Interlude in New Testament Textual Criticism,” JBL 93 (1974), p. 412
 Fee, Revival, p. 28. (Italics added.)
 Metzger, Text, p. 294 (Italics added.) Codex Bezae is the only Greek manuscript that is classed in the Western text type.
 Westcott, B.F and F.J.A. Hort. The New Testament in the Original Greek. 2 vols. London: Macmillan and Co., 1881, p. 133; cited in ibid., p. 132
 Pickering, Identity, pp. 37-38
 Fee, Gordon D. “The Textual Criticism of the New Testament.” In Harrison R.K., Waltke, B.K., Guthrie, D., and Fee, G.D. Biblical Criticism: Historical, Literary and Textual. Grand Rapids, MI: Zondervan Publishing House, 1978, p. 137
 Wenham, John W. Christ and the Bible. 3rd edition. Grand Rapids, MI: Baker Book House, 1994, p. 185
 Sturz, op. cit., pp. 123-126.
 ibid., pp. 123-124
 ibid, p. 126