How Many Witnesses Do We Really Have?

There is an interesting and regularly observable dichotomy found in the major tenets of modern evangelical textual criticism. On the one hand you have modern evangelical text-critics saying,

“…the copies of Homer, Plato, Aristotle, and others from the ancient world have produced adequate copies for us to know what they taught. And as shown below, we have more accurate copies of the original New Testament than they do of their original texts.”

Geisler and Roach, Defending Inerrancy, 80.

Geisler and Roach go on to write,

“Clearly the New Testament is the most well-attested book from all ancient history. If one denies the reliability of the New Testament based upon the number of manuscripts and the interval of time between its original composition and nearest copy, then they would have to thereby discredit the reliability of every work from ancient history.”

Geisler and Roach, Defending Inerrancy, 83.

Along these same lines Daniel Wallace writes,

“Although the vast majority of NT MSS are over a millennium removed from the autographs, there are significant numbers of documents in the first millennium. Naturally, the closer we get in time to the originals, the fewer the MSS. But the numbers are nevertheless impressive – especially when compared with other ancient literature.’

Daniel Wallace, Inerrancy and the Text, Sec. 2.

In sum, we have so many manuscripts. I mean, many times more than other works in antiquity. On the other hand you have textual scholars saying,

“It [the Byzantine Text-Type] is best represented today by Codex Alexandrinus (A 02, in the Gospels; not in Acts, the Epistles, or Revelation), the later uncial manuscripts, and the great mass of minuscule manuscripts. These minuscule manuscripts are cited together under the symbol Byz in the critical apparatus.”

Roger Omanson, A Textual Guide to the Greek New Testament, 23. [Italics Mine]

Omanson goes on to comment,

“About eighty percent of the minuscule manuscripts and nearly all lectionary manuscripts contain the Byzantine text-type.”

Roger Omanson, A Textual Guide to the Greek New Testament, 23.

Wasserman and Gurry follow suit in writing,

“Their [the Byzantine minuscules’] agreement is such that it is hard to deny that they should be grouped. In fact, the editors using the CBGM do group them together, subsuming them in the apparatus under the symbol Byz.”

Wasserman and Gurry, A New Approach to Textual Criticism: An Introduction to the Coherence-Based Genealogical Method, 9.

Given the above quotes we have these two things in play at the same time: 1.) We have more original manuscript copies than any other book from antiquity and 2.) The Byzantine manuscripts really count as one witness because “it is hard to deny that they should be grouped.” And what is the nature of this Byz? According to the experts is it a good source? Omanson calls Byz “corrupt” [24]. Wasserman and Gurry recognize that for nearly the whole existence of modern textual criticism Byz was “disparaged by a majority of New Testament textual critics” as being “the least valuable” [10]. So if we take the Byzantine text-form and subsume it under one symbol, as one witness, how many actual witnesses do we have to the NT according to the scholars?

To answer this question let’s begin by looking at a chart. This chart shows the number of manuscripts we have of other ancient writings.

Author	Work	Date Written	Earliest MSS	Time Gap	Number of MSS
Homer	Iliad	800 BC	c. 400 BC	400	1757
Herodotus	History	480-425 BC	10th C	1350	109
Sophocles	Plays	496-406 BC	3rd C BC	100-200	193
Plato	Tetralogies	400 BC	AD 895	1300	210
Caesar	Gallic Wars	100-44 BC	9th C	950	251
Livy	History of Rome	59 BC-AD 17	Early 5th C	400	150
Tacitus	Annals	AD 100	AD 850	750-950	33
Pliny, the Elder	Natural History	AD 49-79	5th C fragment: 1; Rem. 14-15th C	400	200
Thucydides	History	460-400 BC	3rd C BC	200	96
Demosthenes	Speeches	300 BC	Some fragments from 1 C BC	1100+	340
Greek NT		AD 50-100	AD 130	40	5795

We see that the total number of NT manuscripts is somewhere around 5,800 manuscripts and the closest competitor is Homer with 1,757 manuscripts. Indeed, this is quite a wide margin of witnesses. But before we draw our conclusion as Omanson, Wallace, Geisler, Roach and so many others have, let’s take a look what manuscripts make up that 5,800.

Currently, the total list of minuscules is ~3,000 which is approximately half of the total manuscript count. As we saw above, Omanson says that 80% of those minuscules are Byzantine and Byzantine manuscripts are seen as a single witness as attested to by Omanson, Wasserman, Gurry, and the ECM editors. 80% of 3,000 is 2,400. So according to the experts, 2,400 manuscripts = Byz or exactly one (1) witness. So,

5,800 – 2,399 = 3,401 witnesses.

But we are not done. The manuscript tradition is broken up into four separate kinds of witnesses: papyri, uncials/majuscules, minuscules, and lectionaries. How many of our remaining 3,401 witnesses do you think are lectionaries? We have ~2,400 lectionaries, and, again, as Omanson pointed out under his section dealing with the Byzantine text-type, “nearly all lectionary manuscripts contain the Byzantine text-type.” As such, all 2,400 are in reality Byz or exactly one (1) witness which has already been accounted for above. So,

3,401 – 2,400 = 1,001 witnesses.

What then is the tail of the tape? If the above observations are correct, Homer has 70% more witnesses to the Iliad and Odyssey than we do to the Greek NT. What is more, the vast majority of our “embarrassment of riches” seems to be one corrupt disparaged gem of least value which is called, Byz. We have more copies, sure, but we don’t have more witnesses. What is more, any first year Greek student knows that we don’t count manuscripts to determine a reading, we weigh them. Yet time and time again modern evangelical scholars talk about number, number, number, but those in the know care little for number in making their text-critical decisions. That said, let’s take it a step further.

Note in the quote third from the top that Wallace recognizes that “the vast majority of NT MSS are over a millennium removed from the autographs.” Then take a look at the chart. All but three of the works from antiquity have witnesses earlier than 1,000 years from their original.

The point being that if you make the Byzantine text-form a single witness, Byz, then we do not have an embracement of riches. And if it is an embarrassment of riches the vast majority of those riches are disparaged, said to be corrupt, and of little value. The fact is, that without Byz, we have about half as many witnesses as Homer’s works do. As to the date of the manuscripts, our best manuscripts are said to be Aleph and B which are fourth century manuscripts and Homer’s works have attestation in the fifth century. So even the age of the witnesses are neck-and-neck in this regard.

A few posts ago we had to parse out what modern evangelical text-critics meant by “earlier” or “later” and the equivocation found therein. Now come to find out on the point of copies vs. witnesses, we have a bunch of copies compared to other ancient texts but far less witnesses especially when compared to Homer’s work. We could take this argument further and point out that many of the papyri are mere scraps the size of credit cards having only a handful of words on them, but at this point I think the point is proven. There is an unacknowledged in the modern evangelical text-critical world and depending on whether scholars are talking with scholars or scholars are talking with God’s sheep will determine which side of the dichotomy the scholar chooses.