Aristotle wisely cautioned us to avoid extremes in our pursuit of “the golden mean,” the virtue that often resided between two extremes. I attempt to do so with textual criticism, by this paradoxical mini-creed: Text criticism is not unimportant; text criticism is not all-important. After Robert Lewis Dabney completed an extremely detailed and erudite discussion of “The Doctrinal Various Readings in the New Testament,”[1] he observed that not one significant doctrine would be affected by any text-critical conclusion. I agree with Dabney, both in my willingness to do text criticism whenever the interpretation of a given text demands it, and in agreeing that comparatively little is at stake. The gospel, for instance, is not at stake text-critically, nor is the “word of Christ,” or “the Word of God” at stake. The apostolic gospel existed and was proclaimed orally before there were any apostolic writings, much less twenty-seven of them. That same gospel exists today, and is proclaimed orally today by missionaries in cultures that have no written language.

At the time Paul wrote the letter to the Colossians, the New Testament was still in the process of being formed; indeed, Paul was still writing letters himself. The “Word” that was available to them was the apostolic gospel, perhaps as summarized in 1 Corinthians 15:1–11. A brief argument ad absurdum may help see the point: How would we obey Paul’s command today in countries where the gospel has been preached, but the Bible has not yet been translated? The existence of the proclaimed gospel is not dependent upon individual believers owning or reading a Bible translated into their own language (assuming their culture has a written language).

Mr. Stahl writes: “Those who received the written New Testament as books or letters in its original language had confidence what they were hearing was in fact the Word of God.” Which was it? Did they hear the gospel/Word of God, or did they read it? Prior to the printing press, for fifteen centuries, the Christian church expanded nearly globally, without individuals owning Bibles at all. Indeed, most Christian churches did not have an entire Bible, they merely had copies of the lectionary readings for the year. It is anachronistic to assume that Paul wrote to people who owned Bibles; they did not. The only reason the Ethiopian in the chariot had access to a manuscript of Isaiah was because of his professional duties: “And there was an Ethiopian, a eunuch, a court official of Candace, queen of the Ethiopians, who was in charge of all her treasure” (Acts 8:27, emphasis added). A hand-copied manuscript, whether on animal skin or papyrus, was indeed, in the first century, a “treasure,” and only one who had access to such treasure would have had access to a manuscript.

The claim that “the churches also carefully copied the written Word” is a form of special pleading. We may not assume such a historical matter. Some churches may have had scribes; many, if not most, churches likely did not (especially those that were not in larger metropolitan areas). Note how expensive such manuscripts were in the ancient world: “And a number of those who had practiced magic arts brought their books together and burned them in the sight of all. And they counted the value of them and found it came to fifty thousand pieces of silver” (Acts 19:19). We do not know how many people were practicing black arts, but it may have been as few as the seven sons of Skeva, or possibly more, perhaps fifty. But fifty thousand pieces of silver is over a thousand times what Judas received for betraying Christ; considering that he took his own life in grief, even forty pieces of silver must have been of fairly substantial value. If books were readily available in the first century, Paul would not likely have needed Timothy to bring his from Troas: “When you come, bring the cloak that I left with Carpus at Troas, also the books, and above all the parchments” (2 Tim. 4:13, and the “above all” here, μάλιστα (malista), should probably be “namely,” indicating which books were to be brought, distinguishing parchment—τὰς μεμβράνας (tas membranas)—from papyrus).

It is only plausible, not at all necessary, to interpret Westminster’s “kept pure in all ages” to mean “kept equally pure in all places in all ages,” which the comment appears to assume. Even Stephanus—a Renaissance scholar with access to libraries and scriptoria—did not have an entire Greek text to work with. His original was missing an entire page. It simply is not historically true that Greek and/or Hebrew manuscripts were available everywhere at all times. They probably intended only to affirm that, despite the wide global spread of the church, and despite occasional persecution, copies of original manuscripts survived (almost miraculously); and, insofar as they are available in any given generation, it is they (copies of the autographa) that are authoritative, not the Latin Vulgate. “Authentical” is merely an archaic form of “authentic,” and it was sometimes used to distinguish an original document from a copy thereof. The Assembly knew that they did not have the “authentical” books in that sense; they did not have the autographs. “Authentical” in their context undoubtedly meant that it was the original/authentic sacred writings themselves, not a Latin translation thereof, that was to be considered authentic in the sense that “all controversies in religion” were to be resolved by “appeal unto them” (as opposed to the Vulgate). The Assembly knew about the kinds of imperfections characteristic in ancient manuscripts; they knew that one of them was missing an entire page. Their point was to settle “controversies in religion” by “appeal unto” copies of the original languages, not copies of translations of the original languages.

Neither I nor, I trust, anyone else either knows who Mr. Stahl’s “many Reformed theologians” are, or what difference it makes. Commitment to an eclectic text does not require commitment to any particular textual tradition. Westcott and Hort were (in)famous for their preference for Alexandrian readings, but successive generations have mostly become truly eclectic, preferring internal evidence to external evidence in most cases (and recall that Stephanus himself, the principal originator of the Textus Receptus, employed an eclectic text, even using a Latin text where his Greek manuscript was defective). Anyone who compares available manuscripts, and then selects the reading that is best explained by what we know by the process of copying, embraces an eclectic text, as Stephanus did.

Westcott and Hort are not representatives of the eclectic text tradition. They had strong, presumptive, commitment to Alexandrian manuscripts. Most today regard their viewpoint as naïve, and make text-critical decisions based on internal evidence, not external evidence. Much of Mr. Stahl’s reasoning, therefore, may be a tilting at non-existent windmills.

I also regard all Mr. Stahl’s references to “experts” in the previous article as misleading, whether intentionally or not. The late Bruce Metzger, himself an “expert,” was often in the minority of the committee that produced the UBS text, and the other “experts” permitted him to write the commentary on their work. His commentary demonstrates that there was a range of opinion among those “experts,” as is true in almost all cases where expertise is germane. The frequent references to such “experts” I regard not only as ad hominem but also as a somewhat crass appeal to American populism. Whether an expert or a bumpkin embraces a view is irrelevant to the question of the evidence and reasoning behind the view. Bumpkins are not always wrong, and experts are not always right (they were frequently wrong about Covid), but the frequent reference to unnamed “experts” mars the article with needless smoke through which the reader must cut. If an error is erroneous, its error should be refuted, without regard for its human origin.

I could wish that the matter were as simple as choosing between divine authority and human authority. Who does not prefer divine authority to human authority? Who would not prefer the divine Expert to human experts? But we do not have direct access to God himself (I recall something about a banishment from a garden . . .). We have access to several thousand hand-copied manuscripts of portions of Holy Scripture, every one of which has some demonstrable errors in it, and so we have basically three approaches to sorting through the matter: choose the right family of manuscripts (some Textus Receptus/Majority Text and Westcott and Hort, though they choose different families); choose the largest number of existent manuscripts (Majority Text); choose the reading that accounts for the other readings (Eclectic Text). God’s “singular care and providence” has preserved an enormous number of manuscripts, not one of which is free from obvious human error; respect for that providential care moves some of us to be willing to entertain the full range of what is providentially available.

Mr. Stah’s statistical discussion of percentages was curious, and not pertinent to the question. Do we count all of the letters in the original manuscripts, and calculate how many have variants? An individual with a speech impediment may have an erroneous articulation (an error?) in every word he speaks, yet speak with entire truth. When Dabney wrote about “The Doctrinal Various Readings” in the Greek text, he addressed only variants that might have a substantive consequence for what the Scriptures “principally teach,” namely, “what man is to believe concerning God, and what duty God requires of man” (WSC 3, and also referred to as “faith and life” at WCF 1:2). I don’t recall a single arithmetic statement in his discussion; and he concluded that none of the various readings affected Christian duty or belief. Suppose, for instance, that every sentence in the Hebrew or Greek Bible had a variant in the textual tradition somewhere; would this mean that the Bible is one hundred percent unreliable? No; it would merely mean that human copyists are unreliable. Attempting to quantify the matter would be an enormously difficult task (the equivalent of counting every letter in the Bible, then comparing each letter on each occasion to every known variant, a fool’s errand if ever there were one). It is also an entirely unnecessary task; if our goal is to find in the Bible what we are to believe and what we are to do, we already find more than we can handle, and the Westminster Assembly gave sage advice for reading said Bible: “The infallible rule of interpretation of Scripture is the Scripture itself: and therefore, when there is a question about the true and full sense of any Scripture (which is not manifold, but one), it must be searched and known by other places that speak more clearly” (WCF 1:9). Westminster candidly indicated that some places in Scripture spoke “more clearly” than others and had earlier noted,

All things in Scripture are not alike plain in themselves, nor alike clear unto all: yet those things which are necessary to be known, believed, and observed for salvation, are so clearly propounded, and opened in some place of Scripture or other, that not only the learned, but the unlearned, in a due use of the ordinary means, may attain unto a sufficient understanding of them (WCF 1:7).

Westminster seemed quite content to discover what is “necessary to be known, believed, and observed for salvation,” and that if the entire Bible were consulted (in some place of Scripture or other), people could attain to a “sufficient” understanding of the matter. They said nothing about the “precise meaning” of individual texts or the “specific meaning” of individual texts; they were concerned with “faith and life,” with Scripture’s basic doctrine of salvation.

Neil Postman[2] and Jacques Ellul[3] were very wary about modernism’s movement towards numerical calculation in every intellectual endeavor; even without a thermometer, we can ordinarily tell whether our forehead, or a child’s forehead, suggests a fever; and the treatment for a fever is the same for 100.5 or 101.2. For Ellul and Postman, many/most of life’s important realities could not be calculated numerically, but could be spoken of helpfully and meaningfully. Similarly, the Westminster Assembly was concerned with “faith and life,” with “what man is to believe concerning God and what duty God requires of man,” with what is necessary “for salvation.” None of their concerns were about ten percent of this or twelve percent of that.

It should be noted that Mr. Stahl’s reasoning suggests that an inerrant manuscript text is necessary to one’s faith, and necessary to one’s faith in the gospel and God’s Word. This suggestion is neither psychologically nor biblically true. Psychologically speaking, “truth” and “accuracy” are not identical concepts. A wartime soldier might aim at an enemy’s heart with his rifle and hit the lung instead, with the same result, that the enemy is no longer a combatant. The soldier “truly” shot the enemy, though inaccurately. Indeed, one purpose for aiming at the heart is that it is nestled between the lungs, and the “heart-lung area” is referred to by some specialists in these matters as the critical target area. Similarly, a public speaker may find himself reasoning with a series of negatives: “This is not to say that Paul was not concerned that the Corinthians may not give as generously as the Macedonians, resulting in saints in Jerusalem not having enough food.” In such a situation, at some point the speaker loses momentarily his Aristotelian logic and is not sure whether he has the right number of “nots” in his sentence, fearing that he may be off by one, inverting his intention. His hearers, however, understand his thought; they also may have lost the train of “nots” themselves, but they understand the truth of what the speaker intended.

After the printing press, the more-fluid nature of oral discourse or hand-copied manuscript was replaced with fixed type. The expectations of fixed type differed from those of oral discourse or manuscript, and, yes, when the “Adulterous Bible” was printed—regrettably omitting the “not” from “Thou shalt not commit adultery,” the Bibles were withdrawn from the public and reprinted at no expense to those who had purchased the original, mistaken version. The Holy Scriptures, both Old and New, were generated in that segment of human history in which the sacred writings were neither oral nor printed. For roughly three and a half millennia, oral discourse could be reduced to writing (to the chagrin of Socrates), but not to fixed print, where each individual copy is precisely like the others. It does not take a Media Ecologist to realize that, in this substantial era of human history, human sensibilities shaped themselves to the media they knew (orality and manuscript), not to media they neither had nor anticipated (fixed type). In the brief period when the “Adulterous Bible” was in print, some minister or lay reader must have read aloud the misprinted Decalogue during the liturgy, provoking only laughter, not widespread sexual infidelity.

It is possible that there are individuals today, over a half-millennium after the printing press, whose sensibilities differ from those of humans who preceded them; and perhaps Mr. Stahl is one of them. But his sensibilities are neither universal nor, in all likelihood, even humanly possible, prior to the printing press. And, as to the inference that our faith depends on an inerrantly-reconstructed printed product made from inerrantly-copied manuscripts, I might suggest that “faith comes from hearing, and hearing through the word of Christ” (Rom. 10:17, see also Gal. 3:2, 5), which, when Paul wrote these words, would have been a Word of God fallibly proclaimed orally or fallibly read orally from fallible manuscripts. Fallible apostles preached to fallible hearers, and the gospel nonetheless flourished.

An assumption throughout Mr. Stahl’s article is that we cannot function as faithful Christians unless we place our faith in an inerrant translation culled from errant manuscripts. This is not so; God requires of no generation that it do more than can be done in its own generation. Both Noah and David are described as having been faithful or obedient in his own generation (Gen. 6:9; Acts 13:36), suggesting that God only expects of us what can be expected in our specific moment in human history. But the characteristic trait God expects of us is faithfulness; his servants are routinely commended for being “faithful” (Matt. 24:45; 25:21; Luke 16:10; 1 Cor. 4:2, 17; Eph. 6:21; Col. 1:7; 4:7, 9; 1 Tim. 1:12; 2 Tim. 2:2; Heb. 2:17; 3:2, 6; 1 Pet. 5:12), not “accurate” or “precise.” Those who lived in those 1,500 years when one’s only access to Scripture was to hear audibly the lectionary readings for each Sunday of the church year, could have been very “faithful” to what they heard each week. Those who, like Stephanus, freely translated the Latin of Revelation into Greek to the best of his ability were “faithful.” We who fallibly read fallible translations based on fallible Greek manuscripts can still be by God’s grace, “faithful,” which is all that God expects of us. I am reminded of President Washington’s physicians, who treated him by draining blood from his veins, a practice common in the late eighteenth century. Their “treatment” hastened his demise. Their knowledge of medicine was imperfect, and a physician who did such today would lose, at a minimum, his license to practice. But the President’s physicians did the best they could with the best knowledge available in their generation and treated him as they would have treated an uncle, father, or son. They were medically wrong but ethically right. Approximate knowledge is the only knowledge humans have, and we are judged for being faithful to what we know and for how diligently we went about pursuing it. We will hardly be found unfaithful by all-knowing God if our translation reads “color,” rather than “colour,” and the overwhelming majority of variants in the manuscripts of the Bible are merely such regional variations.


[1] Robert L. Dabney, “The Doctrinal Various Readings in the New Testament,” vol. 1, Discussions: Evangelical and Theological (1891; repr., Edinburgh, UK: Banner of Truth, 1967), 350–90.

[2] Neil Postman, Technopoly: The Surrender of Culture to Technology  (New York: Vintage Books, 1993).

[3] Jacques Ellul, The Technological Society, trans. John Wilkinson (New York: Knopf, 1964). His frequently-repeated thesis was that “the essence of technique is to compel the qualitative to become quantitative.” Cf. also his The Technological Bluff, trans. Geoffrey W. Bromiley (Grand Rapids, Mich.: Eerdmans, 1986).

T. David Gordon is a minister in the Presbyterian Church in America and is a retired professor of religion and Greek at Grove City College in Grove City, Pennsylvania. Ordained Servant Online, December, 2023.

Publication Information

Contact the Editor: Gregory Edward Reynolds

Editorial address: Dr. Gregory Edward Reynolds,
827 Chestnut St.
Manchester, NH 03104-2522
Telephone: 603-668-3069

Electronic mail: reynolds.1@opc.org

Submissions, Style Guide, and Citations


Editorial Policies

Copyright information

Ordained Servant: December 2023

Remembering G. I. Williamson

Also in this issue

Elf on the Shelf or Christ on the Cross?

G. I. Williamson: Encounters with the Life of a Faithful Servant of God

G. I. Williamson’s Farewell Sermon

The Case for the Majority Greek New Testament Text

Theological Daylighting: Retrieving J. Gresham Machen’s Christianity and Liberalism

The Voice of the Good Shepherd: Trumpeter of God: Fulfill Your Office,[1] Chapter 9

Letters to a Younger Ruling Elder, No. 10: Be a Presbyter

Neo-Calvinism: A Theological Introduction by Cory C. Brock and N. Gray Sutanto

An Ode of the Birth of Our Saviour

Download PDFDownload ePubArchive


+1 215 830 0900

Contact Form

Find a Church