Arbitrariness: The Unproved Decisions Behind the Text

This is Part 3 of “By What Standard Is This the Text? A Presuppositional Examination of Modern Textual Criticism.”

What Arbitrariness Means

Calling an argument arbitrary does not mean that its advocate rolled dice or made a choice without thought. A decision may be learned, intricate, and widely respected while still lacking a sufficient ground for the authority claimed on its behalf.

An arbitrary conclusion is one for which no binding reason has been supplied. The critic may possess reasons that explain why he prefers a reading. The question is whether those reasons establish why the church is bound to receive that reading as the Word of God.

Greg Bahnsen’s checklist divides arbitrariness into four forms:

1. Mere opinion
2. Relativism
3. Unargued bias
4. Ignorant conjecture

These defects are related, but they are not identical. A critic may state a personal judgment. A school may permit several incompatible judgments. A method may conceal a controlling preference beneath neutral language. A proposed textual history may fill gaps in the evidence with a story that cannot be checked.

Modern textual criticism is not arbitrary because it has no evidence. It is arbitrary where no revealed and invariant rule governs the passage from evidence to an authoritative text.

Mere Opinion: The Judgment That Ends the Appeal

Reasoned eclecticism gathers several kinds of evidence. External evidence asks about witnesses, age, distribution, and relationships. Internal evidence asks what an author was likely to write and what a scribe was likely to change. The canons are then brought together in a local decision.

But who decides their rank?

One reading may have earlier support while another has wider support. One may fit the author’s style while another more readily explains the rise of its rivals. A harder reading may be favored because scribes tended to smooth difficulties, but it may be rejected because a scribe plainly created nonsense. A shorter reading may look original because scribes expanded texts, yet accidental omission is among the commonest copying errors.

No single canon resolves the conflict. The editor forms a judgment about which considerations matter most in that place.

That judgment may be informed. It may even be correct. Yet the truth of a conclusion and the warrant for a conclusion are different matters. A man can guess the right answer without possessing a rule that proves it. If the final explanation is that the preferred reading seems to offer the best account of the evidence, the method has arrived at expert opinion.

The problem becomes sharper when the conclusion is placed in the main text of a Greek New Testament and then treated as the wording upon which preaching, translation, and doctrine should rest. An editor’s opinion has moved from a scholarly proposal to a practical judgment concerning the boundaries of divine speech.

White’s answer at the end of the Riddle debate illustrates the point. He named sets of standards but denied one objective, universal, and absolute standard. His final court is thus not a rule above the critics. It is the critic applying several rules.

Relativism: More Than One Permissible Text

Relativism does not require a critic to say that truth itself is relative. It appears within a method when incompatible conclusions are equally permitted because no binding higher norm determines between them.

Modern editors openly acknowledge doubtful places. Critical editions mark uncertainty, retain split lines, and revise earlier decisions. The Editio Critica Maior of Revelation, for example, introduced scores of changes to the initial text and many places where the editors did not print one undivided conclusion. Such honesty is better than hiding uncertainty. It also reveals the state of the method.

If one edition prints reading A, another prints reading B, and both results lie within accepted critical practice, which wording binds the conscience? Saying that the evidence is difficult describes the problem; it does not answer it.

Jan Krans expressly presents the text-critical task as unfinished. Methods may be refined, fresh evidence may appear, and readers must live with some uncertainty. Peter Gurry’s discussion of the initial text likewise distinguishes the scholarly starting point of the extant tradition from the authorial text in ways that may not always be demonstrably identical. The CBGM can expose relationships and coherence, but it does not deliver an inspired verdict. Human editors must still decide the priority of readings.

The result is a series of texts rather than one settled text: the current best hypothesis, a prior best hypothesis, and an anticipated future improvement. Each may be responsible scholarship. They cannot all be the same words.

When the question is merely which reconstruction is most useful for further research, provisional answers are expected. When the question is which words are God-breathed and bind the whole church, a provisional plurality is not enough.

Unargued Bias: The Rules Behind the Rules

An unargued bias is not always a conscious prejudice. It is often a controlling assumption that has become so familiar within a field that it is no longer defended.

Consider several common assumptions:

• More recently developed methods are presumed superior because they are newer and use more data.
• A reading known from earlier surviving witnesses is presumed to stand nearer the source.
• Continuous ecclesiastical use is treated as evidence about reception but not as a theological fact governing authenticity.
• The promises and confessional doctrine of preservation are separated from copying history before the manuscripts are examined.
• A local, variant-by-variant method is assumed proper even if it never demonstrates that its collected decisions form one historically possessed text.
• The authorial or initial text is made the goal, while the church’s received and publicly used text is reduced to one witness among others.

Each assumption may be argued. None is neutral.

Elijah Hixson’s work is useful at this point because he has warned popular defenders against the slogan that the earliest manuscript must be best. A paleographical date is often a range, not a timestamp, and the age of the material is not the age of the reading. Peter Gurry has likewise shown that the Byzantine text agrees with the ECM’s reconstructed text across the great majority of the variation units in the Catholic Epistles. The Byzantine tradition once treated largely as a late obstacle is now acknowledged as an important witness to early readings.

That reassessment is healthy as scholarship. It also demonstrates that the evidence did not force the earlier estimate. A framework determined what the evidence was allowed to mean, and a later framework changed the estimate.

The deepest bias is the exclusion of Scripture from the rules governing textual criticism. Krans argues directly that biblical statements about the preservation and reliability of God’s Word do not concern copying processes. But that conclusion is theological and exegetical. It is not a result discovered by collating manuscripts. The divine promises have been assigned a restricted meaning before textual history is reconstructed.

Once that restriction is adopted, the conclusion follows: no biblical promise can control the critic’s account of transmission. The method appears neutral only because its governing theological decision has been placed outside the method.

Ignorant Conjecture: Filling the Missing History

Conjecture is not always wrong. Every historical inquiry draws inferences from incomplete evidence. The fault lies in claiming more than the evidence can bear, especially when the missing facts are essential to the conclusion.

The textual critic does not possess the autographs for comparison. Most copies ever made no longer survive. The exemplars behind surviving manuscripts usually do not survive. The identity, theology, ability, intention, and circumstances of most scribes are unknown. The routes by which readings moved between manuscripts are often reconstructed rather than observed.

Nevertheless, critical explanations commonly describe what must have happened. A reading was expanded to explain a difficulty. A marginal note entered the text. A scribe harmonized the passage. An orthodox copyist strengthened a statement. A shorter form was accidentally or deliberately enlarged. A word was lost by a known mechanical error.

Such proposals may be plausible. Several rival proposals may be plausible as well. Without the source copy, the event cannot be replayed. The critic judges the lost act from its surviving effects.

The CBGM handles some of this uncertainty more carefully than older family-tree models. It maps coherence rather than pretending that every physical manuscript descends through a simple line. But its textual-flow diagrams still depend upon prior decisions about which readings are earlier. The tool arranges relationships after human judgments of priority have been supplied. It cannot independently prove the judgments that feed it.

The same problem governs claims about the Textus Receptus. It is proper to examine Erasmus, Stephanus, Beza, the Elzevirs, their manuscripts, annotations, editions, and printing choices. It is not proper to turn gaps in that record into confident tales of carelessness or ignorance and then use the tales to dismiss the text as a whole.

Unknown history remains unknown. A conjecture does not become knowledge because it accords with the current model.

The Verification Problem

A simple thought experiment exposes the limit. Give a known document to many copyists. Allow their copies to be copied again through several generations. Destroy an unknown portion at each stage. Introduce a mixture of accidental mistakes, deliberate corrections, harmonizations, and faithful copying. Then give only the surviving descendants to a group of critics.

The critics may produce a highly plausible reconstruction. How is its complete accuracy verified? It is compared with the known original.

In the actual New Testament tradition, the autographs are not available for that comparison. The method can be tested for consistency, explanatory power, and performance upon artificial traditions whose originals are retained. It cannot be directly validated against the lost biblical autographs.

This does not prove that every critical decision is false. It proves that the critical process cannot transform probability into divine authority by its own power. Even when it reaches the authentic reading, it lacks the final empirical comparison that would prove it has done so.

The alternative is not blind irrationality. It is a different starting point: the self-attesting Word and God’s revealed work of preservation. The church receives the authentic text as a providential whole and employs manuscripts, versions, citations, editions, and historical study as ministerial witnesses within that prior certainty.

Evidence Beneath Revelation

The confessional position also makes judgments. It must compare editions, identify printing mistakes, study the history of readings, and answer difficult cases. Its claim is not that human reasoning disappears.

Its claim is that human reasoning does not create its own court of final appeal.

The received Greek and Hebrew texts are not authoritative because a sixteenth-century editor possessed infallible skill. They are authoritative because the church confesses that God preserved the immediately inspired Scriptures, kept them pure in all ages, and made them authentical for final appeal. Historical reception is considered within that doctrine, not as an autonomous substitute for it.

All ultimate reasoning contains a necessary return to its first authority. The critic who rejects circularity still uses reason to defend reason and historical judgment to defend historical judgment. The question is not whether one begins with an ultimate commitment. The question is which commitment makes evidence, history, truth, duty, and an identifiable Word of God intelligible.

Mere opinion, relativism, unargued bias, and ignorant conjecture do not prove that every critical reading is mistaken. They prove that modern textual criticism cannot supply the sort of universal warrant required to bind the church to its changing result.

The next test is inconsistency. Do the canons remain stable when they are applied, or do they change places as the desired reading changes?

Notes and Further Reading

• Chris Thomas, “Argument Analysis Checklist”.
• Chris Thomas, “Why New Testament Textual Criticism Can Never Arrive at Truth”, presenting the fuller transmission thought experiment.
• Peter J. Gurry, “The Overall Quality of Byzantine Manuscripts”.
• Peter J. Gurry, “Conjecturing the Initial Text Where the Original Text Is Extant”.
• Peter J. Gurry, “Video Look at the New ECM of Revelation”.
• Jan Krans, “Why the Textus Receptus Cannot Be Accepted”.
• Elijah Hixson and Peter J. Gurry, eds., Myths and Mistakes in New Testament Textual Criticism.