Evidence as Servant, Not Sovereign: The Proper Work of Textual Inquiry

Confessional Bibliology is sometimes presented with a false choice.

Either textual questions must be decided by an allegedly neutral examination of manuscripts, versions, and historical data, or they must be decided by a theological commitment that makes examination unnecessary. On this account, the critic follows evidence while the confessional Christian follows dogma. One investigates; the other assumes.

The choice is false because evidence and doctrine perform different tasks.

The doctrine of Scripture identifies the object being examined: the self-attesting, inspired, infallible, providentially preserved, and ecclesiastically received Word of God. Textual evidence bears historical witness to the transmission of that object. Doctrine does not tell us the date of a manuscript, the wording of a citation, or the location of an omission without examination. A manuscript does not tell us what inspiration is, what God promised to preserve, whether ecclesiastical reception matters, or why any critical principle should possess authority.

Confessional textual inquiry therefore refuses both the rejection of evidence and the sovereignty of evidence.

Manuscripts must be collated. Versions must be evaluated. Citations must be read in context. Grammar, syntax, authorial usage, scribal habits, liturgical use, printed editions, and the history of reception must be studied. Facts inconvenient to a preferred conclusion must not be concealed.

Yet none of these witnesses occupies a throne. They serve beneath the God who speaks in Scripture. They assist the ministerial recognition of His text; they do not constitute the text by an authority independent of revelation and providence.

Evidence is a necessary servant. It is a disastrous sovereign.

The False Opposition Between Faith and Evidence

Scripture never praises credulity, careless judgment, or indifference to facts.

The law requires testimony to be established rather than invented. Proverbs condemns answering before hearing and says that the first account appears right until it is examined. Christ commands right judgment. The Bereans are commended for searching the Scriptures to test apostolic preaching. Paul commands believers to prove all things and hold fast what is good. John tells the churches to test the spirits.¹

Christian faith is not belief without an object or warrant. God acted publicly in history. Luke investigated testimony and wrote an orderly account so that Theophilus might know the certainty of what he had learned. Paul named witnesses of the resurrection, many still living when he wrote. Peter distinguished apostolic eyewitness from cunningly devised fables. John appealed to what the apostles heard, saw, contemplated, and handled.²

Evidence matters because the truthful God acts, speaks, and governs history.

But Scripture also denies that fallen man may stand above revelation as an independent judge. The fear of the Lord is the beginning of knowledge. All the treasures of wisdom and knowledge are hidden in Christ. Every thought must be brought captive to His obedience. The natural man does not receive the things of the Spirit, and the carnal mind is hostile to God.³

Faith and evidence are therefore not two competing sources of information. Faith is the right covenantal posture toward the God whose world supplies the evidence and whose Word interprets its ultimate meaning. Unbelief does not encounter different ink, parchment, grammatical forms, or historical events. It interprets the same realities from a principle of independence.

The question is not whether evidence will be used. The question is whether evidence will be used faithfully, within the truth about God, man, Scripture, providence, and the church.

The Confessional Order

The Westminster Confession gives evidence an honorable but subordinate place.

The testimony of the church may move and induce believers to a high and reverent esteem of Scripture. Scripture displays abundant excellencies: the heavenliness of its matter, the efficacy of its doctrine, the majesty of its style, the consent of all its parts, its God-centered scope, its revelation of salvation, and the perfection of the whole. These are genuine evidences of divine authorship.

Nevertheless, the confession says that full persuasion and assurance of Scripture’s infallible truth and divine authority come from the inward work of the Holy Ghost bearing witness by and with the Word in the heart.⁴

The Spirit does not replace evidence with an unexplained feeling. He bears witness through the objective Word, opening the mind to recognize its divine light and authority. Nor does evidence replace the Spirit by constructing divine authority from historical probabilities. Scripture is received because God speaks; its evidences display the qualities and works of the One speaking.

The same order governs textual inquiry. Westminster 1.8 directs the church to the Hebrew and Greek Scriptures as immediately inspired, providentially kept pure, and therefore authentic. The appeal to the original languages presupposes real linguistic and textual work. Controversies are not settled by invoking the words kept pure while refusing to examine what the Hebrew or Greek says.

Yet the inquiry occurs beneath a prior confession. The church does not investigate whether God might have lost His Word. It investigates particular witnesses and readings within the certainty that He has kept it.

L. Gaussen describes a similar relation between faith and historical science in the recognition of the canon. Historical inquiry supplies documents, catalogues, citations, and facts of reception. It answers objections and confirms the public record. But faith rests finally upon the divine voice through the Holy Ghost, while the history is understood as the theater of God’s faithfulness toward His church.⁵

Textual evidence belongs to that same providential theater.

Evidence Never Interprets Itself

An artifact is not an argument.

A manuscript may display a reading. That is a textual fact about the manuscript. The manuscript’s date, provenance, relationship to other witnesses, and accuracy in a particular book require additional investigation. The conclusion that its reading is canonical requires still further reasoning.

These statements answer different questions:

• Observation: Manuscript A reads one word while manuscript B reads another.

• Historical judgment: The difference probably arose through omission, harmonization, correction, or another transmissional event.

• Textual inference: One reading better accounts for the existence of the others.

• Theological judgment: The reading should be recognized as part of the providentially preserved canonical text.

The first statement does not contain the last three inside it. Each step employs assumptions, classifications, and inferences.

Dating provides an obvious example. If one manuscript is earlier than another, the date is relevant. An earlier witness stands closer in time to the autograph and may preserve an earlier form. But chronological proximity does not guarantee genealogical purity. A late manuscript may accurately copy an ancient exemplar; an early manuscript may reproduce an earlier corruption. The maxim “prefer the oldest manuscript” therefore imports an inference not supplied by the date itself.

The same is true of numerical support. A reading found in many manuscripts may reflect broad and durable transmission, or many copies may descend from a comparatively narrow ancestry. A reading found in few manuscripts may be ancient, or it may be an isolated error. Counting supplies a fact. What the count proves depends upon a transmissional account.

There are no uninterpreted textual facts functioning as conclusions. There are facts interpreted well or badly, with sound or unsound premises, under faithful or unfaithful governing commitments.

What Textual Evidence Can and Cannot Establish

Textual evidence can accomplish a great deal.

It can establish what a particular witness reads. It can reveal copying mistakes, printing errors, marginal additions, transpositions, corrections, harmonizations, spelling changes, and editorial alterations. It can show that a quotation has been misattributed, that a reading existed by a certain period, or that a version cannot support the precise conclusion claimed for it. It can expose a false historical assertion made by a defender of the Received Text just as readily as one made by a defender of a modern critical edition.

David Norton’s history of the Authorized Version illustrates the value of exact historical work. By comparing annotated source materials, early printings, later editions, and the original languages, he distinguishes the translators’ decisions from printer’s errors and subsequent editorial changes. The result does not settle the doctrine of preservation, but it corrects careless claims about the English Bible’s publication history.⁶

Evidence may therefore correct us. A confessional commitment supplies no immunity from a misread page, inaccurate collation, false date, defective translation, or invalid inference. Truth does not become dangerous because an opponent discovers it.

But textual evidence cannot, by itself, establish its own rules of interpretation. It cannot determine whether God inspired words or only concepts. It cannot derive a promise of preservation from the shape of the surviving manuscript record. It cannot decide whether the church’s public reception is theologically relevant. It cannot explain why certainty must be restricted to probability or why an editorial committee’s preferred goal—whether the earliest attainable text, initial text, authorial text, or received text—is the correct goal.

Nor can a collection of artifacts judge God’s truthfulness. The evidence may expose our mistaken account of what God promised. It may require us to revise a historical argument offered in defense of preservation. It cannot demonstrate that a truthful divine promise failed while continuing to be interpreted independently of the Word in which that promise is given.

Evidence tests human claims. Scripture tests the principles by which evidence is made to speak.

The External Witnesses

The external evidence for a reading takes several forms. Each answers some questions well and other questions poorly.

Greek and Hebrew Manuscripts

Manuscripts provide direct witnesses to a textual form. Their age, provenance, material, corrections, marginalia, scribal habits, and relationships matter. A manuscript should be weighed according to its demonstrated character in the relevant book and passage, not treated as uniformly superior or inferior by label.

No manuscript is “the evidence” by itself. Each is one historical witness within a larger transmission. A famous codex does not acquire sovereignty through reputation, and a late minuscule does not become worthless through date alone.

Ancient Versions

Translations may attest that a source-language reading existed when and where the version was made. They can be especially valuable when their manuscripts preserve evidence earlier than surviving Greek or Hebrew copies.

Their limitations must be respected. A translation may not distinguish between two source words. Its surviving text may have been revised toward another language tradition. Retroversion—reasoning backward from the translation to its source—can be uncertain. A version proves only what its linguistic form can actually bear.

Citations by Early Writers

Patristic, rabbinic, and other ancient citations show that a reading was known, used, or remembered. They can also reveal geographic distribution and theological use.

But authors quote with different degrees of exactness. They may cite from memory, paraphrase, conflate passages, adapt grammar to a new sentence, or depend upon a translation. A printed edition of a church father may itself rest upon a complicated manuscript history. The surrounding argument must be read before a quotation is assigned textual weight.

Lectionaries and Liturgical Use

Lectionaries bear witness to the text read publicly in worship. Their repeated ecclesiastical use makes them relevant to reception history.

Liturgical forms may also modify openings, endings, names, or pronouns so that a selected passage can stand alone. A lectionary reading must therefore be distinguished from an ordinary continuous-text manuscript.

Printed Editions and Ecclesiastical Use

Printed Hebrew and Greek editions are comparatively late as physical witnesses, but they may embody earlier manuscripts and editorial traditions. They also provide direct evidence of the text publicly received, translated, preached, and confessed in the Reformation and post-Reformation churches.

Their late date does not erase their history, and their ecclesiastical use does not make every editorial decision inspired. They answer questions of reception more directly than they answer questions about the precise age of a reading.

The responsible critic never forces one kind of witness to answer a question it cannot answer.

Internal Evidence and Transmissional Reasoning

Internal evidence ordinarily includes two related inquiries.

Intrinsic probability asks which reading best fits the author’s language, grammar, style, context, and argument. Transcriptional probability asks which reading most plausibly gave rise to the alternatives through known habits of copying and editing.

Both are legitimate. Neither is mechanical.

Grammar and context matter because the Holy Ghost inspired meaningful language rather than disconnected marks. Authorial usage may reveal that one construction is natural and another suspicious. A reading that makes the sentence nonsensical may preserve a scribal error.

Yet claims about what an author “would have written” can become circular. If an editor first builds a picture of an author from a preferred text and then uses that picture to select the preferred text, the conclusion has been placed in the premise. Biblical writers may also use rare expressions, change style with genre or subject, and state things an interpreter did not expect.

Transcriptional reasoning likewise serves a real purpose. Similar endings may explain an omission. A marginal gloss may enter the text. A scribe may harmonize parallel accounts, simplify grammar, expand a title, or replace an unfamiliar form. A strong reading often explains the origin of competing readings better than its competitors explain it.

But the well-known slogans are not laws. The shorter reading is not always earlier; scribes omit as well as expand. The harder reading is not always original; scribes can create nonsense. The less harmonious reading is not always authentic; the biblical authors themselves may use common language. A possible explanation is not an established history merely because it can be imagined.

The analogy of Scripture also belongs to internal consideration, but it must not be abused. Because one divine Author speaks throughout Scripture, no canonical reading finally contradicts the whole counsel of God. Yet this principle does not authorize an editor to remove a well-attested reading merely because it is difficult or to select a weak reading because it states a doctrine more explicitly. The whole Scripture helps interpret each part; it does not permit theological preference to manufacture the wording of a part.

Useful Criteria That Must Not Become Kings

Many textual criteria contain genuine wisdom. Trouble begins when one is elevated into a sovereign rule.

• Antiquity matters, but the oldest surviving witness is not necessarily the purest.

• Numerical support matters, but manuscripts must be related and weighed rather than merely counted.

• Geographic distribution matters, but widespread error and localized fidelity are both historically possible.

• Difficulty matters, but corruption can produce a harder reading.

• Brevity matters, but omission is among the most ordinary scribal errors.

• Genealogy matters, but a model of relationships remains a human reconstruction and cannot supply evidence absent from its witnesses.

• Authorial style matters, but a circularly constructed style profile can simply reproduce an editor’s prior choices.

• Traditional use matters, but familiarity does not transform error into revelation.

• Ecclesiastical reception matters, but no congregation, printer, or century is individually infallible.

Confessional Bibliology does not replace the slogan “the oldest is best” with “the latest is best,” or critical eclecticism with a crude traditionalism. It refuses to let any isolated criterion become self-authenticating.

The criteria are servants within a larger judgment. Their proper weight depends upon the passage, the witnesses, the transmissional history, the church’s reception, and the biblical doctrine of the text being recognized.

Reception, Presumption, and the Burden of Proof

The historically received text possesses a meaningful presumption.

A presumption is not an infallible decree. It is a reasoned starting point that assigns the burden of proof. Because God preserved Scripture through the public life of His church, a reading broadly received, read, translated, preached, and confessed should not be displaced as though that history carried no theological or evidential significance.

The person proposing a departure should therefore do more than point to an older manuscript or cite a current edition. He should identify the evidence, explain the relevant relationships, account for the received reading’s origin and history, show how the competing reading better explains the record, and defend the principles by which the evidence is weighted.

This burden is proportional. An obvious typographical error in one printing requires little argument. The removal of a long-received clause with extensive ecclesiastical use requires much more. A claim that changes no translation may be stated differently from one that alters the public wording of Scripture.

The presumption also disciplines defenders of the received reading. They may not answer contrary evidence with indignation, invent manuscript support, or treat a historical question as irreverent merely because it is uncomfortable. The received reading must be defended truthfully. A false argument cannot become pious by reaching the desired conclusion.

Reception is evidence of providence. It is neither a substitute for evidence nor an item to be excluded before inquiry begins.

A Ministerial Order of Textual Inquiry

A faithful inquiry can be summarized in nine movements.

1. Confess the identity of the object

Begin with Scripture’s own doctrine: the text under consideration belongs to the inspired, infallible, preserved, and received Word of God. The critic is a servant examining witnesses to a rule already established by God.

2. Define the variant exactly

State every relevant reading without rhetorical labels. Identify whether the difference involves spelling, word order, substitution, omission, addition, or a larger unit. Explain whether it affects translation, interpretation, or doctrine.

3. Establish the evidence accurately

Verify manuscripts, corrections, versions, citations, lectionaries, editions, and reception history. Do not repeat inherited claims without checking them. Distinguish first-hand examination from reliance upon a collation or secondary source.

4. Keep fact and inference distinct

State what a witness reads separately from claims about why it reads that way. Distinguish a manuscript’s estimated date from the age of its reading. Mark conjecture as conjecture.

5. Evaluate witnesses according to their character

Consider relationships, patterns of accuracy, corrections, and the witness’s behavior in the relevant corpus. Avoid granting universal superiority to a document that varies in quality from book to book.

6. Examine intrinsic and transcriptional relations

Study grammar, context, authorial usage, parallel passages, scribal habits, and the ability of each reading to explain the others. Do not convert a useful maxim into a decision-making machine.

7. Account for ecclesiastical reception

Ask how the readings functioned in the historical church. Examine their place in public reading, translation, preaching, confession, and controversy. Give this history theological weight without treating it as new revelation.

8. Reach a proportionate judgment

State the conclusion with the confidence the evidence warrants. Not every variant requires the same certainty. Distinguish the authenticity of the canonical rule from the fallibility of the present argument.

9. Publish the reasoning in a corrigible form

Allow readers to examine sources and answer the case. Correct demonstrated factual errors conspicuously. A ministerial judgment remains accountable to the Word and evidence it claims to serve.

This order does not operate as an algorithm. It is a discipline of submission designed to prevent the critic, the confession, the tradition, or the favored edition from being used dishonestly.

Correction Without Autonomous Reconstruction

The word reconstruction can be used in more than one sense.

A historian may reconstruct the damaged wording of a particular manuscript, the sequence of changes in a printed Bible, or the probable relationship between several copies. Such reconstruction is ordinary ministerial historical work. David Norton’s distinction between the King James translators’ work and later printing or editorial changes is reconstruction in this limited sense.

The problem arises when the authentic text of Scripture is treated as an absent object whose identity is constituted by the critical process. The critic no longer corrects witnesses by a preserved rule; the critic’s method produces the operative rule from an undifferentiated field of variants.

Confessional Bibliology permits correction because particular witnesses are fallible. It rejects autonomous reconstruction because the canonical Word has not ceased to exist in the church’s possession.

The distinction can be stated simply:

Ministerial correction asks whether a witness conforms to the preserved text. Magisterial reconstruction makes the preserved text conform to the critic’s method.

That contrast concerns the structure of authority, not the motives or learning of every scholar. A critic may work with humility, reverence, and extraordinary technical ability while employing principles that give method a constituting role. Another may profess the Received Text while handling contrary evidence arrogantly or incompetently.

Titles and intentions do not settle the methodological question. The relation between Scripture, doctrine, evidence, and conclusion does.

The Moral Discipline of Handling Evidence

Textual criticism is performed by moral persons, not disembodied calculating instruments.

Scholars can become attached to editions, methods, teachers, institutions, controversies, and reputations. Confessional writers share the same vulnerability. Greg Bahnsen’s analysis of self-deception describes how a person may confront adverse evidence while distorting, overemphasizing, suppressing, or misconstruing it through motivated rationalization. The reasoning may retain the appearance of careful evidence-handling while functioning to protect a desired belief.⁷

This diagnosis must first produce self-examination. It does not authorize speculation about an opponent’s hidden motives. We may demonstrate that an inference is invalid or evidence has been omitted. We ordinarily cannot prove why the person made the mistake.

The Christian safeguards are therefore moral as well as technical. Hear before answering. Represent qualifications. Answer the strongest plausible form. Distinguish what a person said from what follows and from what is inferred about his intention. Acknowledge evidence favorable to the opposing conclusion. Correct errors even when they aided one’s argument.

The ninth commandment applies to a manuscript apparatus as surely as to ordinary speech. God’s truth requires no false witness in its defense.⁸

Four Objections Answered

“If doctrine governs the inquiry, the conclusion has been predetermined.”

Doctrine determines the identity of the object and the boundaries of a faithful method. It does not predetermine the date of a codex, the reading of a version, or the resolution of every variant. The doctrine of creation governs Christian science without supplying measurements apart from observation. Likewise, preservation governs textual inquiry without replacing collation, exegesis, and historical reasoning.

“Giving the received text a presumption makes contrary evidence irrelevant.”

No. A rebuttable presumption determines a starting point and burden of proof; it does not forbid rebuttal. Contrary evidence must be heard fully. The presumption rejects the opposite prejudice—that ecclesiastical reception has no significance until a modern method independently approves each reading.

“All textual conclusions are human and therefore only probable.”

Human judgments are fallible, but it does not follow that nothing can be known certainly. Christians know the canon through fallible acts of reading and reasoning. They understand many passages certainly without becoming infallible interpreters. The certainty of the divine rule and the confidence assigned to a particular editorial argument must be distinguished.

“Ministerial textual criticism simply baptizes preference for the Textus Receptus.”

It can be abused that way, but abuse does not define the method. Ministerial criticism requires evidence, permits correction of fallible editions, rejects false historical claims, and places the defender of a received reading under the same duties of accuracy and valid reasoning as everyone else. Its distinctive claim is not that evidence is unnecessary but that evidence serves within revelation, preservation, and reception.

What This Publication Will Require

Confessional Bibliology will not treat the name of a manuscript, scholar, edition, confession, or theological position as a substitute for argument.

Textual claims must identify their evidence. Quotations must be presented in context. The wording of witnesses must be distinguished from conclusions drawn about them. The strongest plausible opposing case must be answered. Degrees of confidence should reflect the record. Demonstrated errors must be corrected.

At the same time, the site will not pretend that textual evidence inhabits a realm religiously neutral toward Christ. It will ask what doctrine of Scripture a method assumes, what goal it assigns to criticism, how it understands preservation, what role it gives the church, and where it locates final authority.

The position is not evidence against theology or theology against evidence.

It is Scripture as supreme, theology as its disciplined confession, history as the realm of providence, the church as witness and keeper, and evidence as the servant of ministerial recognition.

The servant should be honored, trained, heard carefully, and corrected when necessary. It should never be crowned.

The next article will address a distinction frequently obscured in public controversy: The Confessional Text, the Authorized Version, and What We Do Not Mean by KJV-Onlyism.

Notes