Counterevidence Is Evidence

An argument that cannot state the strongest evidence against itself is not ready to judge a text.

That is especially true when the proposed conclusion is grave. To say that a reading weakens the deity of Christ, accommodates a lower creator, divides Jesus from Christ, or removes bodily-resurrection testimony is not ordinary editorial criticism. The claim must survive the best objection that the passage, book, edition, and historical record can supply.

Counterevidence is sometimes treated as a courtesy to the opposing side. In this series it is part of the evidence itself.

A conclusion must report the strongest contrary evidence at every relevant scope and must permit that evidence to reduce, reverse, or defeat the proposed classification.

This is not neutrality between received and critical texts. It is ministerial judgment under the truth.

Four Required Searches

Packet Zero requires every dossier to search four places for counterevidence.

The immediate context

The clause before or after the variation may preserve the title, referent, flesh-language, creator predicate, or resurrection claim thought to be lost. Grammar may show that an omitted repeated subject remains the only possible subject. A narrative action may make an allegedly Gnostic implication impossible.

Immediate context is not always decisive. A direct statement can be weakened even when the reader can reconstruct it from the paragraph. But the context establishes what the local reading can and cannot be made to say.

The same book

An author may state the proposition explicitly elsewhere. John’s Gospel may identify a title used ambiguously in one verse. First Corinthians 15 may control the meaning of a disputed resurrection clause. First John’s repeated confession of Jesus Christ come in the flesh may stand against a separationist construal of one variation unit.

Same-book evidence is especially important because it bears upon authorial usage, recurring vocabulary, argument, and intended reference.

This search is also confessional rather than merely academic: Westminster Confession 1.9 directs the interpreter to search clearer scriptural places when the sense of another place is questioned.¹

The complete critical corpus

The full edition may contain direct and repeated contradictions of the heresy under review. NA28 retains many passages confessing the Son’s deity, creatorhood, true incarnation, suffering, and bodily resurrection. BHS prints the Leningrad-based Hebrew main text even where its apparatus reports alternative forms. No final indictment may be written as though these facts were absent.

Complete-corpus counterevidence determines the highest scope at which a conclusion remains true.

Critical readings that strengthen orthodoxy

The comparison must also identify cases in which a critical reading is stronger than another critical state or stronger, with respect to the defined proposition, than the received comparator.

Packet Zero seeded such questions in the UBS6/intended-NA29 ledger: the unbracketed Son-of-God wording at Mark 1:1, the expansion “Jesus Christ” at Acts 20:21, “from God out of heaven” at Revelation 20:9, and other possible D4 cases. Each still requires collation and grammatical review. They may not be excluded merely because they complicate the thesis.

Counterevidence Can Do More Than Add a Footnote

A weak audit lists an objection and leaves the conclusion untouched. A genuine adverse review allows the objection to change the result.

Counterevidence may:

• reduce D3 to D2 because an indispensable element is not actually asserted;
• reduce D2 to D1 because the positive implication disappears in context;
• reduce D1 to D0 because the supposedly lost predicate remains grammatically explicit;
• preserve D1 locally while defeating a book-level accusation;
• convert an apparent loss into D4 when the newer form strengthens the orthodox assertion;
• keep the row at DU_UNRESOLVED because the evidence is genuinely divided;
• disprove an alleged historical use or transmission link.

An objection that is mentioned but not permitted to perform any of these functions is decoration, not review.

The Adverse-Evidence Table

Every dossier should be able to complete a table like this before publication:

The right column must be filled before the article’s rhetoric is fixed.

Friendly Witnesses Also Require Adverse Review

Counterevidence does not come only from critical editors.

Burgon may preserve an important patristic lead and still overstate the recoverable transmission. Hoskier may report a codex or versional pattern without proving motive. Whitaker may accurately describe heretical rejection of Scripture without supplying the ancestry of a particular modern reading. Maynard may preserve a large historical dossier while individual citations still require inspection. Burgon’s own treatment of corruption by heretics includes evidence of changes made from orthodox motives, a fact that must be admitted when his hypotheses are audited.²

The same rule applies to hostile witnesses. A modern editor may describe a variation unit accurately and minimize its doctrinal consequence. A patristic writer may identify a heretical use but report an opponent polemically. A manuscript image may establish wording but say nothing about the scribe’s theology.

Sources are witnesses with proper jobs:

• editions and images establish text;
• grammar establishes assertion;
• historical witnesses establish reported use or accusation;
• collations establish distribution;
• confessions identify doctrinal propositions;
• none of them performs every task.

The adverse question is always: What could this source actually know, and what does it actually prove?

Falsifiability in a Confessional Project

Confessional commitment does not make the investigation unfalsifiable.

The project begins with a confession concerning the authentic Hebrew and Greek Scriptures and with a received textual control. It also asks empirical and analytical questions:

• What wording does NA28 print?
• What main-text change does UBS6 introduce?
• Does BHS print a different Hebrew form or merely report one?
• Does a sentence contain every element of AR-02 or GN-04?
• Is a reading attested before its alleged heretical inventor?
• Does the complete critical corpus directly contradict the proposed heresy?

These questions can receive answers inconvenient to a prior expectation. If they could not, the apparatus, manuscripts, witnesses, and grammar would be unnecessary.

The confessional position supplies the authority and theological standard. It does not grant permission to misstate the evidence. Whitaker’s insistence upon the Hebrew and Greek fountains is an appeal to the authentic text as judge, not an exemption from careful comparison.³

Counterevidence Protects the Stronger Indictment

There is a practical reason to publish every adverse fact: overclaiming hides the real offense.

Suppose a critical corpus retains many explicit anti-Arian passages while weakening five especially direct predicates. If the series says, “The critical text teaches Arianism,” a reviewer can answer by quoting the retained passages. The whole discussion then turns upon the exaggerated corpus claim.

If the series instead demonstrates:

• the exact received predicates;
• the exact critical changes;
• five sound D1 findings;
• the retained D4 evidence;
• the loss of concentration or redundancy;
• the absence of a proven D3 corpus entailment;

the opponent must answer the actual textual losses. The modesty is methodological, not rhetorical. It produces a harder case to evade.

The same is true if D2 or D3 is proved. A conclusion that has already disclosed the strongest orthodox counterevidence cannot be dismissed by producing evidence it has hidden.

The 169-Change Ledger as an Adverse Test

The UBS6/intended-NA29 dataset offers a controlled test. Packet Zero records 169 main-text changes across Acts, Mark, and Revelation. The series could search those rows only for omissions useful to its thesis. That would make the number impressive and the audit worthless.

Article 56 must therefore report:

• every D0 row;
• every D1, D2, or D3 row;
• every D4 row;
• every unresolved row;
• the proposition and scope assigned to each;
• the strongest adverse reading of every material classification.

The final pattern must emerge from the complete table. It may be a pattern of weakening, strengthening, neutrality, mixed effects, or no material relation to either heresy. The result has not been authorized in advance.

This is what makes the final indictment capable of correction. If a later apparatus, manuscript image, or grammatical argument changes a row, the ledger records the revision and the synthesis must be rebuilt.

Counterevidence and the Burden of Proof

The burden rests on the person making the stronger claim.

To claim D1, show the lost testimony.

To claim D2, show the positive element that favors the proposition.

To claim D3, show every indispensable element and defeat the contextual objection.

To claim a corpus pattern, publish the neutral and contrary rows.

To claim heretical use, identify the teacher, source, reading, and date.

To claim scribal intent or transmission, supply evidence beyond doctrinal compatibility.

Counterevidence need not prove the opposite in order to defeat the claim. It may be enough to show that the conclusion does not necessarily follow. This is particularly important under the Martin Rule: a possible heretical use cannot satisfy the burden for formal entailment.

A Rule for Writing the Article

The public article should not bury adverse evidence in the final paragraph. It belongs beside the claim it tests.

A sound order is:

1. state the proposition and reading;
2. explain the prima facie doctrinal effect;
3. present the strongest immediate objection;
4. inspect the same-book evidence;
5. inspect corpus and D4 evidence;
6. state what survives the adverse test;
7. separate the effect verdict from provenance and motive;
8. identify what evidence would reopen the judgment.

This order allows the reader to watch the claim become narrower, stronger, or fail.

Verdict

Counterevidence is not a concession granted after the argument. It is one of the materials from which the argument is made.

Every article in this series must report:

• the strongest contextual evidence against its first reading;
• the strongest same-book evidence;
• the strongest complete-corpus evidence;
• critical readings that preserve or strengthen orthodoxy;
• weaknesses in friendly historical witnesses;
• facts that lower or defeat a provenance claim.

The evidence must be allowed to change the classification. Otherwise the series would merely reproduce the selective method it criticizes.

What This Does Not Prove

This article does not assume that every objection is sound, that opposing evidence always cancels a local loss, or that confessional judgment must wait for scholarly unanimity. Counterevidence must itself be verified, weighed, and placed at the correct scope.

It also does not assign final D4 status to the seeded UBS6 examples. Those rows remain open until the editions and grammar are audited.

Text and Edition Record

• Project control: Packet Zero, Governing Rules 9, 10, 13, and 14.
• Required adverse scopes: immediate context, same book, complete critical corpus, and stronger critical readings.
• Source-quality control: edition/image, primary historical witness, specialist analysis, audited polemic, discovery lead.
• English Scripture: Authorized Version.
• All seeded textual examples retain open status.

Bibliography

• Burgon, John William. The Causes of the Corruption of the Traditional Text of the Holy Gospels. Chapter 13.
• Hoskier, Herman C. Codex B and Its Allies. 2 vols.
• Whitaker, William. A Disputation on Holy Scripture. First Controversy.
• Westminster Assembly. The Westminster Confession of Faith, chapter 1.
• The Critical Text and the Two Ancient Heresies: Packet Zero. Confessional Bibliology Research Center, 2026.

Revision History

• 2026-07-24: First Packet Three draft; established four mandatory counterevidence searches, verdict-changing adverse review, source-role limits, and exhaustive reporting for the 169-change ledger.

Notes and Sources

Source-control note: The requirement to seek and publish adverse evidence is a project publication gate. Counterevidence may narrow, reverse, or leave unresolved a proposed classification. It does not automatically establish originality or erase a verified local loss.

Previous in the series: “Local Reading and Complete Corpus.”

Next in the series: “Textual Effect, Historical Use, and Scribal Intent.”