Recursive Heresy and the Boundary of Christian Confession

A controversy can end too easily in one of two failures. The first is timidity. After tracing a method to a doctrinal contradiction, the critic retreats to the harmless observation that Christians may disagree. The second is presumption. After finding the contradiction, he leaps from a public argument to the secret condition of another man’s soul. One failure refuses to name a boundary. The other claims a jurisdiction no essay possesses.

This series has tried to avoid both. Wesley Huff publicly commends the Christian Scriptures, defends the resurrection, teaches the Trinity, and recommends explicitly Trinitarian resources.¹ Those facts matter. They are not small concessions inserted for politeness. They are evidence concerning his avowed faith, and any account that suppresses them would misrepresent its subject.

The same man also presents an open historical reconstruction as the route by which the church identifies the original text. In his popular account, readings may be removed from the received Bible on the judgment of ancient and reliable manuscripts; the resulting changes do not affect any major doctrine; and modern translations may therefore be trusted as accurate representations of the originals. The preceding parts tested those claims case by case. They found some strong critical readings, some historical stories stated more confidently than the evidence permits, and several losses that cannot be described merely as differences in wording. At 1 John 5:7, the shorter text leaves a precise numerical premise unproved: the scriptural Godhead contains exactly three persons, neither fewer nor more.

That finding does not prove that Huff believes in four divine persons. It proves that his stated textual and apologetic premises do not close the door against a fourth. His orthodox profession supplies the closure his public derivation lacks.

The final question is therefore not whether one defective inference instantly makes a man a heretic. It is whether an error can become self-protecting: the method removes a premise; inherited orthodoxy is invoked to say that nothing has been lost; that inherited orthodoxy then shields the method that could not derive it. If this cycle is knowingly defended after its consequence is understood, the error no longer remains a private defect in textual procedure. It reaches the boundary of Christian confession.

This essay calls that cycle recursive heresy. The phrase does not name a new ecclesiastical offense or empower a writer to pronounce an anathema. It describes an argumentative structure by which a heterodox consequence borrows the authority of the orthodoxy it weakens. The structure must be exposed, the boundary stated, and the judgment assigned to the right category. Doctrine is judged by what it says. A person is judged only according to public profession, demonstrated understanding, correction, persistence, and the authority competent to hear the case.

What recursive heresy means

An ordinary error can be isolated. A reader misunderstands a verse, an editor misdates a manuscript, or a teacher overstates the force of a variant. The claim is corrected without rearranging the whole system. Recursive error behaves differently. Its conclusion becomes a premise in the defense of the method that failed to establish the conclusion.

The sequence in this controversy has four turns.

First, an editorial rule excludes or demotes a disputed reading. The rule may appeal to age, geographical spread, genealogy, internal probability, or a committee’s reasoned judgment. At 1 John 5, it prints the earthly witnesses without the heavenly witness.

Second, the doctrinal effect is measured against an inherited Christian system. The Trinity still stands, the advocate says, because Father, Son, and Holy Ghost appear throughout the New Testament. The textual change therefore affects no major doctrine.

Third, exact entailment reveals a remainder. The undisputed passages establish one God, identify three divine persons, distinguish them, and display their coordinated action. They establish at least three. The conclusion “there are exactly three persons in the Godhead” adds exclusivity. Unless some premise numbers the Godhead or excludes every additional divine person, the conclusion is stronger than the stated evidence.

Fourth, the inherited conclusion is returned to the method as proof of harmlessness. We know there are exactly three because Christianity teaches the Trinity; therefore the reading that gives numbered unity can be removed without doctrinal consequence. Yet the question under examination was whether the retained text supplies the exact doctrine. The confession has paid the method’s debt and is then presented as evidence that no debt existed.

That is the recursion. It does not require a conspiracy, conscious dishonesty, or hostility toward the Trinity. A scholar may inherit an orthodox conclusion so securely that he never notices where his proof has ceased. The logical circle exists before the moral explanation is known.

The pattern also explains why “no major doctrine” is an unstable reassurance. If major doctrine means a conclusion already held by the church, the claim can survive almost any local subtraction. The creed functions as a reserve account. Each deleted premise is declared nonessential because the balance still looks healthy. But a doctrine’s present possession and its derivation from the proposed canon are different assets. A church may remember more than its new method can prove.

Recursive error becomes more serious when correction is answered by the same circle. Suppose the critic identifies the missing exclusivity premise. The advocate replies that no Christian would infer a fourth person. That is sociologically plausible and logically irrelevant. The countermodel need not be a likely belief; it need only satisfy the retained premises while falsifying the conclusion. An open triad—three revealed divine persons without a proved maximum—does exactly that.

Suppose the advocate then appeals to the ecumenical creeds. The creeds truthfully confess the exact Trinity, but their authority does not show that the shorter critical corpus entails every proposition they contain. It shows that the church historically received and formulated that proposition. Unless the argument permits ecclesiastical tradition to complete Scripture’s premises, the appeal changes the rule in the middle of the case.

Suppose, finally, he cites threefold benedictions and baptismal language. Those passages are indispensable evidence for the Trinity. They are still lists of three, not universal quantifiers over the Godhead. A photograph containing three lamps does not prove that no fourth lamp exists outside the frame. Naming Father, Son, and Holy Ghost does not, by grammar alone, state “these three and no other are the one God.”

The recursive error is thus exact. It is not “textual criticism leads to atheism,” nor “everyone who prefers the shorter reading denies the Trinity.” It is this: a method that cannot derive an indispensable boundary uses the inherited boundary to certify the adequacy of the method. Once identified, the cycle requires an answer. Either produce a valid closing premise from the admitted canon, revise the claim that the method preserves the exact doctrine, or revise the method’s authority over the text.

The boundary is a proposition, not a mood

Historic Christian confession does not mean feeling warmly toward Jesus, using the word Trinity, or naming three divine persons in worship. It makes claims. The Westminster Shorter Catechism gives the numerical boundary with admirable economy: there are three persons in the Godhead—Father, Son, and Holy Ghost—and these three are one God, equal in power and glory. Westminster Confession 2.3 adds distinction, consubstantiality, coeternity, and the personal relations by which the three are not interchangeable.²

The number belongs to the doctrine. Remove “three” and the doctrine changes. Replace it with “at least three” and the confession becomes compatible with a fourth divine person not yet revealed or not named in the familiar formulas. Replace it with “Father, Son, and Spirit are divine” and tritheism remains open unless their one undivided essence is also confessed. Say “one God who appears in three ways” and modalism remains open unless personal distinction is confessed. Orthodoxy is bounded because error can preserve much Christian vocabulary while denying one governing relation.

The exact proposition is therefore richer than a count, but it is not poorer than one. Christian confession asserts one and only one God; exactly three divine persons; the full deity and true personality of each; real personal distinction; one undivided essence; equality of power and glory; and the eternal relations by which the Father is unbegotten, the Son begotten, and the Spirit proceeding. Different Reformed standards arrange those truths differently, yet none treats the maximum number of divine persons as optional.

The Comma does not single-handedly state that entire doctrine. “There are three that bear record in heaven” names and numbers the heavenly witnesses; “these three are one” joins the numbered witnesses in unity. The words require interpretation, and the exact doctrine arises cumulatively with passages teaching monotheism, personal distinction, deity, and relation. The received reading is not a magic incantation. Its relevance is that it supplies a kind of premise the shorter text lacks: a numbered heavenly unity.

Even here the Martin Rule forbids exaggeration. The Comma says “three”; it does not add the word “only.” Closure depends on the contextual comprehensiveness of the heavenly witness and its identification with the Father, the Word, and the Holy Ghost. The received case is a cumulative good-and-necessary-consequence argument, not a refusal to reason. Part 11 stated that limitation because exactness must govern the defender as strictly as the critic.

The shorter-text advocate may attempt a different closure. He might argue that the singular baptismal “name” of Father, Son, and Holy Ghost presents a complete covenantal identity; that the apostolic benediction embraces the whole source of salvation; or that the canon’s revelation of God is complete and nowhere names another divine person. Such arguments deserve a hearing. What they may not do is hide their universal premise. “The formula names three” is not “the formula exhaustively names every divine person.” “No fourth is revealed” is not “Scripture reveals there can be no fourth” unless the completeness relevant to that inference is established.

The boundary matters because an open-ended Godhead is not merely a thinner explanation of the Trinity. It is a different proposition. A man who says, “I worship Father, Son, and Spirit but allow that other divine persons may exist,” does not confess the historic Trinity. His three may be the same three names, and his worship may imitate Christian form, but his doctrine denies the precise limitation expressed by tri-.

Scripture warns against treating the content of confession as indefinitely elastic. The apostolic test concerns abiding in the doctrine of Christ, not merely speaking respectfully of Christ. Paul tells the church to mark and reject divisive teaching after patient admonition; John refuses religious partnership to teaching that does not bring the apostolic doctrine.³ These texts do not hand an online writer a judicial gavel. They do show that doctrinal boundaries are part of Christian charity. A boundary that can never exclude anything is not a boundary, and a confession unable to state what contradicts it cannot guard the flock.

The issue must therefore be expressed without spectacle. An open-ended Godhead is incompatible with the Christian doctrine of God. A textual method is doctrinally defective at this point if its admitted premises prove only an open triad while its advocate claims an exact Trinity has been reconstructed. That judgment concerns propositions. It is true whether the person making the argument is devout, careless, confused, inconsistent, self-deceived, or in the process of correction.

The Category Rule judges doctrine and person separately

The Category Rule prevents a doctrinal judgment from impersonating a verdict on a soul. It places the evidence into distinct columns.

The first column is public profession. Huff publicly presents himself as a Christian apologist, defends central Christian claims, and offers teaching and recommended resources on the Trinity. Nothing examined in this series justifies erasing that profession. A critic who writes as though Huff openly worships four gods would be bearing false witness.

The second column is derivation. Does the textual and apologetic method he publicly describes entail the exact doctrines with which he reassures his audience? On the numerical closure of the Trinity, this series has answered no. The answer is an argument about premises, not a report about his private belief.

The third column is implication. What would follow if the method were applied consistently and no inherited premise were borrowed? The result at this point is an open triad: the critical corpus identifies three divine persons but, under the proof map tested here, does not exclude every additional person. Implication is not profession. People regularly affirm conclusions stronger than their stated reasons warrant.

The fourth column is culpability. Did the person understand the gap? Was it clearly presented? Did he have access to the adverse evidence? Did he suppress it, misunderstand it, defer judgment, or revise his view? Did fear, loyalty, career, or another motive govern his attention? Those are factual and pastoral questions. They cannot be answered by moving a sentence from the second column into the fourth.

Bahnsen’s work on self-deception makes the restraint especially important. A person can avow one belief while relying inferentially on another; human belief sets can be inconsistent. But inconsistency alone does not prove self-deception in the morally charged sense. That judgment requires evidence of adverse knowledge and motivated management of evidence—rationalization, selective attention, or some comparable evasion.⁴ This series has not supplied that evidence about Huff’s inward conduct. It has tested publicly available arguments.

The older Reformed writers observed the same jurisdictional limit. James Ussher distinguished the outward call and profession by which the church recognizes Christians from the inward saving work known infallibly to God. He also placed severe ecclesiastical censure after instruction, evidence, and obstinate resistance, not before them.⁵ That order protects truth and neighbor together. The church does not call error harmless; neither does it pretend omniscience.

This is a lay publication. It is not Huff’s congregation, presbytery, employer, or examining court. It can quote, analyze, warn, invite correction, and state the doctrinal consequence of premises. It cannot administer church discipline. It cannot determine regeneration. It cannot know whether an argument already abandoned in private remains a present conviction.

The distinction does not make public criticism toothless. A teacher who speaks publicly may be answered publicly. Listeners may be warned that a reassurance has not been proved. Churches may ask their own officers to articulate the missing premise. Scholars may be pressed to distinguish historical probability from divine authority. The critic may say, without qualification, that an open Godhead is not the Trinity.

What he may not say without additional evidence is that every defender of the shorter reading knowingly embraces an open Godhead, that Huff has consciously chosen such a doctrine, or that a failed derivation proves personal apostasy. Those conclusions would violate the same rule of exact entailment this series has applied to textual claims.

The Category Rule also prevents profession from ending the inquiry. “He says he is Trinitarian” answers the first column; it does not repair the second. A sincere confession may be logically unsupported by a stated method. Indeed, that is the central allegation: inherited orthodoxy is doing work the reconstruction cannot do. Respect for a profession requires examining whether the teacher’s public reasons preserve the thing professed, not using sincerity as an exemption from logic.

Doctrine and person must therefore be judged separately without being divorced. A proposition can place pressure on a profession. Persistent response to correction can reveal what a person is willing to affirm. But the movement from one category to another occurs through evidence: understanding, avowal, persistence, and competent judgment. It does not occur by rhetorical acceleration.

The conditional verdict and the way back

The final verdict must keep its conditions visible. If a teacher understands that his admitted textual premises establish no more than an open triad; if he agrees that historic Christianity confesses exactly three persons in the one God; if he is shown that his reassurance imports this exact conclusion rather than deriving it; and if he nevertheless knowingly and persistently defends the open-ended result as sufficient Christian doctrine, then that defense is incompatible with Christian confession. He cannot coherently affirm both that the Godhead may contain more than three persons and that the Godhead contains exactly three.

The conclusion concerns the confession maintained under those conditions. It does not establish that the conditions presently obtain in Huff’s case. His public Trinitarian profession is evidence against attributing the open conclusion to him as a personal creed. The unresolved fact is how he would answer the demonstrated derivational gap: produce closure, limit his “no major doctrine” claim, revise the authority assigned to the critical method, or accept the open consequence.

That is why the charge is conditional rather than evasive. A conditional can draw a real boundary. “If a man knowingly and persistently teaches that additional divine persons may exist, he does not confess the Christian Trinity” is not softer than an unsupported personal accusation. It is more exact. It tells every reader, including the author, where the line lies and what evidence would be needed to apply it to an individual.

The way back is also exact. It is not the abandonment of manuscript study. Parts 4 and 8 found serious reasons for shorter or critical readings; Part 9 did not pretend that the late surviving Greek support for the Comma is strong. Historical comparison remains a legitimate servant. Whitaker’s hierarchy places every human judgment—including churches, councils, and scholars—beneath the Holy Ghost speaking in Scripture.⁶ The correction is to refuse historical reconstruction magisterial power over the church’s final textual norm, or else to supply from the admitted critical canon a valid argument that closes the exact doctrinal proposition.

Nor does the way back require pretending that every inherited proof was equally sound. Confessions may be tested by Scripture. Proof texts may be improved. A doctrine may rest cumulatively on many passages rather than one. What cannot be done is to remove a premise, retain the conclusion by inheritance, and describe the result as a complete derivation without identifying the substitute.

For the textual scholar, repentance may take the modest form of corrected speech. “The shorter text preserves abundant evidence for Father, Son, and Spirit, though I have not demonstrated from it the exclusive number of divine persons” is honest. “Most central Christian teaching remains widely attested, though individual propositions and events are lost under some critical decisions” is defensible. “The current edition is a reasoned and revisable scholarly reconstruction” accurately names the product. None of those sentences gives the reconstruction the confessional office of an authentic text kept pure in all ages.

For the church, the remedy is source stewardship. It must know not only what it confesses but which scriptural premises bear each claim. It must not spend theological capital inherited from one textual corpus while telling the next generation that the excluded readings contributed nothing. A creed detached from its proof map can survive for a time through memory. It cannot expose the borrowing that makes its future revision seem inevitable.

For the apologist, the remedy is to stop reassuring before the audit is complete. The skeptic’s exaggerated claim that “thousands of variants destroyed the Bible” should be answered with proportion, evidence, and a doctrine of providence. It should not be answered by making all changes doctrinally weightless. Some variants are spelling; some change grammar; some remove events; some exchange predicates; some alter the premises available for exact confessional claims. Truth does not need a slogan broad enough to hide these differences.

For the critic of the critic, the remedy is equal discipline. Do not inflate every omission into apostasy. Do not turn a late manuscript count into an inspired statistic. Do not invent motives, repeat the Erasmus-wager legend, or make the received case depend upon historical claims the sources will not bear. A defense that violates truth to protect Scripture contradicts the Scripture it names.

The series ends, then, where it began: with authority. “One Bible, many versions” can be a useful description of translation. It cannot settle which words constitute the one Bible. A manuscript archive supplies witnesses. An editor supplies judgments. A translation supplies English. A confession supplies an inherited doctrinal map. None may silently perform the office of another.

The Martin Rule asks for the exact proposition, every indispensable premise, valid entailment, closure against countermodels, and a ledger of borrowed claims.⁷ Applied to these cases, it has not produced one theatrical verdict. Some modern readings are historically strong. Some received readings preserve real propositions modern translations omit. Some doctrines survive cumulatively; some exact formulations remain dependent upon a disputed premise. The method has been required to live with the differences.

At the final boundary, charity and precision speak together. Huff’s public profession should be reported truthfully, and so should his method’s numerical gap. No verdict on his heart follows from the gap; no profession makes the gap disappear. If correction is received, the recursion breaks. If an open Godhead is knowingly and persistently defended, the boundary has been crossed. Christian confession is protected by naming that boundary while refusing to claim God’s knowledge of the heart.

Notes and Further Reading

Series navigation: Previous: “The Method on Trial” · Part 14 of 14 · End of series