Providence Redefined: A Manuscript Treasury Is Not a Text

<!– EDITORIAL INTEGRATION NOTE: This new series article must remain a companion to, not a silent replacement for, the published CB article “The Hidden-Providence Escape Hatch: Continuity When Convenient, Recovery When Necessary” (slug: hidden-providence-escape-hatch). Review title, thesis, links, and overlap before publishing both. –>

The doctrine of providence is sometimes made to carry more than it says. A Christian looks at the immense number of surviving manuscripts, ancient versions, quotations, lectionaries, printed editions, and translations, and he rightly gives thanks to God. The evidence is not nothing. It is a remarkable history of transmission through persecution, copying, travel, preaching, translation, print, and ordinary human labor.

But a large treasury of manuscript evidence is not yet a text.

That statement is not an insult to the manuscripts. It does not deny that God has preserved His Word through them. It does not demand one perfectly uniform copy in every place. It means only that a collection of witnesses containing different readings cannot, simply by existing, tell the church which reading is the authoritative wording at a disputed point. The witnesses are evidence. The text is the norm the church receives from that evidence. The difference between the two is the question.

Begin with the affirmation that must be made

Lane Keister has stated his view in emphatic terms. In 2024 he wrote that God, by an act of special providence, “most certainly did preserve the original text” of both Testaments “through the manuscripts.” He added that the New Testament is not second-class in preservation, inspiration, clarity, infallibility, sufficiency, or any other category relating to Scripture.¹

That is an explicit affirmation of providential preservation. It must be acknowledged before any criticism is offered. This article does not say that Keister denies preservation, denies providence, denies the New Testament’s authority, or believes Christians have no Bible. Those would be false descriptions of his stated position.

His earlier essay on the canons of New Testament textual criticism supplies an equally important part of the picture. Keister says that textual criticism has no explicit biblical guidelines, that his own canons will not match every other scholar’s, and that the canons have probabilistic force only. He also says that the critic’s judgment and imagination are an essential and unavoidable part of the discipline.² He speaks positively of God’s providential preservation of the text in all ages and calls textual criticism an exercise in reading the book of God’s providence.³

These statements put the real question in view. If God has preserved the original text through the manuscripts, and if the manuscripts contain variants, by what rule is the church to identify the original text among them? Is the preserved text identical with the total collection of witnesses? Is it a set of readings distributed across the witnesses? Is it the text printed in a particular edition? Or is it the reading a critic judges most probable after applying fallible and non-universal canons?

To ask those questions is not to reject providence. It is to ask what providence has accomplished in the doctrine being asserted.

Material preservation and formal preservation

The distinction at the center of this article is simple.

Material preservation means that biblical words, readings, books, and textual evidence have been preserved somewhere in the history of transmission. They may be found in manuscripts, versions, patristic citations, lectionaries, printed editions, translations, or other witnesses. The claim may be very robust: God has not allowed His Word to vanish from the earth.

Formal or identifiable preservation means that the church possesses an authentic original-language text whose wording can function as the final rule in a controversy. This does not mean that each individual is infallible in his private reading. It does not mean that the church is inspired in every historical decision. It means the church has a real text to which she may appeal, rather than only a field of data from which an edition must repeatedly be constructed.

The categories are not enemies. Formal preservation does not despise material preservation; it requires it as one of the ordinary means through which God has preserved and delivered His Word. The contrast arises only when material survival is treated as though it automatically supplied formal identity.

An illustration may help. A court archive may preserve every draft, letter, copy, correction, and marginal note associated with a law. That archive is valuable. Yet the archive is not, by itself, the statute book. If multiple drafts conflict, a court needs to know which form has legal force. More documents improve the historical record; they do not automatically determine the binding text.

So it is with manuscripts. A manuscript is not unimportant because it is not the final rule. It is important precisely because it bears witness to the history of the Word’s transmission. But an aggregate of manuscripts, especially one that contains alternatives, still needs a rule by which the church says, “This is the reading we receive as Scripture.”

A treasury contains alternatives

Consider what a manuscript treasury actually contains. It contains readings that agree with one another. It contains spelling differences. It contains transpositions. It contains harmonizations. It contains omissions and expansions. It contains corrections by later hands. It contains readings in the text and readings in margins. It contains witnesses whose relationship to one another is debated. It contains versions that reflect translation choices as well as source texts. It contains patristic citations that may be quotations, allusions, paraphrases, or memories.

None of that is scandalous. It is simply the history of copying.

But because the treasury contains alternatives, it cannot be identified with the final text without further explanation. If it is said that every reading in the treasury is equally Scripture, then incompatible readings make God speak incompatible words. If it is said that only some readings are Scripture, then the question becomes: which ones, and by what authority?

The answer cannot be merely, “The manuscripts preserve the text.” That sentence tells us where the evidence resides. It does not identify the reading when the evidence differs. Nor does it solve the problem to say, “The original text is distributed among the manuscripts.” A distributed text is still not a text the church can read until its wording is identified.

The task of identifying a text is unavoidable. The controversy is over the authority assigned to that task.

Confessional Bibliology says that the task is ministerial. The church receives an authentic Hebrew and Greek text as the public rule preserved for her; historical investigation, grammar, comparison, and textual study serve that rule. Modern eclectic reconstruction commonly says that the task is constitutive at disputed points: the reading becomes the church’s operative text because the critic’s method selects it as most likely original.

The difference is not between people who use evidence and people who refuse to use evidence. It is between two accounts of where final textual authority resides.

Keister’s canons make the question sharper, not smaller

Keister’s own 2017 canons illustrate why the question must be faced. He explicitly says that the canons have probabilistic force. He lists age, geographical diversity, genealogical relationships, the more difficult reading, explanatory power, continuity of attestation, the majority of manuscripts, and patristic evidence. He also gives caveats to each. Older manuscripts tend to attest older readings but are not always decisive. The harder reading is a tendency, but it has a limit: a reading that directly conflicts with Scripture cannot be original. Continuity matters, but hidden providence may account for an isolated witness. Majority evidence carries weight but is not automatically correct.⁴

There is much to respect in the candor. Keister does not pretend that a single slogan or one manuscript ends every case. He acknowledges the value of the Byzantine tradition, the weight of patristic evidence, and the possibility that different canons may point in different directions. He states that judgment is unavoidable rather than disguising the role of the critic.

But the candor makes the epistemic issue plain. If the canons are probabilistic, potentially conflicting, not explicitly prescribed by Scripture, and dependent on the critic’s judgment, then what gives the final result the authority of the Word of God? A scholar may honestly say, “On balance, this reading has the better claim.” That is an intelligible historical conclusion. It is not yet an explanation of how that conclusion becomes the church’s universal, invariant, and final norm.

The question is not answered by adding more canons. More fallible canons may refine a probability. They do not turn probability into a self-attesting authority. Nor is it answered by the good character or learning of the critic. A faithful scholar is still a fallible scholar. The church’s final rule cannot rest on the personal weight of whoever applies the rules most persuasively.

This is why the doctrine of providence must not be reduced to the survival of information. Providence, as the Confession uses it in Chapter 1, is connected to the church’s actual possession of an authentic text. If the result of providence is only a collection that requires perpetual reconstruction, then the historic conclusion has changed even if the word providence remains.

The proper role of ordinary means

The argument should not be mistaken for an attack on ordinary means. God ordinarily preserves His Word through copying, teaching, preaching, translation, printing, and the custody of His church. Manuscripts are ordinary means. Linguistic study is an ordinary means. Comparing witnesses is an ordinary means. Historical research is an ordinary means.

The question is whether ordinary means are servants of a text the church has received or whether they are made the authority that continually supplies the text the church lacks.

Ussher’s language is instructive. He speaks of the “marvellous preservation of the Scriptures,” saying that God has by His providence preserved them “and every part of them.”⁵ Yet he also acknowledges differences of reading in Hebrew and Greek copies and says that in most cases the true reading may be discerned by circumstances and the comparison of other places.⁶ He does not oppose providence to study. Nor does he define preservation as an indefinitely shifting balance of probabilities. Scripture remains the certain rule; the original languages remain authentic; study serves the Word.

The point is not that every seventeenth-century judgment should be repeated without examination. The point is that the historical Reformed doctrine of preservation did not treat God’s ordinary means as a substitute for the text they preserved. The means were not the final court. They were the means by which God gave the church the Word by which all other courts were judged.

Four questions a manuscript-providence account must answer

A view that says God has preserved the original text through the manuscripts must answer four questions clearly.

1. What is the preserved text?

Is it the total body of manuscripts? If so, how can a body containing differences function as a single text? Is it a set of readings? If so, where is that set publicly identified? Is it an edition? If so, which edition, and what gives it authority over its competitors?

2. What is the rule for choosing among differences?

Is the rule majority, age, geographical distribution, genealogy, internal coherence, scribal habits, continuity, a combination of these, or something else? If several rules conflict, which one rules the others? Is that priority established by Scripture, by the church’s historic reception, or by the individual critic’s assessment?

3. What authorizes the rule?

This is the deepest question. Even if a critic’s criteria are highly useful, what makes them the authorized court for determining the exact words of God? A probability rule cannot simply declare itself final. A scholarly consensus cannot simply appoint itself supreme. A reading may be historically preferable and still lack an account of how it bears canonical authority.

4. How can the church finally appeal to the result?

The Westminster Confession does not say merely that the church may consult the Hebrew and Greek materials. It says she is finally to appeal to the original-language Scriptures in religious controversies.⁷ A final appeal requires something more determinate than “the reading our current method favors with high confidence.” What happens when the next edition changes it? What did the church appeal to before the latest revision? What does a pastor say when the method assigns a different degree of probability to the words he is commanded to preach?

These questions do not demand intellectual dishonesty. They demand a doctrine of authority adequate to the claims being made.

The pastoral difference

This distinction reaches beyond an academic debate. A pastor must read a text, expound its words, apply them to his people, and defend doctrine by them. A translator must know what he is translating. A catechist must teach children that Scripture is the Word of God, not a rough approximation of what God once said. A believer facing a footnote needs more than the statement that the issue probably does not matter. He needs to know whether the Word before him is the Word God has preserved for His church.

Confessional Bibliology does not answer by inviting a believer to trust his own private ability to resolve every variant. It directs him to the received original-language text, the faithful translation of that text, the whole counsel of God, and the church’s ordinary ministry. It calls scholars to serve, not replace, the rule. It allows the church to say, with confidence appropriate to God’s promise, that she possesses the Scriptures God gave her.

A manuscript treasury is a blessing. It may strengthen a historical case, illuminate a reading, expose a copying error, and deepen appreciation for God’s ordinary care. But it is not the same thing as the text. The providence that preserves evidence must be shown to preserve a rule.

Until that is shown, a critical-text advocate cannot simply borrow the full confessional conclusion of providential preservation. He may affirm providence in a broad and important sense. He may rightly praise the survival of manuscripts. But he must still account for the authentic text WCF 1.8 says the church finally appeals to. A corpus of alternatives, however rich, is not yet that text.

Why the church cannot finally appeal to an apparatus

An apparatus is one of the clearest demonstrations of the difference between witnesses and a text. It is a useful scholarly instrument. It preserves variant readings, identifies evidence, and alerts the reader to places requiring study. A responsible minister may consult it. A translator may benefit from it. A researcher may spend years examining it.

But an apparatus cannot preach the sermon, settle the doctrinal dispute, or serve as the church’s final court. At the decisive point it gives alternatives. The church must finally read one wording as the text. If the response is that the editor’s main text resolves the problem, then the authority question simply moves to the editor. Why does that editor’s selection, rather than the alternative in the apparatus, carry canonical authority? If the response is that the main text may later be revised, then the church has not been given a final text but an interim decision.

The argument does not require contempt for critical editions. It requires clarity about their function. They may be powerful tools for historical inquiry. They cannot, on their own authority, convert a probability into the Word God has commanded the church to receive. A method that recognizes this limitation may serve ministerially. A method that silently makes its changing main text the rule of faith has assumed an authority it has not demonstrated.

The providential question is therefore not answered when a Christian points to an impressive apparatus and says, “God preserved the evidence.” The right response is gratitude, followed by the necessary question: “Where is the text to which the church is finally to appeal?”

Sources and correction

This article begins with Lane Keister’s express affirmation of providential preservation and does not claim that he denies the preservation, inspiration, authority, or usefulness of Scripture. It critiques the authority claimed by a probability-governed method, not personal motives or spiritual condition. The article is a companion, not a replacement, for the existing CB discussion of continuity and hidden providence; its overlap must be reviewed before publication. Corrections supported by the cited primary sources will be added in a dated update.

Series navigation: Previous: The Autograph Nobody Has · Part 6 of 18 · Next: The Spirit’s Witness, the Church’s Custody, and the Sanctifying Word