“Tweaking,” “Plus,” and “Minus”: The Rhetoric by Which Critical Revision Becomes Neutral

Revision note: Substantially revised July 20, 2026. The argument has been reordered according to the classical rhetorical movement, compressed where necessary, and given one consolidated Notes and Further Reading section.

“Plus” and “minus” can be useful terms because they describe the formal relation between two readings without prematurely deciding whether an ancient scribe added or omitted words. But the terminology does not make the modern editor’s decision neutral. A reading placed in the apparatus is not functioning like a reading printed in the main text; a passage bracketed as doubtful is not being presented like an unqualified part of Scripture; and a congregation does not ordinarily hear a variation unit—it hears one continuous Bible. Fair terminology should postpone conclusions that have not yet been proved. It should not conceal decisions that have already been made.

That symmetry prepares us to examine the vocabulary of the debate.

Lane Keister objects when Received Text advocates say that the critical text “omits” words or that modern critics have “removed” passages. Such expressions, he argues, assume the very point in dispute. If the longer reading was not original, then a shorter text has not omitted inspired words; it has declined to print words that entered the transmission later. Keister recommends the formally comparative terms plus and minus rather than omit and insert. Elsewhere, in rejecting a claim about the program of nineteenth-century critics, he writes that they were “tweaking the text based on new evidence.” These statements appear in his chapter-by-chapter review of Why I Preach from the Received Text.

Keister explicitly affirms preservation. His disagreement concerns its definition and textual location, not whether God preserved Scripture. Keister has identified a real problem. Textual advocates often smuggle their conclusion into their verbs. A defender of the Received Text says that the critical text deleted a phrase, as though its originality were already established. A defender of the critical text says that the Received Text added the phrase, as though its secondary character were already established. Both formulations can beg the question.

Yet Keister’s solution is incomplete. Plus and minus can suspend a verdict about the history of transmission. They cannot suspend the editorial decision about what will be printed as the text. Nor can tweaking tell us whether the change is theologically or ecclesiastically small. These terms are useful within a carefully delimited task. When allowed to govern the entire discussion, however, they compress several different questions into a vocabulary that makes critical revision sound merely descriptive. No accusation about Keister’s motive is needed. The issue is the function of the language.

Keister’s Objection at Its Strongest

Suppose one Greek text reads A B C, while another reads A C. Before the textual history has been adjudicated, at least three explanations are possible:

1. B was original and disappeared during transmission. A scribe, an exemplar, or a branch of transmission omitted it.

1. B was not original and entered during transmission. A scribe, an exemplar, or a branch added it.

1. The surviving evidence does not permit a confident account of direction. We can describe the difference while withholding judgment about its origin.

If we immediately call the shorter reading an “omission,” we have chosen the first explanation before arguing for it. If we immediately call the longer reading an “addition,” we have chosen the second. At this stage, A B C has a plus relative to A C and A C has a minus relative to A B C state the formal relation without assigning guilt to a scribe or priority to a reading.

That is a genuine gain in precision. The same care should govern Received Text argumentation. It is not enough to place a critical edition beside Scrivener, circle words absent from the former, and count every absence as the removal of Scripture. The conclusion depends upon whether the words are authentic. That question cannot be answered by typography, subtraction signs, or an English comparison alone.

Keister is also right to resist caricatures of textual critics as men casually cutting words from the Bible. Conservative critics ordinarily profess the opposite intention: to identify the words God inspired by weighing the surviving evidence. Their judgments may be mistaken. Their doctrine of preservation may be inadequate. Their method may assign the wrong role to ecclesiastical reception. None of those criticisms warrants inventing a malicious purpose.

Thus two principles should be granted without reservation:

• Do not use a historical verb before establishing the historical direction. “Added,” “interpolated,” “omitted,” and “deleted” require argument when they describe ancient transmission.

• Do not infer a hostile motive from an editorial result. A critic can exclude a received reading because he sincerely believes it secondary.

These principles protect the discussion from circularity and slander. But they do not establish the neutrality of the resulting text. Intention, transmissional history, editorial action, and ecclesiastical effect are different subjects. A fair analysis must keep all four in view.

Four Questions Hidden in One Variation Unit

The vocabulary becomes clearer when the questions are separated.

Level of analysis: Formal comparison

Question being asked: How do two extant readings differ?

Proper form of language: “Reading A has these words; Reading B lacks them,” or “A has a plus relative to B.”

Level of analysis: Transmissional history

Question being asked: What happened during copying?

Proper form of language: “The shorter reading likely arose by omission,” or “the longer reading likely arose by expansion,” followed by evidence.

Level of analysis: Editorial action

Question being asked: What did an editor or committee print?

Proper form of language: “The edition retained, excluded, bracketed, or relegated the words to the apparatus.”

Level of analysis: Ecclesiastical consequence

Question being asked: What text is read, translated, preached, memorized, and used in controversy?

Proper form of language: “The words appear in the church’s continuous text,” or “they are absent from it or marked as doubtful.”

Plus and minus belong primarily to the first level. They are relational descriptions. They may help the scholar avoid prejudging the second level. But the third and fourth levels remain even when the second is uncertain. An editor may honestly say, “I do not know whether these words were lost from one branch or added in another.” He must still decide whether his edition will print them in the main text, enclose them in brackets, or confine them to the apparatus. A translation committee must decide whether they will appear in the body, a footnote, or nowhere. A minister must decide what he will read as the sermon text.¹

Suspending the verdict about scribal direction does not suspend the editor’s decision. This distinction also permits more exact use of words such as remove. Saying “a fourth-century scribe removed these words” is a historical claim and requires proof. Saying “a twentieth-century edition removed these words from the main text of its predecessor” is an editorial comparison and may be established directly by comparing editions. Saying “this translation no longer places the passage in the body where its predecessor placed it” is likewise a verifiable publishing fact. Neither statement, by itself, proves what the apostle wrote.

The language is not forbidden. Its reference must be identified.

A Plus Is Always a Plus Relative to Something

The apparent neutrality of plus and minus depends upon a reference text that can easily disappear from view. If Scrivener contains a clause that NA28 lacks, Scrivener has a plus relative to NA28. Reverse the reference point, and NA28 has a minus relative to Scrivener. Neither statement tells us which reading is original. The signs record relation, not authority.

An apparatus, however, is constructed around a main text. Once an editor prints the shorter reading in the main line, the longer reading may be represented as a plus in certain witnesses. That notation is perfectly intelligible. Yet the main text is not merely one more coordinate in a neutral comparison. It is the edition’s proposed text. It will supply the base for translation, citation, exegesis, and usually public reading.

The reference point therefore matters twice. Grammatically, it tells us what plus means. Editorially, it tells us which reading has been installed as the text from which other readings are measured. This is why “the TR has a plus” can communicate more than its speaker may intend. In strict apparatus usage it means only that the TR is longer at that locus relative to the chosen base. In ordinary discourse, however, plus readily sounds like extra: something attached to the text rather than belonging to it. Conversely, “the critical text has a minus” may sound like deprivation even if the phrase is meant only relationally.

The cure is not to ban either expression. It is to name the reference and limit the conclusion:

• “Scrivener contains these words; NA28 does not.”

• “Relative to NA28, the Received Text has a plus at this locus.”

• “This formal difference does not yet decide whether the words are original.”

Accountable language states what the comparison proves and what it leaves open.

The Apparatus May Be Symmetrical; the Main Text Is Not

At the level of data, two readings can be displayed symmetrically. Both can be transcribed. Both can be assigned witnesses. Both can be studied without ridicule. This is an indispensable virtue of a good apparatus. But a continuous edition is necessarily asymmetrical. It prints one sequence of words. Even an edition that uses brackets, diamonds, or other signs to express uncertainty must decide which words occupy the line. The apparatus may preserve knowledge of the alternative; it does not give the alternative the same textual function.

This distinction is especially important in theological debate. To say that no reading has been “lost” because every known reading remains somewhere in an apparatus confuses preservation of evidence with possession of a text. A variant documented below the line is available to the specialist. It is not thereby presented to the church as the unqualified wording of Scripture.

A congregation does not ordinarily read an apparatus. It reads the main text of a translation derived from an edited Greek text. A child memorizes the words in the body, not the sigla beneath them. A pastor may consult the note, but he must finally preach a wording. An elder answering doctrinal controversy cannot treat every listed variant as simultaneously canonical.²

Formal symmetry is therefore not ecclesiastical equivalence.

This does not mean that every apparatus decision is wrong or that a received reading is original merely because generations read it. It means that textual location performs a real act. Main text, brackets, footnote, and absence are not interchangeable containers. They communicate different judgments about status, and those judgments shape the church’s use.

Neutral notation can be useful. Neutral textual authority is impossible.

What Does “Tweaking” Measure?

Keister uses tweaking while denying that nineteenth-century critics set out to undermine and replace the traditional text. His correction contains an important historical caution. Claims about what critics “set out” to do are claims about purpose. They require evidence from prefaces, correspondence, published principles, and institutional history. An unfavorable result does not prove a hostile intention.

But tweaking does not simply correct a claim about motive. It also suggests scale. Sometimes the word is apt. The great mass of the Greek New Testament is shared across the Received Text and modern critical editions. Many variants concern spelling, movable letters, word order, synonymous forms, or differences that do not survive translation. Describing every one of these as a crisis would be irresponsible. The church should be told the extent of agreement as well as the places of dispute. Yet tweaking becomes misleading when it functions as a blanket description of the whole revision. It can collapse at least three different measures:

1. Numerical measure: What proportion of the total words differs?

1. Semantic measure: Does a particular difference affect a translatable word, clause, or passage?

1. Authority measure: Who or what now determines which wording is printed as Scripture?

The first may be small while the second is significant at selected places and the third represents a fundamental change in textual government. A small percentage does not decide the status of any one disputed reading. Nor does overwhelming agreement explain whether criticism ministers to a received text or continuously constitutes the text from an evidentiary aggregate.

A printer may alter only a few lines of a legal instrument, a creed, or a will; the small proportion of altered words does not by itself establish that the revision is insignificant. The analogy does not prove that the Received Text is correct. It proves only that magnitude cannot be measured by word count alone.

Keister’s own position illustrates the point. He identifies as Sturzian, accepts the longer ending of Mark, accepts the traditional reading at 1 Timothy 3:16, and remains open on John 7:53–8:11. Those are not the judgments of a man who thinks every difference is meaningless. He properly believes that individual readings must be assessed. But once that is granted, tweaking cannot settle the assessment. It describes the total revision from a distance while the debate concerns particular words and the rule by which they are judged.

The preservationist should therefore make two concessions and one denial. He should concede the vast agreement between the texts. He should concede that many individual differences are minor. He should deny that either concession proves all disputed readings minor or makes the governing method confessionally adequate.

Tweaking describes magnitude. It cannot decide authority.

Typography Is an Editorial Argument

The rhetoric of neutrality is not confined to spoken labels. Editions argue typographically. Words printed in the uninterrupted main text are offered for ordinary reading. Words enclosed in brackets are offered under a sign of doubt. Words transferred to a footnote are made available while being denied the same place. Words recorded only in a technical apparatus remain accessible to trained readers but disappear from most ecclesiastical use. Words not reported at all are functionally unavailable to the edition’s ordinary reader.³

None of these acts is necessarily dishonest. Editors need ways to represent evidence and uncertainty. Brackets can be more transparent than silent exclusion. Notes can alert readers to serious alternatives. An apparatus can preserve an extraordinary amount of historical information. Nevertheless, a bracket is not an absence of judgment. It is a judgment rendered typographically. A footnote is not the same thing as main text. The scholar who says, “The reading has not vanished; it is in the apparatus,” has answered an archival question, not the church’s canonical question.

This point applies equally to a Received Text edition. Printing a reading in the main text does not prove it apostolic. A traditional editor cannot convert reception into demonstration merely by declining to print an apparatus. Silence about alternatives can conceal real questions just as effectively as neutral-sounding notation can conceal editorial consequences.

Both sides owe the reader transparency. The critical editor should state why the received reading has been demoted, bracketed, or excluded. The Received Text editor should identify genuine intramural differences and explain why the chosen form is retained. Typography must serve an argued judgment, not substitute for one.

Two Opposite Ways to Beg the Question

The debate commonly moves between two loaded descriptions. The Received Text advocate says:

The critical text omits inspired words.

---

1. Lane G. Keister, ‘An Answer to Chris Thomas’ (November 26, 2024), for the language of tweaking, plusses, and minuses considered here. ↩
2. B. F. Westcott and F. J. A. Hort, The New Testament in the Original Greek (1881), and the introductions to Nestle–Aland editions illustrate how a main text and apparatus necessarily embody editorial judgments. ↩
3. Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament, 4th ed. (Oxford, 2005), for a standard account of witnesses, variants, and editorial method. ↩

The critical-text advocate says:

The Received Text adds uninspired words.

---

1. Lane G. Keister, ‘An Answer to Chris Thomas’ (November 26, 2024), for the language of tweaking, plusses, and minuses considered here. ↩
2. B. F. Westcott and F. J. A. Hort, The New Testament in the Original Greek (1881), and the introductions to Nestle–Aland editions illustrate how a main text and apparatus necessarily embody editorial judgments. ↩
3. Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament, 4th ed. (Oxford, 2005), for a standard account of witnesses, variants, and editorial method. ↩

Each sentence contains the conclusion inside the description. The first assumes the longer reading is original; the second assumes it is secondary. Repetition cannot supply the missing proof. A fair opening description would be:

The Received Text contains these words, while the named critical edition does not.

---

1. Lane G. Keister, ‘An Answer to Chris Thomas’ (November 26, 2024), for the language of tweaking, plusses, and minuses considered here. ↩
2. B. F. Westcott and F. J. A. Hort, The New Testament in the Original Greek (1881), and the introductions to Nestle–Aland editions illustrate how a main text and apparatus necessarily embody editorial judgments. ↩
3. Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament, 4th ed. (Oxford, 2005), for a standard account of witnesses, variants, and editorial method. ↩

That sentence establishes a public fact. The argument may then proceed to manuscript support, versional and patristic evidence, transcriptional explanations, intrinsic considerations, historical reception, confessional doctrine, and the proper relation of those categories. After that work, one may conclude that words were lost in one line or expanded in another.

But there is a third way to beg the question:

The difference is merely a plus or minus.

---

1. Lane G. Keister, ‘An Answer to Chris Thomas’ (November 26, 2024), for the language of tweaking, plusses, and minuses considered here. ↩
2. B. F. Westcott and F. J. A. Hort, The New Testament in the Original Greek (1881), and the introductions to Nestle–Aland editions illustrate how a main text and apparatus necessarily embody editorial judgments. ↩
3. Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament, 4th ed. (Oxford, 2005), for a standard account of witnesses, variants, and editorial method. ↩

The terms themselves need not be prejudicial. The word merely is the problem. Formal description is made to carry a theological verdict it cannot bear. A twelve-verse passage and a movable nu may both be classified as variants, but only at a level of abstraction too broad to guide the church. The existence of a variation unit says nothing by itself about its size, meaning, support, history, or canonical consequence. Responsible writing should therefore distinguish presence and absence before explaining addition and omission. It should not pretend that the distinction makes the outcome unimportant.

Critical Revision Without a Charge of Bad Faith

The phrase critical revision can also be used carefully. It need not mean that an editor knowingly corrupted Scripture. It can mean that an edition produced by critical principles differs from a previously received or previously published text. The revision is an observable relation between textual artifacts. Its rightness remains a separate question.

Similarly, replacement can name an institutional result without claiming a destructive intention. When churches, seminaries, publishers, and translations cease using one Greek base and adopt another, one textual standard has functionally displaced another. The historians may then investigate why. Some actors may have regarded the change as modest correction, others as recovery of a more ancient text, and still others as a major methodological advance. The institutional fact does not reveal every participant’s motive.⁴

This discipline matters because preservationists can weaken a strong case with an unnecessary psychological claim. The most important criticism is not that every critical editor wanted to overthrow the Bible. It is that the critical method locates the church’s access to exact wording differently. The received form no longer supplies the presumptive textual rule; the surviving witnesses supply evidence from which specialists construct the current main text.

That architecture can be described and criticized without judging anyone’s heart. The same discipline should govern criticism of Received Text advocates. One should not infer that a minister refuses evidence, fears scholarship, or worships a translation merely because he affirms the received text. His published arguments may reveal errors or inconsistencies. Those are enough. Motive is not a shortcut to refutation. The goal is not bloodless prose. It is accountable prose: forceful about acts, standards, and consequences; restrained about intentions that have not been demonstrated.

Why the Vocabulary Matters to Preservation

If preservation means only that all relevant evidence survives somewhere in the manuscript aggregate, then plus and minus fit naturally. The inspired wording remains to be selected from available alternatives, and the apparatus displays the field of possibilities.

The confessional claim examined in this series is stronger. The Hebrew and Greek Scriptures were “kept pure in all ages” by God’s singular care and providence. The autographs perished, but the inspired text continued in faithful apographs as the church’s public rule. Copies may contain errors, editions may require ministerial correction, and no printer or editor is inspired. Yet the authentic text is not reduced to a set of rival possibilities awaiting perpetual reconstruction.

From that standpoint, the location of a reading matters. A word in an apparatus may be evidentially preserved while ecclesiastically displaced. A passage under brackets may remain physically visible while its canonical status is denied or suspended. A reading may survive in thousands of facsimiles while no longer governing the translation placed in the pew.

The preservation question is therefore not settled by changing the nouns. Calling a disputed clause a plus does not tell us whether God preserved it as Scripture. Calling its exclusion a tweak does not tell us whether criticism ministered to the church’s text or revised the church’s text under another authority.

Keister is correct that the decisive question is whether the reading is original. Confessional Bibliology adds that the doctrine of preservation governs how the church expects that originality to be historically available and publicly recognizable. The debate concerns both textual evidence and the theological conditions under which evidence is judged.

A Standard for Fair Textual Description

The following protocol can prevent rhetoric from doing the work of argument:

1. Name the texts being compared. Do not say “the TR” or “the CT” when the local claim requires Scrivener, Stephanus, NA28, UBS5, or another specified edition.

1. State the readings before naming their history. Give the actual Greek forms and a sufficiently literal translation. Identify presence, absence, spelling, order, or substitution.

1. Name the reference point. If plus or minus is used, say relative to which text. Reverse the comparison when helpful so the relational character remains visible.

1. Separate formal relation from genealogical explanation. “These words are present here and absent there” is a fact. “A scribe added them” or “a scribe omitted them” is an inference requiring evidence.

1. Describe the modern editorial act. State whether the reading is in the main text, bracketed, footnoted, placed only in the apparatus, or excluded. This can be reported without deciding ancient direction.

1. Describe the ecclesiastical consequence. Will the wording normally be translated, read, preached, and memorized? Availability to a specialist and use by a congregation are not the same condition.

1. Measure the variant honestly. Distinguish orthography, untranslatable order, minor translatable difference, substantial clause, and extended passage. Do not inflate the small or miniaturize the large.

1. Give the opposing argument its strongest form. Do not assume that the critical editor hates Scripture or that the Received Text advocate rejects evidence. Address the stated doctrine and method.

1. Reach a conclusion. Neutral comparison is a stage of inquiry, not its final goal. A church needs a text, and an edition necessarily proposes one.

This standard permits the careful use of plus, minus, addition, omission, retention, exclusion, and revision. No single term must carry every level of analysis. The writer states which claim he is making and supplies the evidence appropriate to it. Fair terminology should postpone conclusions that have not yet been proved. It should not conceal decisions that have already been made.

Notes and Further Reading