How the Discipline Works Now

For ten sessions we have used a method without emphasizing its name. We considered manuscript age and relationship, ancient versions, patristic evidence, copying habits, authorial style, and the reading that best explains its rivals. That approach is commonly called reasoned eclecticism.

This session names it, compares it with a more internally focused method, and introduces the computer-assisted Coherence-Based Genealogical Method. The methods disagree at the margins. They do not produce radically different New Testaments.

The continued debate is a sign of a discipline testing its assumptions in public.

1. Reasoned eclecticism

Eclectic means an editor does not follow one manuscript or textual family at every passage. The reading is judged locally. An Alexandrian reading may prevail at one verse, a Byzantine reading at another, and a reading preserved chiefly in versions at a third.

Reasoned means the decision rests on arguments other readers can inspect:

• which witnesses support each reading;
• their dates and relationships;
• whether several witnesses are independent;
• which change a copyist is more likely to make;
• which form fits the author’s vocabulary, style, and argument;
• which reading best explains the origin of the alternatives.

Bruce Metzger’s Textual Commentary made the committee’s reasoning accessible at many important variants. Kurt and Barbara Aland and the Institute for New Testament Textual Research helped build the manuscript infrastructure behind modern hand editions.

Reasoned eclecticism is the broad method behind the Nestle-Aland and United Bible Societies texts and therefore behind many modern translations. Committees and translators can still differ at individual verses while working from the same general approach.

2. Work the method: 1 Thessalonians 2:7

The difference is one Greek letter.

Ancient Greek manuscripts commonly used continuous writing. The preceding word ends with n, so the boundary could appear as either one n or two. A copyist could easily drop or repeat the letter.

External evidence

The Vulgate reads parvuli, “little ones,” and therefore supports “infants,” not “gentle.” Codex Bezae should not appear in a Pauline witness list; the Pauline manuscript designated D is Claromontanus.

Internal evidence

“We became infants among you, like a nursing mother caring for her own children” creates a jarring shift. Paul is first the child and immediately the mother. “Gentle” makes a smooth, single image.

That roughness favors “infants.” A scribe can easily turn the mixed metaphor into the expected adjective. Paul also shifts family imagery freely in this chapter: infant, nursing mother, and father. “Infants” fits his rhetoric even though it initially surprises the reader.

The decision

Early and diverse witnesses, the harder reading, and Paul’s style converge on “infants” or “little children.” Current critical editions print that form. Some translations retain “gentle,” sometimes with a note, because the single-letter confusion is ancient and the smoother reading remains well represented.

The Ad Fontes NT comparison currently displays older or edition-specific texts that all read “gentle”; its publisher-note panel records the “little children” alternative associated with later critical editions. The tool’s limits are part of reading it responsibly.

3. Confidence ratings are not footnote ratings

UBS editions historically assigned letter grades to selected decisions:

These grades belong to particular editions and committee decisions; they can change when a text is re-edited. They do not measure a verse’s theological importance.

A translation footnote does not imply C or D uncertainty. The omissions of Mark 16:9–20, John 7:53–8:11, and Acts 8:37 received high-confidence judgments even though English Bibles give them prominent notes. A footnote tells the reader that the evidence or reception history is worth seeing.

4. Thoroughgoing eclecticism

G. D. Kilpatrick and J. Keith Elliott are associated with an approach often called thoroughgoing eclecticism. It gives unusually strong weight to internal evidence.

Its advocates emphasize several real problems:

• manuscripts are mixed rather than pure representatives of tidy families;
• a witness can align with different clusters in different books or passages;
• early scribes could polish grammar as well as preserve it;
• authorial vocabulary and style are genuine historical evidence.

The approach asks which reading the author is most likely to have written, even when external support is limited.

What it gets right

Family labels are approximations. Manuscripts do not vote as permanent parties. Internal evidence cannot be reduced to a tie-breaker used only after counting witnesses.

Where the danger lies

Arguments from style can become elastic. An editor may call a reading original because it sounds like Paul, or secondary because it sounds too much like Paul. The same feature can be explained in opposite ways.

External evidence restrains editorial taste. A reading with no witness requires an unusually strong explanation. Most editors therefore keep external and internal evidence together rather than letting a reconstructed authorial style control the decision.

Claims about Elliott’s publications should remain modest and verifiable. His contribution lies in his textual work and defense of thoroughgoing eclecticism; unrelated lexicon or synopsis attributions are unnecessary here.

5. The CBGM without the mystique

The Coherence-Based Genealogical Method, developed in Münster under Gerd Mink and collaborators, helps editors analyze large sets of textual relationships. It is used in the Editio Critica Maior and has influenced current hand-edition texts.

The CBGM does not discover a physical family tree in which one surviving manuscript is literally copied from another. It models relationships among textual states and readings.

In simplified form:

1. Editors begin with a provisional decision at each place of variation.
2. Those local decisions help define which reading is prior to another at each variant.
3. Software compares witnesses across many variants and identifies possible genealogical relationships between their texts.
4. Editors examine whether a proposed reading coheres with that network.
5. When a decision changes, the relationships are recalculated and reviewed.

The process is iterative. It combines editorial judgment and computational comparison.

What the method contributes

• It takes mixture seriously: one manuscript need not belong permanently to one family.
• It can analyze far more relationships than a person can hold in memory.
• It exposes where local decisions fit—or strain—the wider textual network.
• It makes assumptions and data available for scholarly criticism.

What it cannot do

• It cannot identify the autographs by computation alone.
• It does not date a reading merely by assigning it a strong relationship.
• It depends partly on editorial judgments used to establish local priority.
• It cannot replace transcriptional probability, authorial style, manuscript date, or versional evidence.

The computer does not vote on inspiration. It helps scholars organize evidence. The CBGM is a tool, not an oracle.

6. Jude 5: who saved Israel from Egypt?

Jude reminds his readers of the exodus and subsequent judgment. The subject varies.

Alexandrinus belongs with “Jesus,” not “the Lord.” P72’s unusual form is “God Christ.”

Earlier Nestle-Aland editions printed “the Lord,” and the committee regarded the decision as exceptionally difficult; the historical UBS rating was D, not B. Metzger himself preferred “Jesus.”

Work on the Catholic Letters for the ECM, including CBGM analysis, contributed to the reassessment reflected in NA28 and later critical hand-edition text: Jesus.

Why “Jesus” has a strong claim

It is the harder reading. A copyist encountering Jesus as the one who saved Israel from Egypt could replace the name with the expected “Lord” or “God.” The reverse change is less natural unless a scribe deliberately sought a more explicit Christological statement.

The external evidence is strong but divided. The CBGM did not manufacture a new reading; “Jesus” stands in major ancient witnesses. The method helped editors reconsider how those witnesses and readings cohere.

Why the reading matters

Jude identifies the preincarnate Son with the Lord who delivered Israel. Paul similarly writes that the rock accompanying Israel was Christ. The reading belongs to the apostolic pattern of reading Israel’s Scriptures in relation to the Son.

One changed word did not reinvent Christian doctrine. It refined a close textual decision with public reasons.

7. Three methods, one broadly stable text

All three are eclectic. None follows one manuscript family everywhere. Their disagreements affect a relatively small portion of the New Testament and rarely touch a doctrinal formulation.

Methods improve one another by criticism. Thoroughgoing eclecticism exposed the weakness of rigid family labels. The CBGM offers tools for analyzing mixture. Traditional internal analysis reminds CBGM users that coherence is not the same as originality.

A discipline that rechecks its assumptions is not necessarily unstable. It may be doing the patient work its evidence requires.

Practice

Inspect the 1 Thessalonians 2:7 note and Jude 5 comparison.

1. Why does the app show only “gentle” in its displayed editions while noting “little children” separately?
2. Which side at 1 Thessalonians 2:7 has earlier versional support?
3. At Jude 5, which reading is hardest and why?
4. Explain the CBGM without using “the computer chooses the text.”
5. Name one evidence type the CBGM organizes and one it cannot replace.

Discussion

1. Why is internal evidence both necessary and dangerous?
2. Is “gentle” a defensible translation choice if an edition’s Greek text reads “infants”?
3. Does a changed critical decision at Jude 5 unsettle or reassure you?
4. What would responsible transparency about a method’s assumptions look like?
5. How can several methods disagree while the text remains broadly stable?

Working vocabulary

Reasoned eclecticism. Establishing each reading through stated use of external and internal evidence without permanent loyalty to one textual family.

Thoroughgoing eclecticism. An approach that gives exceptional weight to internal evidence and authorial style.

CBGM. The Coherence-Based Genealogical Method, a computer-assisted framework for studying relationships among textual states and readings.

Coherence. The degree to which a proposed relationship at one variant fits relationships observed across many variants.

Initial text / Ausgangstext. The reconstructed earliest recoverable form targeted by an edition; not a claim to possess the physical autograph.

Scriptio continua. Ancient writing without spaces between words.

Before the final session

Read: Chapter 15, “What We Have, What We Don’t, and Why It’s Enough.”

Do one thing: Choose the passage from this course that has stayed with you most. Read it again, footnote included, and write one sentence about how you hear it differently now.

Bring: Your Bible, your sentence, your vocabulary pages, and the answer you gave in Session 1 about where your confidence rests.

Think about: Has the ground of your confidence changed, or has the course helped you describe that ground more accurately?

A closing word

Thanks be to God for scholars who compare, question, calculate, and revise in public. Their methods are tools used by fallible people, not new sources of revelation. Through them the church continues the ordinary work of listening carefully to the witnesses it has received—and hears again the Lord who saved his people and remains the same forever.

“Jesus Christ is the same yesterday and today and forever!”

— Hebrews 13:8 (NET)

Scripture quotations in this session are from the NET Bible, with the KJV referenced where the lesson compares translations. Confessional citations follow the Kolb-Wengert edition of the Book of Concord.