Textual Reading, Translation, and Interpretation: Three Different Questions

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.

The English Bible Is Not the Greek or Hebrew Edition

This series uses the English Standard Version to make the wording of modern critical editions accessible to English readers. That use is necessary, but it creates a danger. A reader can begin to treat the ESV as though it were simply the Nestle-Aland or Masoretic text converted into English. Once that assumption is made, every ESV expression is credited to the critical edition, every interpretive choice is treated as a fact of the original language, and every doctrine that can be defended from the English translation is said to have been derived from the Greek or Hebrew main text.

None of those inferences is automatic.

A textual edition and a translation are not the same artifact. A Greek or Hebrew reading answers one question. The English words selected to represent that reading answer another. The meaning assigned to those words answers a third. A doctrinal proposition is then formulated from the interpreted evidence. These stages are closely related, and translation always requires some interpretation, but their overlap does not erase their different objects, evidence, or burdens of proof.

The distinction is decisive for the experiment undertaken in this series. If the ESV renders a reading different from the one printed in NA28 or BHS, then a coherent English proposition may belong to the ESV without belonging to the stated original-language corpus. If the ESV follows the edition exactly but selects one disputed meaning in English, the wording belongs to the edition while the disambiguation belongs to the translation. If both the edition and translation remain ambiguous but a Reformed commentator supplies a confessional interpretation, the doctrine may belong to the commentator’s synthesis rather than to the wording by necessary consequence.

Article 7 required every alleged contradiction to have an identified source. The present article establishes how that source is to be identified. It asks three questions in order:

1. What textual reading is actually operative?
2. How has that reading been represented in English?
3. What does the resulting wording mean in its grammatical, literary, historical, and canonical context?

Only after those questions have been answered may the series ask whether the exact Westminster proposition is expressed or follows by good and necessary consequence.

The governing rule is simple:

A translation is a witness to the translators’ textual and interpretive decisions. It is not a transparent substitute for the edition from which those decisions were made.

This rule neither disparages translation nor requires every Christian to become a textual critic. It merely refuses to make an English sentence prove what a specified Greek or Hebrew text does not say.

Three Questions That Must Not Be Collapsed

The three questions can be stated in a compact form.

These are analytical distinctions, not watertight compartments. Editors interpret evidence when they select a reading. Translators interpret grammar and meaning when they choose English words. Interpreters depend upon both textual and translational decisions. Yet the fact that every stage involves judgment does not turn every judgment into the same kind of judgment.

An editor may correctly select a Greek reading that a translator renders poorly. A translator may accurately render a reading that an interpreter misconstrues. An interpreter may correctly explain an English sentence even though that sentence represents a different textual reading from the one declared as the control. A theologian may state an orthodox doctrine from an interpretation that the passage permits but does not require. Unless the layers remain identifiable, the series cannot tell where a difficulty begins or where a proposed repair occurs.

The distinctions also preserve fair criticism. It would be unjust to accuse NA28 of saying “will be exposed” in 2 Peter 3:10 when NA28 prints a Greek negation. It would be equally unjust to accuse the ESV translators of hiding their textual freedom when their own preface expressly states that they sometimes depart from the preferred readings of the editions they ordinarily use. Accuracy requires attribution before evaluation.

The Confessional Order of Authority

Westminster Confession 1.8 gives translation both a necessary dignity and a subordinate place. The Old Testament in Hebrew and the New Testament in Greek, it says, were immediately inspired and are therefore authentic; in religious controversy the church is finally to appeal to them. The same paragraph then commands translation into the common languages of the nations because the original languages are not known to all God’s people.¹

The order matters. The confession does not oppose the original-language text to the translated Bible as though the latter were spiritually unimportant. Translation is the divinely appointed means by which the Word dwells plentifully among peoples who do not know Hebrew and Greek. Yet the translation is not made the final court of appeal over the words it is intended to render.

Confession 1.9 then distinguishes text from interpretation. Scripture is its own infallible interpreter, and a disputed sense must be searched by comparison with clearer places. The appeal is not merely to an isolated dictionary meaning, still less to the first English gloss that comes to mind. It is to Scripture speaking through its words, grammar, contexts, and canonical relations.

This confessional order agrees with the bibliological distinction already defended in the supplied project materials: Confessional Bibliology does not place the KJV above the Hebrew and Greek but evaluates a translation by its fidelity to the original-language text it claims to represent.² The same standard must govern the ESV. It cannot be treated as final authority merely because it is the English control for this series. If it legitimately conveys the operative Greek or Hebrew, it is useful evidence for the English reader. If it follows another reading, resolves an ambiguity, supplies an implication, or obscures a grammatical feature, that relation must be recorded.

The governing order is therefore not:

ESV wording → assumed original wording → Reformed interpretation → Westminster doctrine.

---

1. Westminster Confession of Faith 1.8–1.9. See the Orthodox Presbyterian Church’s Confession of Faith. ↩
2. Chris Thomas, “Confessional Bibliology vs. KJV-Onlyism,” October 24, 2024, supplied PDF, 1. The article locates final authority in the Hebrew and Greek rather than in the KJV and describes translations as faithful insofar as they accurately represent those originals. ↩

It is:

identified Hebrew or Greek reading → tested English representation → contextual interpretation → doctrinal inference.

---

1. Westminster Confession of Faith 1.8–1.9. See the Orthodox Presbyterian Church’s Confession of Faith. ↩
2. Chris Thomas, “Confessional Bibliology vs. KJV-Onlyism,” October 24, 2024, supplied PDF, 1. The article locates final authority in the Hebrew and Greek rather than in the KJV and describes translations as faithful insofar as they accurately represent those originals. ↩

That order does not guarantee agreement with any predetermined textual position. It can expose failures in a critical edition, a received-text translation, a confessional interpreter, or the critic’s own argument. Its value is precisely that it allows no participant to borrow the authority of a different layer without disclosure.

Five Relationships between an Edition and a Translation

The comparison will ordinarily yield one of five relationships. These are descriptive categories rather than final grades.

Direct correspondence

The translation represents all material features of the operative reading and does not prematurely eliminate a doctrinally relevant ambiguity. Normal grammatical adjustments do not defeat direct correspondence. Greek participles may become English finite verbs; Hebrew word order may change; an idiom may receive a natural equivalent.

Direct correspondence establishes that the English clause can be used as an accessible representation of the declared text. It does not yet establish that the interpreter’s doctrine follows.

Legitimate formal adjustment

The translation changes surface form to preserve meaning across languages. This category recognizes that formal identity can produce semantic distortion. An article may need to be omitted, a pronoun made explicit, or a repeated conjunction varied. The adjustment must be explained if it affects a doctrinal argument.

Interpretive selection

The translation follows the same textual reading but selects one meaning, syntactic relation, referent, or discourse force among alternatives. The choice may be highly probable or even demonstrably correct. It remains an interpretive selection, and the dossier must test the excluded possibilities before claiming necessary consequence.

Textual departure

The English rendering presupposes a materially different Hebrew or Greek reading from the declared main text. The departure may follow an apparatus reading, another edition, an ancient version, a conjectural emendation, or a translator’s judgment. This category does not decide whether the departure is good or bad. It ensures that the resulting proposition is attributed to the reading that actually generates it.

Apparatus or footnote reporting

The translation’s main text follows one option while a footnote reports another. The note is part of the published translation artifact and can alert the reader to uncertainty or alternatives. It does not give the main text two simultaneous propositions. A doctrine that depends upon the note must be identified as depending upon the noted reading.

These relationships prevent a recurring equivocation. “The ESV contains the words” can mean that the words stand in the main text, that they appear only in a footnote, that they paraphrase an implication, or that they translate a different source reading. Those are four different evidentiary facts.

Control Pattern One: Matthew 5:22

Matthew 5:22 is an example in which the current ESV substantially tracks the shorter NA28 main reading.

NA28 prints:

πᾶς ὁ ὀργιζόμενος τῷ ἀδελφῷ αὐτοῦ ἔνοχος ἔσται τῇ κρίσει

---

1. Westminster Confession of Faith 1.8–1.9. See the Orthodox Presbyterian Church’s Confession of Faith. ↩
2. Chris Thomas, “Confessional Bibliology vs. KJV-Onlyism,” October 24, 2024, supplied PDF, 1. The article locates final authority in the Hebrew and Greek rather than in the KJV and describes translations as faithful insofar as they accurately represent those originals. ↩

A close diagnostic gloss is:

“Everyone who is angry with his brother will be liable to the judgment.”

---

1. Westminster Confession of Faith 1.8–1.9. See the Orthodox Presbyterian Church’s Confession of Faith. ↩
2. Chris Thomas, “Confessional Bibliology vs. KJV-Onlyism,” October 24, 2024, supplied PDF, 1. The article locates final authority in the Hebrew and Greek rather than in the KJV and describes translations as faithful insofar as they accurately represent those originals. ↩

The ESV likewise reads, “everyone who is angry with his brother will be liable to judgment.” The received Greek comparison includes εἰκῇ, commonly represented as “without cause.”³

At the first two layers the record is straightforward:

The third question remains open. Does the shorter clause assert that every instance of anger, without distinction of object, cause, moral quality, duration, or manner, is sinful? The absence of an explicit adverb does not answer that interpretive question by itself. The participle occurs inside a discourse about murder, contempt, alienation, liability, and reconciliation. The wider canon records divine anger and describes Jesus looking upon opponents “with anger” in Mark 3:5. The interpreter must determine whether Matthew condemns anger absolutely or anger of the murderous and contemptuous moral species developed in the context.

The point here is not to settle that exegesis. It is to show why the layers matter.

The textual difference is real: NA28 lacks a locally explicit qualifier present in the received comparison. The ESV makes that difference visible. Yet the proposition “all anger without exception is sinful” is an interpretation of the shorter wording, not another name for the wording itself. If that interpretation is used to create a contradiction with Christ’s anger, the critic must establish its scope. If a defender appeals to Mark 3:5 to limit the condemned anger, the defender must show that the distinction arises from the canonical evidence rather than merely from the need to protect Christ’s sinlessness.

Matthew 5:22 therefore displays textual correspondence with an unresolved interpretive question. It must not be mislabeled either as a translation departure or as an automatic contradiction.

Main Text, Footnote, and Apparatus Are Not Interchangeable

Modern editions and translations often present several kinds of information on the same page. Visual proximity can create a false equivalence among them.

The edition main text is the wording the editors print in the continuous reading line. In an eclectic New Testament edition it ordinarily represents the editors’ preferred reconstruction. In a diplomatic Hebrew edition it represents the base manuscript according to the edition’s stated policy.

The critical apparatus reports selected variants, witnesses, and editorial information. Its presence means that the edition documents alternatives; it does not mean the main text asserts every alternative or that all readings have equal probability.

The translation main text is the wording intended for continuous public and private reading. It is the proposition most English readers will naturally attribute to Scripture.

The translation footnote may offer a literal gloss, an alternate rendering, a different textual reading, a manuscript notice, or an explanation. The ESV rightly calls such notes integral to its work. Yet a note beginning “some manuscripts” performs a different function from the main clause. It reports rather than adopts the alternate reading.

The commentary or study note belongs to another layer again. It may defend the translation, explain doctrine, harmonize passages, or cite evidence absent from the page. It cannot be treated as wording contained in either the original-language main text or the translation.

The following attribution rules will govern this series:

1. A doctrine supported only by an apparatus reading will be credited to that reading, not to the edition’s main text.
2. A doctrine stated only in an English footnote will not be counted as directly expressed by the translation’s main line.
3. A footnote can establish that readers were alerted to an alternative; it cannot make the main text semantically contain the alternative.
4. A translation that follows an apparatus reading may be textually defensible, but its proposition belongs to the adopted alternate reading.
5. A commentator’s harmonization must disclose every premise and cannot be credited to the verse merely because it appears beneath it.

These rules are especially important for familiar passages such as Matthew 6:13 and 1 John 5:7–8. A longer wording can remain visible in a note while being absent from the continuous critical text. Visibility is not main-text inclusion, and liturgical familiarity is not textual adoption.

Layer-Specific Error and Repair

A difficulty must be corrected at the layer where it occurs. Otherwise the proposed solution can conceal the original failure.

Several combinations are possible.

A translator may correct what he judges to be a defective edition reading by following another witness. The resulting English may be true and textually superior, but the coherence belongs to the alternate reading. For the declared main-text experiment, this is a textual repair.

A translator may follow the same reading while making its best contextual meaning explicit. If the grammar and canon warrant the choice, the translation can rightly clarify the text. The dossier should still record the choice so that the doctrine’s dependence upon it is visible.

An interpreter may resolve an apparent contradiction by distinguishing time, respect, moral species, or referent. If the distinction is textually warranted, the resolution belongs to interpretation and may qualify for C1. If the distinction is imported only because the system requires it, coherence may be repair-dependent or the doctrine may remain underdetermined.

A doctrinal system may survive the loss of one local proof because other passages supply the premises. That is proof migration, not textual restoration. The missing words do not reappear simply because the doctrine remains true.

The series will therefore ask of every repair: Who supplied it, at what layer, and from what evidence?

Strong Objections

“All translation is interpretation, so the distinction is artificial”

All translation involves interpretation, but it does not follow that textual selection, translation, and interpretation are identical. A translator deciding whether οὐχ is present answers a different question from a translator deciding how to render the present participle, and both answer a different question from a theologian deciding how the proposition bears on eschatology.

The distinction is analytical in the same way that grammar and exegesis are analytically distinct though inseparable in practice. It enables responsibility to be assigned to the decision that produced the result.

“Ordinary readers cannot check Greek and Hebrew”

Westminster 1.8 anticipated that fact and required translation. Ordinary readers can know Scripture truly through faithful translations without reproducing the entire scholarly process. The present series, however, makes technical claims about what particular critical editions can sustain. Such claims require checking the editions.

The burden belongs chiefly to teachers, translators, ministers, and scholars who invoke the original text—not to every believer reading at home. A minister should not tell an English reader that “the Greek says” what he has inferred only from an English translation.

“The ESV footnotes already disclose the alternatives”

Often they do, and that disclosure is valuable. It establishes that the translation is more transparent than a bare quotation of the main line. It does not erase the distinction between the adopted wording and the reported alternative. Nor does every note provide the evidence necessary to adjudicate the variant.

“The ESV translators may be right to correct their base edition”

Certainly. Psalm 22:16 may represent precisely such a judgment. The present method does not assume that the edition is right and the translation wrong. It requires the correction to be named. If the alternate reading is original, the later textual dossier should defend it. What may not happen is crediting the BHS main line with a piercing verb it does not print.

“A translation should communicate meaning, not preserve ambiguity”

A translation sometimes must choose. The choice can be responsible and useful. But when a doctrinal argument depends upon the choice, the underlying ambiguity and reasons for resolution must be examined. The ordinary reader needs a readable Bible; the present experiment needs an auditable inference.

“The received text and the KJV have the same translation problems”

Yes. The three questions apply without favoritism. The KJV may select a sense, supply English words, or follow a reading that must be identified. Confessional Bibliology is weakened, not strengthened, if it demands layer precision from modern versions while treating a received-text translation as though it were the inspired Greek and Hebrew.

“If the final doctrine is orthodox, the route does not matter”

The route matters because this series tests derivation. A true conclusion can be reached through an invalid argument or borrowed from a source not admitted into the experiment. Westminster 1.6 claims not merely that its doctrine is compatible with Scripture but that the whole counsel is expressly set down or follows by good and necessary consequence. Provenance is part of that claim.

“Canonical interpretation allows theology to clarify every difficult verse”

Canonical interpretation is necessary, but it has rules. A clearer passage can determine referent, scope, moral category, or doctrinal relation. It cannot make one textual reading contain another. Nor can the interpreter use the completed confession as an unstated premise and then claim to have reconstructed that confession from the tested text.

“These controls make theological construction impossible”

They make undocumented construction impossible. Systematic theology legitimately gathers passages, states implications, distinguishes senses, and draws consequences. The controls simply require each premise to be visible and each inferential step to be warranted. A doctrine that truly arises from the corpus will survive that demand.

Verdict

Textual reading, translation, and interpretation are three different questions because they have different objects and can yield different answers at the same verse.

The textual question asks which Hebrew or Greek words belong to the declared corpus. It distinguishes main text from apparatus, eclectic selection from diplomatic presentation, and a fixed edition from the wider manuscript tradition. It must be answered before an English proposition can be attributed to NA28, UBS6, BHS, BHQ, or a received comparison.

The translation question asks what source reading the English represents and how it represents it. The ESV is an exceptionally useful reader-facing witness because of its essentially literal aim and its textual notes. Yet its own preface acknowledges departures from its ordinary bases. It is therefore neither NA28 nor BHS in English, and its 2025 text cannot be assumed to reproduce the later UBS6 main text.

The interpretive question asks what the wording means in its grammatical, literary, historical, and canonical context. A translation may embody an interpretation, but the printed choice does not make every excluded alternative impossible. Doctrinal compatibility remains weaker than good and necessary consequence.

The control passages establish four distinct relations. Matthew 5:22 shows the ESV following the shorter NA28 reading while leaving the moral scope of anger to interpretation. Second Peter 3:10 shows a positive ESV predicate that does not transparently represent NA28’s negative clause. Psalm 22:16 shows the ESV following an alternate textual judgment rather than directly translating the BHS main line. John 1:18 shows the ESV changing its English formulation while NA28’s μονογενὴς θεός remains unchanged.

These facts do not predetermine which reading, translation, or interpretation is correct. They determine who owns each proposition. A coherent ESV rendering cannot be credited automatically to a critical Greek or Hebrew main text. An orthodox footnote cannot be counted as the main text. A Reformed interpretation cannot be described as a necessary consequence until its premises are recovered from the declared corpus. And a contradiction cannot be manufactured by combining propositions from different layers without disclosure.

The series will therefore proceed in a fixed order: establish the operative reading, test the English representation, interpret the resulting wording, trace the doctrinal inference, and record the provenance of every repair. Only then can it answer whether Reformed theology has truly been reconstructed from the critical corpus or merely carried into it.

Safeguards 15–19 add a source-artifact requirement to that order. Every quotation or transcription used in an audit must identify whether the researcher inspected a facsimile, scan, OCR product, diplomatic transcription, corrected text, modernized edition, translation, or excerpt; whether the source was directly accessed; and whether the quoted wording was verified against an image or authoritative text. Historical meaning is established before a modern interpretation is tested, and material corrections remain visible in the revision history.

Notes and Further Reading