Textual Criticism of the Hebrew Bible witnesses, apparatus & the preservation of Scripture

Why Textual Criticism Matters — And Why It Does Not Threaten Scripture

"The grass withers, the flower fades, but the word of our God will stand forever." — Isaiah 40:8

1.1 What Textual Criticism Is

Textual criticism is the discipline that compares the surviving manuscript witnesses of an ancient text to recover, as nearly as the evidence allows, the wording the original author wrote. It applies to every text from antiquity for which we lack the autograph: Homer, Plato, Caesar, Livy, Josephus, the New Testament, and the Old Testament. It is not a sceptical project; it is a humble one. The textual critic asks: given that we do not possess the autograph and that the manuscript witnesses we possess sometimes differ, how do we reconstruct what the author originally wrote?

For the New Testament we have something like 5,800 Greek manuscripts, plus tens of thousands of versional and patristic witnesses, all dating to within centuries of the apostolic period. For the Old Testament the situation is different. The autographs of Moses, Samuel, Isaiah, and the rest were written between roughly 1400 BC and 400 BC. Until the discovery of the Dead Sea Scrolls in 1947, our oldest substantial Hebrew manuscripts were the great Masoretic codices of the tenth and eleventh centuries AD — separated from the autographs by between fifteen hundred and two thousand four hundred years. The Dead Sea Scrolls have now pushed the documentary chain back by roughly a thousand years for many books. We also have important early witnesses in translation: the Greek Septuagint (third century BC onward), the Samaritan Pentateuch (preserved in a parallel tradition since at least the second century BC), the Aramaic Targums (first century BC onward), the Syriac Peshitta (second century AD onward), and the Latin Vulgate (fourth century AD).

1.2 The Shape of the Data

The most important single fact about the textual data is this: the mass of OT witnesses agree on the great mass of the text. Tov estimates that something like 90–95 percent of the consonantal Hebrew text is identical across all major witnesses; the variants concentrate in a small percentage of cases, and the great majority of those are themselves minor — differences in spelling, word order, the presence or absence of a conjunction, the use of a synonym. Genuinely substantive textual variants — places where a clause or sentence differs in a way that materially affects meaning — exist, but they are rare. And in the small number of places where the substantive variants matter, the careful collation of witnesses gives us a high degree of confidence about which reading is original.

Compare this with the textual situation of other ancient works. For Tacitus's Annals, we have a single eleventh-century manuscript for the first half and a single eleventh-century manuscript (now lost, copied) for the second. For Caesar's Gallic Wars, the earliest manuscript is from the ninth century, and there are perhaps a dozen significant copies. For most of classical Greek and Roman literature the textual evidence is comparatively thin, late, and uncontested only because no one questions a Stoic philosopher the way they question Moses. The OT, by these standards, is extraordinarily well attested.

1.3 Why It Does Not Threaten Scripture

Three observations matter for the Christian reader who fears that textual criticism is a hostile discipline.

First, no major doctrine of the Christian faith depends on a disputed OT textual variant. The Trinity, the deity of Christ, the atonement, justification by faith, the resurrection, the sacraments, the second coming — none of these doctrines is taught only by a verse with a doubtful text. They are taught across the canon, with massive cross-reference, in passages whose text is secure. A Christian could read the OT in the Masoretic Text, or in the Septuagint, or in the Dead Sea Scrolls 4QSam-a, and end up with the same Christian theology.

Second, the doctrine of inspiration applies to the autograph; the doctrine of providential preservation applies to the textual transmission. The Reformed confessions distinguish these carefully (WCF 1.8). Inspiration produced an inerrant Hebrew and Greek original. Providential preservation has kept that original substantially intact across the manuscript tradition. Neither doctrine requires that any particular copy be perfect; both doctrines together require that the church across the ages has had ready access to the substantial content of God's Word, and that is precisely what we observe.

Third, the alternative — abandoning the science of textual criticism altogether — is not piety; it is wilful ignorance. The fact that variant readings exist is a fact, whether we look at them or not. To bury our head in the sand and pretend that the Hebrew text we use in seminary or read in our English Bibles arrived from heaven without a manuscript history is to abandon the very confidence the science actually warrants. The Reformed tradition has always insisted on the careful study of the original languages and the manuscript tradition (cf. WCF 1.8). Calvin, Beza, the Westminster divines, and Owen all engaged the textual questions of their day with rigour. The contemporary Reformed Bible school student should do the same.

The Masoretic Text — Origin, Development, Reliability

"They [the Jews] were entrusted with the oracles of God." — Romans 3:2

2.1 The Name and Period

The "Masoretic Text" is the standardised Hebrew text of the Old Testament produced by Jewish scribes known as the Masoretes (from the Hebrew m-s-r, "to hand on, transmit"). The Masoretic period stretches roughly from AD 600 to 1000, with its great codices clustering around the late ninth and early tenth centuries. The Masoretes inherited a consonantal Hebrew text that had already been remarkably stable for at least a millennium; their distinctive contribution was to fix that text with extraordinary precision, add a system of vowel points (which classical Hebrew, like other Semitic languages, did not write), develop a system of cantillation accents indicating how the text was to be chanted in the synagogue, and surround the text with a marginal apparatus (the masora) noting unusual readings, statistical observations, and cross-references.

The Masoretes were heirs to a much longer Jewish scribal tradition. The forerunners of the Masoretes — the Sopherim ("scribes"), active from roughly the fifth century BC through the second century AD — had already developed strict rules for the copying of Torah scrolls. The Talmudic and post-Talmudic tradition codifies these rules in fascinating detail: scribes were to count letters, words, and verses for each book; were to leave a specific number of blank spaces between sections; were to wash before writing the divine name; were to destroy any scroll containing more than three errors per column. The Masoretes added to this inheritance a comparable scrupulousness for the vowels and accents.

2.2 Ben Asher and Ben Naftali

Two great families dominated the late Masoretic period in Tiberias (Galilee), the centre of Hebrew textual scholarship in this era. The Ben Asher family, working across five generations from roughly AD 780 to 930, produced the textual tradition that came to be recognised as authoritative throughout most of the Jewish world. The Ben Naftali family produced a parallel tradition with minor differences (chiefly in vocalisation and accentuation; the consonantal text is essentially identical). By the high Middle Ages the Ben Asher tradition had prevailed; Maimonides in the twelfth century explicitly endorsed Aaron ben Asher's text as authoritative, and the Ben Naftali tradition gradually receded into a few scattered manuscripts.

The patriarch of the Tiberian school was Aaron ben Moses ben Asher (d. c. AD 960), the fifth generation of the family. The two great codices of the Masoretic period are both attributed to him or to his school.

2.3 The Aleppo Codex

The Aleppo Codex (Heb. Keter Aram Tzova, "the Crown of Aleppo"), produced about AD 925, is the codex Aaron ben Asher himself is said to have vowelled and added the masora to. The consonantal text was written by Shlomo ben Buya'a slightly earlier; Aaron ben Asher's contribution was the vocalisation and the masoretic apparatus. Maimonides in the twelfth century travelled to Aleppo to consult it and pronounced it authoritative. The codex was held in the Aleppo synagogue for centuries.

In 1947, in the wake of the UN partition vote that created the State of Israel, anti-Jewish riots in Aleppo set the synagogue on fire and damaged the codex. Roughly forty percent of the codex — including most of the Torah — was lost or destroyed in the chaos that followed. The surviving portion (about 60 percent, including most of the Prophets and Writings) was smuggled to Israel and is now held at the Shrine of the Book in Jerusalem. The Aleppo Codex is the textual base of two modern critical editions: the Hebrew University Bible (HUB) and the Mikra'ot Gedolot Haketer.

2.4 The Leningrad Codex

The Leningrad Codex (technically Codex Leningradensis, abbreviated L in apparatus notation; also called the St Petersburg Codex), dated to AD 1008–1009, is the oldest complete manuscript of the entire Hebrew Bible. Its colophon states that it was copied by Samuel ben Jacob in Cairo, corrected against a Ben Asher manuscript, and intended for a patron in Damascus. It was acquired by the Russian Imperial Library (now the National Library of Russia in St Petersburg) in the nineteenth century and remains there today.

Because the Aleppo Codex was partially damaged in 1947, the Leningrad Codex has become the de facto standard manuscript for modern critical editions. It is the textual base of:

• Biblia Hebraica (BHK) — Rudolf Kittel's third edition (1937), the first to use the Leningrad Codex.
• Biblia Hebraica Stuttgartensia (BHS) — 1977, the standard scholarly Hebrew Bible from the 1980s through the early 2000s.
• Biblia Hebraica Quinta (BHQ) — in progress since 2004, the successor to BHS.
• The Westminster Leningrad Codex — the digital scholarly edition used by most Bible software (Logos, Accordance, BibleWorks).

The Leningrad Codex is thus the manuscript behind virtually every modern English translation of the Old Testament made in the last fifty years.

2.5 The Stability of the Masoretic Tradition

How accurately did the Masoretes transmit the text they inherited? The discovery of the Dead Sea Scrolls in 1947 made it possible to answer this question with documentary evidence for the first time. Among the scrolls is a complete copy of Isaiah (1QIsa-a, the Great Isaiah Scroll, c. 125 BC), some thousand years older than the Aleppo Codex. When scholars compared 1QIsa-a with the Masoretic Text of Isaiah, they found astonishing stability. The two texts agree at a rate of roughly 95 percent at the level of individual words; the disagreements are overwhelmingly minor (spelling variations, the presence or absence of a conjunction, a single letter variation in a personal name). Across a thousand years of transmission, the Hebrew text of Isaiah barely moved.

This is the empirical vindication of the Masoretic tradition. The Sopherim and the Masoretes inherited a text that was already remarkably stable; they transmitted it for another millennium with comparable scrupulousness; the result is a text whose consonantal stability is documented across nearly twenty-five hundred years.

לֹֽא־יָמ֡וּשׁ סֵפֶר֩ הַתּוֹרָ֨ה הַזֶּ֜ה מִפִּ֗יךָ "This book of the Law shall not depart from your mouth, but you shall meditate on it day and night, that you may be careful to do according to all that is written in it." Joshua 1:8 — the charter text of Israelite scribal piety

The Masoretes and Their Apparatus

"Forever, O LORD, your word is firmly fixed in the heavens." — Psalm 119:89

3.1 The Three Layers Added to the Consonantal Text

The text inherited by the Masoretes from the Sopherim consisted of consonants only — letters with no vowels, no punctuation, no chapter or verse divisions. This is how ancient Hebrew was written; vowels were supplied by the reader from oral tradition. By the early medieval period, with Hebrew no longer a vernacular and the diaspora widening, the risk of losing the oral tradition of vocalisation became acute. The Masoretes responded by developing three concurrent systems of notation:

1. Vowel points (Heb. niqqud) — small marks placed above, below, or within consonants to indicate their accompanying vowel. The Tiberian system, perfected by the Ben Asher school, uses seven principal vowel signs plus the shewa for half-vowels and silent breaks. Tiberian pointing has become the universal standard; older Palestinian and Babylonian systems survive only in scattered manuscripts.
2. Cantillation accents (Heb. te'amim) — marks indicating both the musical chant of the synagogue and, more importantly for the reader, the syntactic phrasing of each verse. The accent system is binary: every word receives either a disjunctive (separating it from what follows) or a conjunctive (binding it to what follows) accent, with the disjunctives ranked into four classes (emperors, kings, dukes, counts). The result is a precise, intricate system of clause structure annotation embedded in every verse of the Hebrew Bible.
3. The masora — the body of marginal notes. This is what gives the Masoretes their name. The masora is organised in two layers (described in 3.3 below).

3.2 The Significance of the Pointing

It is sometimes claimed that the Masoretic vowel pointing is "late" and therefore unreliable. The claim is half true and seriously misleading. The pointing as a written notation is indeed late (seventh to tenth century AD). But the traditional pronunciation the pointing encodes is far older — it represents the oral reading tradition of the synagogue, which had been transmitted continuously alongside the consonantal text for centuries. The Masoretes did not invent a pronunciation; they codified one that they had received.

This matters because in a small but non-trivial number of cases, the pointing yields a different meaning than other plausible pointings of the same consonants would yield. For example, the consonants d-b-r (דבר) can be vocalised as davar ("word"), dever ("plague"), dibber ("he spoke"), or dabber ("speak!"). Context normally settles the matter, but in some cases the Masoretic pointing is itself an interpretive commitment. The Reformed tradition has generally followed the Masoretic vocalisation as the inherited traditional reading, while recognising that the consonantal text is primary and that the pointing is a reliable but secondary witness to original meaning.

3.3 Masora Magna and Masora Parva

The masora proper — the marginal scribal apparatus — comes in two layers, distinguished by where it appears on the manuscript page:

• Masora Parva (Mp; the "small masora") — brief notes in the side margins next to specific words. The Mp typically gives the number of occurrences of a particular form in the OT, flags unusual readings, and notes when a word is unique (hapax legomenon) or when a particular spelling is exceptional. Example: a small note "ב" (Hebrew letter bet, numerical value 2) in the margin means "this form occurs 2 times in the OT."
• Masora Magna (Mm; the "great masora") — longer notes at the top and bottom of the page (and sometimes between columns). The Mm expands on the Mp by giving the actual references where a flagged form occurs, by listing all members of a particular linguistic category, and by recording lists of similar items across the canon.

To these two are sometimes added a third: the Masora Finalis (Mf; the "final masora") — listings collected at the end of each book or at the end of the whole Bible, including chapter and verse counts, halfway points, and other large-scale statistics.

3.4 Qere and Ketiv — The Living Footnotes

The most theologically significant feature of the Masoretic apparatus is the Qere/Ketiv system. Ketiv (Heb. "what is written") refers to the consonantal text as it stands on the page. Qere (Heb. "what is read") refers to a marginal note indicating that the reader should pronounce something different from what is written. The Masoretes preserved the consonants of the inherited tradition without alteration but used the margin to record alternative readings — sometimes corrections of obvious copyist errors, sometimes pious euphemisms, sometimes alternative spellings.

The most famous instance is the divine name YHWH (יהוה). The ketiv is the four-letter Tetragrammaton; the qere is Adonai ("Lord") or, in certain contexts, Elohim ("God"). The Masoretes wrote the consonants of YHWH but pointed them with the vowels of Adonai, signalling to the reader to pronounce the substituted word out of reverence. (The hybrid form "Jehovah" in older English Bibles is a misreading of the YHWH consonants with the Adonai vowels.) The Masoretes' restraint here is important: they refused to alter the consonantal text even when their piety required a different pronunciation; they used the margin to communicate the difference.

3.5 Tiqqune Sopherim — The Eighteen Emendations

One especially fascinating masoretic note records the so-called tiqqune sopherim ("emendations of the scribes") — eighteen places (some traditions list a few more or fewer) where the Sopherim are said to have made deliberate alterations to the inherited text for reverential reasons, typically to avoid an anthropomorphic or potentially blasphemous-sounding statement about God. Example: Genesis 18:22 reads, "Abraham still stood before the LORD"; some traditions hold that the original read "the LORD still stood before Abraham" but was altered for theological propriety. The fact that the Masoretes preserved the record of these alleged emendations — refusing to hide the textual history even where it ran against later scribal tendencies — is a striking testament to scribal honesty.

It should be added that not all the tiqqune sopherim are equally certain; some appear to be Masoretic theological reflections on textual difficulties rather than records of actual emendation. But the principle they witness to is the same: the Masoretes inherited a text they regarded as sacred and would not alter without disclosing what they had done. This is not the textual posture of an editorialising scribal class; it is the textual posture of conservative custodians.

The Dead Sea Scrolls — Discovery, Range, Significance for the OT Text

"And he opened the scroll and found the place where it was written…" — Luke 4:17

4.1 Discovery and Inventory

In the winter of 1946–47, a Bedouin shepherd named Muhammad edh-Dhib ("Muhammad the Wolf") was searching for a lost goat in the cliffs above the northwestern shore of the Dead Sea. He threw a stone into a cave and heard pottery breaking. Inside the cave he found a collection of clay jars containing leather scrolls — the first installment of what would become the most important manuscript discovery of the twentieth century. Between 1947 and 1956, eleven caves in the cliffs near Qumran were excavated, yielding roughly nine hundred manuscripts in tens of thousands of fragments. About a quarter of them — roughly 230 — are biblical manuscripts; the rest include sectarian documents (the Community Rule, the War Scroll, the Damascus Document), Bible commentaries (pesharim), liturgical texts, and other Second Temple Jewish literature.

The biblical scrolls cover every book of the Hebrew Bible except Esther — which is sometimes attributed to the Qumran sect's discomfort with Esther's lack of explicit reference to God, but may simply be an accident of survival. The most-attested books at Qumran are the books most attested in early Jewish piety generally: Psalms (36 manuscripts), Deuteronomy (32), Isaiah (21), Genesis (20). The dates of the biblical scrolls range from roughly 250 BC to AD 70, with the bulk falling in the second and first centuries BC.

4.2 The Great Isaiah Scroll (1QIsa-a)

The most famous of the DSS biblical manuscripts is 1QIsa-a, the Great Isaiah Scroll, found in Cave 1 in 1947. It is a complete leather scroll of all sixty-six chapters of Isaiah, written in formal Hebrew script around 125 BC. It is now the showpiece of the Shrine of the Book in Jerusalem. Before 1QIsa-a, the oldest substantial Hebrew manuscript of Isaiah was a thousand years younger. The scroll therefore offered the first hard test of how well the Masoretic textual tradition had transmitted the Hebrew text over a millennium.

The result, repeatedly noted by Tov, Würthwein, and every careful student of the question, was vindicating. The consonantal text of 1QIsa-a agrees with the Masoretic Text at something like 95 percent of word-for-word readings. The differences are overwhelmingly orthographic (matters of spelling, especially the use of matres lectionis — consonants used as vowel-indicators), with a small number of genuine textual variants. A particularly striking case is Isaiah 53:11, where the Masoretic Text reads "he shall see [the fruit] of the travail of his soul," but 1QIsa-a, 1QIsa-b, and the Septuagint read "he shall see light" — a reading now followed by most modern translations (NIV, ESV, NRSV) and one that arguably enhances the resurrection-overtone of the verse. The variant is theologically significant but does not threaten any major Christian doctrine.

4.3 The Three (or Four) Text Families

The DSS revealed that the Hebrew text of the OT in the second century BC was not uniformly transmitted through a single channel. Tov and others now distinguish (with refinements) several text types attested at Qumran:

4.4 The Significance for OT Textual Criticism

The textual significance of the DSS is enormous, and four claims are now established by consensus among textual critics.

First, the DSS push the documentary chain back roughly a thousand years. For Isaiah, for Psalms, for Deuteronomy, for many other books, we now have manuscript witnesses from before the time of Christ — and they show that the text Jesus and the apostles read was substantially the text we have today.

Second, the DSS demonstrate the antiquity of the proto-Masoretic tradition. The MT is not (as some 19th-century scholars suspected) a late medieval reconstruction of a heavily edited Hellenistic text. It is a faithful preservation of a textual tradition already dominant in the late Second Temple period.

Third, the DSS confirm that the LXX is a real translation of a real Hebrew text. Before 1947, scholars debated whether the LXX's many divergences from the MT reflected a genuinely different Hebrew Vorlage or whether they were free translation choices. The DSS gave us actual Hebrew manuscripts (4QJer-b, 4QSam-a, others) whose readings underlie the LXX. The LXX is now established as a translation of a real, demonstrably different Hebrew textual tradition for some books.

Fourth, the DSS reveal the textual plurality of Second Temple Judaism without undermining the substantial unity of the text. There were multiple text types in circulation in the second and first centuries BC; the proto-MT was already dominant; the differences among the text types, while real, fall well short of demonstrating wholesale recomposition. The OT text was stable enough to be recognisably the same book across the various traditions.

The Septuagint (LXX) — Origin, Technique, Theological Importance

"All Scripture is breathed out by God…" — 2 Timothy 3:16, quoting from the Greek tradition

5.1 The Letter of Aristeas and the Traditional Story

The traditional story of the Septuagint's origin is preserved in the Letter of Aristeas, a Hellenistic Jewish document written in Greek probably in the mid-second century BC, purporting to be by a Greek courtier in the service of Ptolemy II Philadelphus (r. 285–246 BC). According to Aristeas, Ptolemy II wished to acquire for the great library of Alexandria a copy of the Jewish Law in Greek translation. He wrote to the high priest in Jerusalem, who sent seventy-two scholars — six from each of the twelve tribes — to undertake the translation. The translators were lodged on the island of Pharos, where they completed the work in seventy-two days. When the translation was finished, the assembled Jewish community of Alexandria pronounced it perfect and forbade any alteration.

The Letter of Aristeas is legendary in many of its details (the twelve-tribe representation is anachronistic; the perfect agreement of seventy-two translators working independently is improbable). But the core of the story is credible. There is good reason to believe that the Greek Pentateuch was produced in Alexandria in the third century BC under Ptolemaic patronage, by Jewish scholars who were both Hebrew-competent and Greek-fluent. By the second century BC the project had expanded; by the first century AD virtually the whole Hebrew Bible was available in Greek. The name Septuagint (Latin septuaginta, "seventy"; Greek hebdomēkonta) was originally applied only to the Pentateuch (the work of the "seventy" of the Aristeas legend) but came to denote the whole Greek OT.

5.2 Translation Technique

The Septuagint is not a single translation; it is a translation library, with different translators working on different books at different times. The translation technique varies accordingly. The Pentateuch is generally the most carefully translated, with a fairly close formal correspondence to its Hebrew Vorlage. The Prophets vary by book: Isaiah is somewhat free; Jeremiah is in a famously different (and shorter) form from the MT, suggesting a different Hebrew Vorlage. The Writings vary widely: Psalms is close to the MT; Proverbs is freer; Job is heavily reorganised; Esther has substantial Greek additions not in the MT. Daniel exists in two Greek forms (Old Greek and Theodotion's revision), with significant differences from the MT and notable additions (the Prayer of Azariah, Susanna, Bel and the Dragon).

The LXX has been called both the world's first major translation project and the world's first systematic mistranslation. Both descriptions overshoot. The LXX translators were generally competent, generally faithful, sometimes inspired in their lexical choices, and occasionally constrained by the limits of the Greek language to handle Hebrew idioms that did not transfer naturally. Where the LXX differs from the MT, the differences fall into three classes: (1) translation choices (a different Greek word, a different syntactic rendering, a slight paraphrase); (2) underlying Hebrew Vorlage differences (the translator was working from a Hebrew manuscript that genuinely differed from what the Masoretes would later preserve); and (3) interpretive expansions (the translator clarified or expanded a difficult Hebrew text for the Greek-speaking reader). Distinguishing these three classes is the daily work of LXX textual criticism.

5.3 The LXX as the Apostles' Bible

By the first century AD the LXX was the standard Bible of Greek-speaking Jewish and (then) Christian communities throughout the Mediterranean. It is the Bible Paul read in the synagogues of Asia Minor and Greece. It is the Bible Apollos used in his eloquent preaching from "the Scriptures" (Acts 18:24–28). When the New Testament authors quote the OT, they quote the LXX more often than they translate the Hebrew directly — by some counts, two-thirds or more of NT OT quotations agree with the LXX against the MT in some detail.

The most famous illustration is Hebrews 10:5–7, which quotes Psalm 40:6–8 in its LXX form: "Sacrifices and offerings you have not desired, but a body you have prepared for me." The MT of Psalm 40:6 reads, "you have not desired sacrifice and offering, but my ears you have dug for me." The shift from "ears" to "body" is in the LXX, and the author of Hebrews builds an extended Christological argument on the LXX wording. Another classic is Matthew 1:23, which quotes Isaiah 7:14 in its LXX form ("the virgin [Greek parthenos] shall conceive"), where the Hebrew word 'almah can mean "young woman" without necessarily specifying virginity. The LXX rendering, made centuries before Christ, becomes the foundation of the Christological apologetic for the virgin birth.

5.4 The Reformed View of the LXX

Two convictions hold together in the historic Reformed view of the LXX. First, the Reformed confessions identify the Hebrew Old Testament as authentical — the inspired original (WCF 1.8: "The Old Testament in Hebrew… being immediately inspired by God, and by his singular care and providence kept pure in all ages, are therefore authentical"). The LXX is a translation; it is not itself the inspired original. Second, the Reformed tradition has consistently honoured the LXX as the apostles' Bible — as an important textual witness, as a venerable Christian-era translation, and as evidence of the providential preparation of the Hellenistic world to receive the gospel.

The Reformed view thus sits between two errors. On one side is the Roman Catholic and Eastern Orthodox tendency (especially evident in the older Catholic preference for the Vulgate, and the Orthodox elevation of the LXX as the OT of the church) to treat a translation as authentical. On the other side is the King-James-Only-style dismissal of the LXX as a paraphrastic Hellenistic distortion irrelevant to the original. Neither position fits the evidence. The Hebrew is original; the LXX is the apostles' Bible; both deserve their proper place in Reformed exegesis.

פִּתֻּחֵי חוֹתָם קֹדֶשׁ לַיהוה "For thus says the LORD: 'Heaven is my throne, and the earth is my footstool; what is the house that you would build for me?'" — Isa 66:1, quoted in Acts 7:49 from the LXX. Acts 7 — Stephen quotes the LXX as Scripture in his Sanhedrin defence

5.5 The Recensions — Aquila, Symmachus, Theodotion

Three additional Greek translations of the OT were produced in the second century AD as Jewish responses to the Christian appropriation of the LXX. Aquila (c. AD 130) produced an extremely literal, hyper-Hebraic rendering, aiming at word-for-word equivalence even where the result was barely intelligible Greek. Symmachus (c. AD 170) produced a freer, more idiomatically Greek translation, more readable than Aquila but less literally tied to the Hebrew. Theodotion (c. AD 180) produced a revision aimed at bringing the LXX into closer alignment with the proto-Masoretic Hebrew; Theodotion's version of Daniel actually replaced the Old Greek Daniel in the Christian textual tradition and is the form of Daniel in most LXX manuscripts and editions today.

Origen of Alexandria (c. AD 240) gathered these four Greek versions (LXX, Aquila, Symmachus, Theodotion) plus the Hebrew text and a transliterated Hebrew into a six-column comparative edition called the Hexapla ("six-fold"). The Hexapla itself is lost, but its readings are preserved in marginal notes in many later manuscripts and in patristic commentaries. The Hexapla is the foundation of all modern critical work on the Greek OT.

The Samaritan Pentateuch

"Our fathers worshiped on this mountain, but you say that in Jerusalem is the place where people ought to worship." — John 4:20

6.1 The Samaritans and Their Bible

The Samaritans are a distinct religious community, descended in part from the inhabitants of the northern kingdom of Israel after the Assyrian conquest (722 BC), and in part from peoples settled there by the Assyrians. By the post-exilic period the Samaritans and the returning Judeans had become bitterly opposed (the books of Ezra, Nehemiah, and Haggai give the Jerusalem perspective on this conflict). The decisive break came probably in the fourth or third century BC, when the Samaritans built their own temple on Mount Gerizim (just north of the city of Shechem / modern Nablus) and codified their own version of the Hebrew Bible.

Critically, the Samaritan canon includes only the five books of Moses — the Pentateuch — and rejects the Prophets and the Writings. The Samaritan Pentateuch is therefore not a full OT but a Pentateuchal tradition, preserved in a community parallel to but separate from rabbinic Judaism. The earliest surviving SP manuscripts are medieval (the famous Abisha Scroll, kept in Nablus, is traditionally dated to the eleventh century AD or earlier; the oldest dated SP manuscript is from the twelfth century). But the textual tradition the SP preserves is much older, as the discovery of proto-Samaritan manuscripts at Qumran (notably 4QpaleoExod-m) confirms — the textual type was already in circulation by the second century BC.

6.2 The Distinctive SP Readings

The SP differs from the MT in roughly 6,000 readings — a large absolute number but small in proportion to the total volume of the Pentateuch. The differences fall into three broad classes:

1. Harmonising additions. The SP frequently expands one passage with material drawn from a parallel passage elsewhere in the Pentateuch. For example, in the plague narratives of Exodus the SP often duplicates the Lord's commands to Moses by inserting the actual execution of the command as a separate event, even when the MT presents the execution implicitly. The SP also frequently smooths chronological or logical difficulties by harmonising parallel narratives.
2. Orthographic and linguistic modernisation. The SP regularly fills in matres lectionis (vowel-letter spellings) that are written defectively in the MT, and occasionally substitutes a more current Hebrew form for an archaic one.
3. Sectarian theological alterations. A small but important number of SP readings are theologically motivated. The most famous is the SP's reading of the tenth commandment, which incorporates an additional commandment requiring the building of an altar on Mount Gerizim. Several other passages where the MT speaks of "the place that YHWH your God will choose" (e.g., Deut 12:5; 12:11; 16:6) are rendered in the SP as "the place that YHWH your God has chosen" — past tense, with Mount Gerizim assumed.

The sectarian theological alterations are clearly later Samaritan polemical additions, not original readings. The harmonising additions are more interesting from a textual standpoint: they show a scribal culture happy to clarify and smooth the text, contrasting with the more conservative proto-Masoretic tradition that preserved difficulties unaltered.

6.3 The Significance of the SP

The SP has been used and overused in OT textual criticism. Two cautions are necessary. First, the SP is not an independent witness to a pre-exilic Hebrew text; it is a parallel medieval transmission of an early text-type with substantial sectarian editing. Second, individual SP readings should be evaluated case by case, with attention to the SP's known harmonising and sectarian tendencies. When the SP agrees with the LXX against the MT in a non-tendentious reading, the agreement may indicate that both preserve an older form. When the SP agrees with the MT, the convergence of two independent traditions strongly supports the reading. When the SP stands alone with a harmonising or sectarian alteration, it is most likely a secondary modification.

The Reformed evangelical position has been to use the SP as a useful but secondary witness, valuable particularly for confirming the antiquity of MT readings against the LXX, but rarely decisive on its own. The text of the Pentateuch in modern critical editions (BHS, BHQ) follows the MT and notes SP variants in the apparatus.

The Targums and the Peshitta

"And they read from the book, from the Law of God, clearly, and they gave the sense, so that the people understood the reading." — Nehemiah 8:8

7.1 The Origin of the Targums

After the Babylonian exile, Hebrew gradually receded as the daily vernacular of the Jewish people, displaced by Aramaic (the lingua franca of the Persian and Hellenistic Near East). By the second century BC, ordinary Jews in Palestine, Babylonia, and the wider diaspora needed Aramaic translation to follow the Hebrew Scripture readings in synagogue. The translator (Aramaic meturgeman) would render the Hebrew into Aramaic verse by verse — at first orally, but in due course writing the translation down. These written translations are the Targums.

Nehemiah 8:8 may already attest to the practice: when Ezra reads the Law to the assembled people, the Levites "gave the sense, so that the people understood the reading." Whether this means Aramaic translation, paragraph-by-paragraph explanation, or both is debated. By the first century AD the Targumic practice is well established; the Babylonian Talmud (tractate Megillah) regulates the public reading and translation of Scripture in detail.

7.2 The Major Targums

7.3 The Targumic Method

The Targums are not slavishly literal translations; they are interpretive renderings that incorporate the synagogue's traditional exegesis. Three characteristic Targumic moves are worth noting.

Anti-anthropomorphism. Targum Onqelos consistently softens anthropomorphic language about God. Where the Hebrew says "the hand of the LORD," Onqelos substitutes "the power of the LORD" or "a plague from before the LORD." Where the Hebrew says "the LORD walked in the garden," Onqelos rephrases. The instinct is reverential, but it has the effect of distancing God from the immediate language of the Hebrew.

The Memra. The Targums regularly use the Aramaic term Memra ("Word") as a circumlocution for direct divine action. Where the Hebrew says "the LORD did X," the Targum says "the Memra of the LORD did X." There has been long scholarly debate about the relationship of the Targumic Memra to the Johannine Logos. Most contemporary scholarship rejects a direct genetic relationship while noting the shared intellectual milieu — both terms are ways of speaking of God's effective speech as a quasi-personal agent of his action in the world. The early Jewish-Christian reader of John 1 would have heard the Logos against a background that included Memra theology, though the Christian doctrine of the personal pre-existent Word goes far beyond what Targumic Memra requires.

Messianic expansion. The Targums regularly identify messianic passages and expand them with explicit reference to the coming Messiah. Targum Jonathan on Isaiah 52:13 reads, "Behold my servant the Messiah shall prosper." The Targums therefore preserve a pre-Christian Jewish messianic reading of OT prophecy that confirms the apostolic exegesis was not a Christian invention — it engaged Jewish messianic expectation already present in Second Temple and rabbinic interpretation.

7.4 The Peshitta — the Syriac Bible

The Peshitta (Syriac pəšiṭtā, "simple, straightforward") is the standard Bible of Syriac-speaking Christianity. It was produced in stages between the second and fourth centuries AD, probably with some Jewish predecessors (a Jewish-Aramaic translation of parts of the OT, later adopted and adapted by Syriac Christians). The Peshitta OT is a largely literal translation from a proto-Masoretic Hebrew Vorlage, with occasional influence from the LXX. It is the OT of the Syriac Orthodox, Assyrian, Maronite, and Indian (Mar Thoma) churches.

Textually, the Peshitta is valuable as an independent witness whose agreements with the MT overwhelmingly support the proto-Masoretic tradition. Where the Peshitta agrees with the LXX or with a DSS reading against the MT, the convergence may witness to an older reading. Where the Peshitta agrees with the MT against other witnesses, its independent testimony strengthens the proto-MT.

7.5 A Note on the Vulgate

For completeness, the Latin Vulgate deserves brief mention. Jerome's late-fourth-century Latin translation of the OT was produced, deliberately, from the Hebrew (Hebraica veritas, "the Hebrew truth") rather than from the LXX. Jerome learned Hebrew (an unusual achievement for a Western Christian of his day) and consulted Jewish teachers in his retirement at Bethlehem. His translation became the standard Bible of Western Christianity for over a millennium. Critically, Jerome's Hebrew Vorlage was substantially the proto-Masoretic text type — confirming once again the dominance of that tradition in the late ancient period. The Vulgate is therefore valuable both as a textual witness and as the conduit through which Hebrew readings entered Western European biblical culture before the Reformation.

Modern Critical Editions — BHS, BHQ, HBCE

"Of making many books there is no end…" — Ecclesiastes 12:12

8.1 From Bomberg to BHS — A Brief History

Printed Hebrew Bibles begin with the Soncino edition of 1488 and reach a watershed in the great rabbinic editions of Daniel Bomberg in Venice. The Mikra'ot Gedolot ("Great Scripture") or Second Rabbinic Bible (1524–25), edited by Jacob ben Chayyim ibn Adonijah and printed by Bomberg, became the textus receptus of the Hebrew Bible for the next four centuries. Every European Christian Hebraist from Luther and Calvin to the seventeenth-century post-Reformation orthodox to the nineteenth-century textual scholars worked from the Ben Chayyim text. The KJV (1611) translates from the Ben Chayyim Hebrew. The text is solid; the Ben Chayyim edition is, in effect, an eclectic Masoretic Text drawing on a number of medieval Spanish and Ashkenazi manuscripts available to Ben Chayyim.

The transition to a critical edition based on a single early manuscript came in the early twentieth century. Rudolf Kittel's Biblia Hebraica (BHK, 1st ed. 1906) initially used the Ben Chayyim text but, in its third edition (1937), switched to the Leningrad Codex (L) as the base text — for the first time printing the actual text of the oldest complete Hebrew Bible manuscript. The third Kittel edition (BHK3) was the standard scholarly Hebrew Bible for forty years.

8.2 Biblia Hebraica Stuttgartensia (BHS)

The Biblia Hebraica Stuttgartensia (BHS), edited by Karl Elliger and Wilhelm Rudolph and published by the Deutsche Bibelgesellschaft in Stuttgart, appeared in fascicles from 1968 to 1976 and was published as a complete volume in 1977. It is the de facto standard Hebrew Bible of the late twentieth and early twenty-first centuries — the Hebrew Bible owned by every seminary student and consulted by every careful expositor of the OT for roughly four decades.

BHS is a diplomatic edition: it prints the text of a single chosen manuscript (the Leningrad Codex) as faithfully as possible, with variant readings from other witnesses recorded in a footnote-style apparatus at the bottom of each page. The Masoretic apparatus (Mp) is also reproduced in the side margins; the Masora Magna is provided in a separate companion volume. The apparatus draws on the Cairo Codex of the Prophets, the Aleppo Codex (where available), other Masoretic manuscripts, the DSS, the LXX, the SP, the Targums, the Peshitta, the Vulgate, and a substantial roster of conjectural emendations. The result is a single working Bible for the academic study of the OT.

8.3 Biblia Hebraica Quinta (BHQ)

The Biblia Hebraica Quinta (BHQ, the "fifth" Biblia Hebraica, succeeding BHK1, BHK2, BHK3, and BHS) is the in-progress successor edition, published in fascicles by the Deutsche Bibelgesellschaft beginning in 2004. As of 2025, roughly two-thirds of the OT books have appeared in BHQ form; the project is expected to be complete in the 2030s.

BHQ retains the Leningrad Codex as base text and the diplomatic-edition philosophy but expands the apparatus in three ways. First, the apparatus is more extensive and includes additional textual witnesses (especially the DSS, where the relevant evidence has appeared since BHS was published). Second, each fascicle includes a fuller textual commentary explaining the editors' decisions on the most significant variants. Third, the Masora Magna is integrated with each fascicle rather than relegated to a separate volume. The result is a more accessible, more thoroughly documented working Bible, though still operating within the same fundamental diplomatic-MT framework as BHS.

8.4 The Hebrew Bible: A Critical Edition (HBCE)

The Hebrew Bible: A Critical Edition (HBCE), formerly known as the Oxford Hebrew Bible, is a different kind of project entirely. Begun in the early 2000s under the general editorship of Ronald Hendel (UC Berkeley), HBCE is an eclectic edition: rather than printing the text of a single manuscript, it aims to reconstruct a hypothetical archetype — the earliest recoverable form of each book — by drawing eclectically from all available witnesses (MT, DSS, LXX, SP, Targums, Peshitta, Vulgate). Where the editors judge that the LXX or a DSS reading preserves an earlier form than the MT, the HBCE text prints the earlier form and relegates the MT reading to the apparatus.

HBCE is published by the Society of Biblical Literature; only a few volumes have appeared as of 2025 (Proverbs by Michael Fox, 2015; Deuteronomy in progress; others to follow). It is the methodologically most ambitious of the three editions, and also the most controversial. Two responses are in order from the Reformed evangelical perspective.

Positive engagement. HBCE forces the explicit textual-critical question for each major variant; it surfaces evidence that BHS and BHQ leave in the apparatus margin; it pushes the discipline toward greater methodological rigour. As a reference and as a check on BHS/BHQ readings, HBCE is a valuable contribution.

Critical reserve. HBCE departs from MT on the basis of editorial judgement about which reading is "earliest." Such judgements are inherently more conjectural than the diplomatic-edition approach. The Reformed evangelical, with the confessional commitment to the providentially preserved Hebrew text, will be slower to adopt eclectic departures from MT without strong supporting witness. The Reformed convictions about providential preservation suggest that the Masoretic tradition has unusual claim on our trust precisely as the tradition the church has historically received and used.

1. Biblia Hebraica Kittel (BHK)
   1st ed. 1906, 3rd ed. 1937 (first to use L)
   The first major critical edition of the Hebrew Bible. The third edition switched the base text from Ben Chayyim to the Leningrad Codex (L), establishing the diplomatic-MT-on-L framework that has governed German-tradition critical editions ever since.
2. Biblia Hebraica Stuttgartensia (BHS)
   1977, ed. Elliger & Rudolph
   The standard scholarly Hebrew Bible from the 1980s through the early 2000s. Still in print and still used as a working Bible by most seminarians and pastors. Diplomatic edition of L with critical apparatus.
3. Biblia Hebraica Quinta (BHQ)
   2004– (in progress)
   The successor to BHS. More extensive apparatus, integrated Masora Magna, fuller textual commentary. Still diplomatic of L. Expected completion in the 2030s.
4. Hebrew University Bible (HUB)
   1965– (in progress)
   An Israeli project producing a diplomatic edition of the Aleppo Codex (A). Only Isaiah, Jeremiah, and Ezekiel published so far. Very full apparatus including LXX readings in Greek script, Talmudic and rabbinic citations, and medieval Hebrew variants.
5. Hebrew Bible: A Critical Edition (HBCE)
   2015– (in progress)
   The Oxford / SBL eclectic edition under Hendel. Aims to reconstruct a pre-Masoretic archetype. Departs from MT where the editors judge that other witnesses preserve an earlier reading. Methodologically ambitious; engaged critically by Reformed evangelicals.

8.5 Which Edition Should the Reformed Student Use?

For pastoral and devotional use, an English Bible translated from the MT (ESV, NASB, NIV, NKJV; the KJV uses the Ben Chayyim, which is substantially the same) is fully adequate. For seminary study and exegesis, BHS or BHQ is the standard working Bible. BHS remains the most widely available and most affordable; BHQ is preferable where the fascicle for the relevant book has appeared. HBCE should be consulted where available as a check and a stimulus to engagement with the textual data, but its eclectic departures from MT should be evaluated case by case rather than adopted wholesale.

The Apparatus — How to Read It

"Test everything; hold fast what is good." — 1 Thessalonians 5:21

9.1 The Purpose of the Apparatus

A critical edition of an ancient text consists of two layers: the printed text (what the editor judges the original most likely said) and the apparatus (a footnote system recording the variant readings the editor declined to adopt and the manuscripts in which they are found). The apparatus is therefore the student's window into the textual evidence that lies behind the printed text. Every serious user of BHS or BHQ has to learn to read the apparatus; otherwise, the apparatus is just noise at the bottom of the page.

The apparatus is structured as a sequence of variant units, each tied to a specific verse and word in the text above. Each variant unit lists the alternative reading(s) and the witnesses that attest each reading. The format is compressed for space — a single variant unit may fit in a single line — so it relies heavily on abbreviations and conventional symbols (the sigla).

9.2 The Most Important Sigla

The following are the most important sigla used in BHS and BHQ. Variations exist between editions; the student should consult the front matter of whichever edition is in hand for the authoritative list.

• 𝔐 / MT The Masoretic Text — usually the base text of the printed page.
• 𝔊 / LXX The Septuagint, in its Old Greek form. Specific LXX manuscripts can be cited as 𝔊ᴬ (Codex Alexandrinus), 𝔊ᴮ (Codex Vaticanus), 𝔊ˢ (Codex Sinaiticus).
• 𝔖𝔓 / Smr The Samaritan Pentateuch.
• 𝔗 / Tg The Targums (Aramaic). Specified as 𝔗ᴼ (Onqelos), 𝔗ᴶ (Jonathan), 𝔗ᴺ (Neofiti), 𝔗ᴾᴶ (Pseudo-Jonathan).
• 𝔖 / Pesh The Peshitta (Syriac).
• 𝔙 / Vg The Vulgate (Latin).
• α' / σ' / θ' Aquila, Symmachus, and Theodotion (Hexaplaric Greek revisions).
• Q / DSS The Dead Sea Scrolls. Specific manuscripts are cited by find-site and abbreviation: 1QIsa-a, 4QSam-a, 4QJer-b, etc.
• L The Leningrad Codex.
• A The Aleppo Codex.
• C The Cairo Codex of the Prophets.
• Or Origen / Hexaplaric tradition.
• Ms / Mss One or several Hebrew manuscripts (specified by Kennicott number or de Rossi number in older works).
• Edd Printed editions (Bomberg, Ben Chayyim, Soncino, etc.).

9.3 The Most Important Notations

The apparatus uses a set of Latin and German abbreviations to indicate the kind of variant being recorded. The most important:

• add. addit / addunt — adds (the witness adds this word or phrase to the MT).
• om. omittit / omittunt — omits (the witness lacks a word or phrase the MT has).
• ins. inserit — inserts (often used for an explanatory addition).
• tr. transponit — transposes (the witness has the words in a different order).
• cf. confer — compare (drawing attention to a parallel or related reading).
• l lege — read (the editor proposes that the text be read differently).
• prps. proposuit / proponitur — has proposed / is proposed (a conjectural emendation).
• c cum — with (the variant is supported with this witness).
• vid. videtur — apparently, seems (the reading is uncertain in the witness).
• frt. fortasse — perhaps.
• dl. dele — delete (the editor proposes removing the word).
• gloss. a marginal gloss has been incorporated into the text.
• dittog. dittography — a scribal error of duplicating a letter or word.
• haplogr. haplography — a scribal error of writing once what should be written twice.
• homoeotel. homoeoteleuton — a scribal omission caused by two lines ending similarly.

9.4 A Worked Example

To make the system concrete, consider a representative apparatus note. Suppose the text of 1 Samuel 14:41 reads in BHS:

What the apparatus is telling us: the MT of 1 Samuel 14:41 contains a short, somewhat puzzling Hebrew clause; the LXX preserves a much longer reading that explicitly mentions Urim and Thummim and is internally coherent; the Vulgate partly preserves the longer reading; the editors of BHS judge that the longer LXX reading is original and the shorter MT reading reflects an early scribal omission (probably through homoeoteleuton — a scribe's eye jumping from one phrase to a similar phrase later in the verse). Most modern English translations follow the LXX here (ESV, NIV, NRSV: "Therefore Saul said, 'O LORD God of Israel, why have you not answered your servant this day?…'").

9.5 Forming a Judgement on a Variant

The apparatus does not decide the textual question for you. It gives you the evidence; you must form a judgement. Reformed evangelical textual criticism has historically held to a fairly conservative bias toward MT — the providentially preserved text the church has received — while remaining open to following another witness when (a) the textual evidence strongly favours it; (b) the variant is a plausible older reading; and (c) the MT reading is genuinely difficult on internal grounds. The classic textual-critical principles apply:

• Lectio difficilior potior — "the harder reading is preferred." Scribes tended to ease difficulties, not introduce them.
• Lectio brevior potior — "the shorter reading is preferred." Scribes tended to expand, not condense (with the caveat that omission by homoeoteleuton is also common, so this principle is not absolute).
• External evidence — the number, age, and independence of the witnesses supporting each reading.
• Internal evidence — what reading best explains the origin of the others; what reading fits the author's style and theology.

The Reformed evangelical student should learn to weigh these factors with a presumption in favour of the proto-Masoretic reading and a willingness to follow the evidence where it leads in genuinely difficult cases.

Major Textual Variants and Their Doctrinal Implications

"The sum of your word is truth, and every one of your righteous rules endures forever." — Psalm 119:160

10.1 1 Samuel 13:1 — Saul's Regnal Years

The Masoretic Text of 1 Samuel 13:1 reads, literally translated, "Saul was [...] year[s] old when he became king, and he reigned two years over Israel." The numeral for Saul's age is missing; the duration of his reign reads "two years," which is plainly too short (compare Acts 13:21, "forty years"). The verse is one of the clearest examples of textual damage in the MT. The LXX of 1 Samuel 13:1 omits the verse entirely in many manuscripts; in others it has "and Saul was thirty years old [or one year old] when he began to reign." Modern English translations handle the difficulty differently:

The textual question is real; the doctrinal stakes are negligible. No Christian doctrine depends on Saul's age or the exact length of his reign. The variant illustrates the kind of textual problem the OT genuinely has — and how modest such problems are.

10.2 Psalm 22:16 — "They Pierced" or "Like a Lion"?

Psalm 22:16 (Heb. 22:17) is one of the most famous and theologically charged textual variants in the OT. The Masoretic Text reads כָּאֲרִי יָדַי וְרַגְלָי (kā'ărī yāday wəraglāy), traditionally translated "like a lion my hands and my feet." The Septuagint reads ōryxan cheiras mou kai podas mou — "they have pierced my hands and my feet." Many other early witnesses, including the Peshitta, the Vulgate, and some Hebrew manuscripts, support the LXX reading. The difference in the Hebrew is between kā'ărī (כארי, "like a lion") and kā'ărū or kārû (כארו or כרו, "they pierced" or "they dug") — a difference of a single letter (yod vs vav, which are very similar in many Hebrew scripts).

The textual question is therefore: does the original Hebrew have kā'ărī ("like a lion") or kā'ărū ("they pierced")? Reformed evangelical scholarship now generally favours kā'ărū for three reasons. First, the LXX witnesses to it (third or second century BC). Second, a Hebrew DSS manuscript (5/6HevPs, mid-first century AD, before Christian influence on Jewish textual transmission) actually reads kā'ărū. Third, the MT kā'ărī ("like a lion") leaves the Hebrew sentence verbless and difficult; the kā'ărū ("they pierced") reading gives a coherent sentence. The variant is therefore best understood as an early scribal substitution of yod for vav, perhaps for theological reasons (downplaying a christologically charged verse) or perhaps as an accidental misreading. The doctrinal payoff: the verse predicting the piercing of the messianic sufferer's hands and feet was already present in pre-Christian Hebrew tradition — the Christian application to the crucifixion is exegetically warranted, not Christian eisegesis.

10.3 Genesis 4:8 — The Missing Words of Cain

Genesis 4:8 in the MT reads, "Cain said to Abel his brother. And when they were in the field, Cain rose up against Abel his brother and killed him." The Hebrew is grammatically incomplete: "Cain said" demands an object — what did he say? The verse has clearly suffered a small scribal loss. The LXX, the Samaritan Pentateuch, the Vulgate, the Peshitta, and the Targums all preserve the words: "Cain said to Abel his brother, Let us go out to the field." Most modern English translations restore the missing words on this strong cross-witness evidence (ESV: "Cain spoke to Abel his brother. And when they were in the field…" with footnote; NIV: "Now Cain said to his brother Abel, 'Let's go out to the field.'"). This variant is a textbook case of a small scribal omission that the versions enable us to confidently restore.

10.4 Deuteronomy 32:8 — "Sons of Israel" or "Sons of God"?

Deuteronomy 32:8 is one of the most theologically interesting variants. The MT reads, "When the Most High gave to the nations their inheritance… he fixed the borders of the peoples according to the number of the sons of Israel (bnê Yisrā'ēl)." The LXX reads, "according to the number of the angels of God (angelōn theou)." A Hebrew DSS manuscript (4QDeut-q) reads "the sons of God (bnê 'ĕlōhîm)." The variant is therefore:

• MT: sons of Israel (bnê Yisrā'ēl)
• LXX: angels of God (angelōn theou) — translating a Hebrew Vorlage with bnê 'ĕlōhîm
• 4QDeut-q (DSS): sons of God (bnê 'ĕlōhîm)

The DSS reading and the LXX agree against the MT, and on internal grounds the "sons of God" reading is the more difficult and more likely original — it makes sense of God "fixing the borders of the peoples" according to a heavenly host (one divine being per nation, with Israel being YHWH's own portion in v. 9). The MT reading "sons of Israel" appears to be a theological correction by a scribe uncomfortable with the implied divine council. The ESV (2011 revision) and the NRSV both follow the DSS/LXX reading. The doctrinal implications are limited but real: the original reading supports the OT picture of a divine council under YHWH, a picture the NT inherits and develops (Michael Heiser and others have explored this in The Unseen Realm). The Reformed evangelical can affirm the divine-council reading without anxiety — it is biblical and traditional.

10.5 The Long Ending of Deuteronomy 32

The Song of Moses (Deut 32) and the surrounding chapters present another set of variants. The MT and the LXX preserve essentially the same Song, but with significant differences in vv. 43–47 (the closing benediction and Moses's final instructions). The LXX of Deut 32:43 is substantially longer than the MT, with additional clauses calling the heavens and the angels to rejoice. Hebrews 1:6 quotes one of the LXX clauses of Deut 32:43 ("let all God's angels worship him") and applies it to Christ. A DSS manuscript (4QDeut-q) supports the longer reading. The Reformed evangelical reader will note that the apostles' citation of a longer reading in the LXX of Deuteronomy 32:43 confirms again that the LXX was not freely paraphrasing — it preserved a real Hebrew tradition that the writer of Hebrews could cite as Scripture.

10.6 The Joshua Plus (Joshua 8:30–35 and the LXX Reordering)

Joshua 8:30–35 in the MT records the building of an altar on Mount Ebal and the reading of the Law there. In the LXX, this passage appears in a different location — placed after the Ai narrative ended and before the Gibeonite deception. A few DSS Joshua fragments support the LXX placement. The textual question is which ordering is original. Scholars are divided; the issue is one of literary structure rather than doctrine, and the underlying historical event (the Mount Ebal ceremony) is preserved in both traditions. The variant illustrates how the OT text was sometimes transmitted with minor reorderings even in the late Second Temple period.

10.7 Jeremiah — Two Editions

The most extensive OT textual variation is in Jeremiah. The MT of Jeremiah is roughly one-seventh (about 14 percent) longer than the LXX of Jeremiah, and the two texts also order some material differently — most notably the Oracles Against the Nations, which appear in the middle of the book in the LXX but at the end in the MT. The DSS preserve both forms: 4QJer-a and 4QJer-c align with the MT; 4QJer-b and 4QJer-d align with the shorter LXX Vorlage. The discovery confirmed what scholars had long suspected: Jeremiah circulated in two ancient editions, a shorter (preserved in the LXX) and a longer (preserved in the MT), both demonstrably attested in pre-Christian Hebrew manuscripts.

The Reformed evangelical response to the Jeremiah variation is not panic but careful exegesis. Two ancient editions of a prophetic book both circulating in inspired Hebrew form is consistent with the prophet's own role in compiling and re-issuing his oracles (cf. Jer 36, where Baruch produces a second scroll after Jehoiakim destroys the first). The MT preserves what looks like the prophet's final, fuller edition; the LXX preserves an earlier shorter edition. Both are substantially the prophet Jeremiah; neither overturns Christian theology. The canonical Reformed practice is to use the MT (the longer, fuller, traditionally received form) while engaging the LXX form as a witness to the prophet's textual history.

The pattern across all these variants is consistent: the textual evidence is real, the questions are interesting, the modern critical editions handle them carefully, and the doctrinal stakes are modest to zero. The Christian who is told that the OT is "full of contradictions and uncertainties" has been told a simplification. The Christian who is told that "the Hebrew Bible is perfect to the last yod" has also been told a simplification. The truth is the more remarkable: a millennia-old text, copied and recopied by hand across thirty centuries, has come down to us with such substantial stability that the modern critical apparatus consists chiefly of minor variants and a handful of more significant readings — none of which threatens the Christian gospel.

The "Goliath's Height" Question — Three Witnesses Compared

"There went out a champion out of the camp of the Philistines, named Goliath, of Gath, whose height was…" — 1 Samuel 17:4

11.1 The Three Witnesses

Goliath of Gath's height is recorded in 1 Samuel 17:4. The three earliest witnesses give three different numbers, though only two of them differ significantly.

11.2 The Arguments For and Against Each Reading

The case for the MT "six cubits and a span" (9' 9"). The MT is the received tradition; it is the reading the church has used for two thousand years; it is internally consistent with the rest of the narrative (Goliath is a champion, the situation calls for an extraordinary opponent). Some scholars also note that the unusual height fits an Anakite or Rephaite lineage (Num 13:33; Deut 2:10–11); the OT records giants of comparable height (e.g., Og of Bashan and his iron bedstead, Deut 3:11). The MT reading is not impossible on physiological grounds — known cases of acromegaly (such as Robert Wadlow, who reached 8' 11" in the twentieth century) demonstrate that human beings can reach extreme heights.

The case for the LXX/4QSam-a "four cubits and a span" (6' 9"). The LXX and 4QSam-a are the older witnesses, and they agree. The LXX of Samuel as a whole is frequently shown by 4QSam-a to preserve older readings than the MT (4QSam-a is one of the most "non-MT" of the DSS biblical manuscripts). The shorter height is internally coherent with the narrative — it makes Goliath a giant in the Philistine champion sense but not an impossible figure. The shorter height also makes better sense of why David's confidence in YHWH (rather than his own strength) is the narrative's point: a 6' 9" Goliath is intimidating but not physically impossible to defeat by skill and surprise; a 9' 9" Goliath is so extraordinary that any sling-stone strike on him would seem miraculous, undermining the narrative's emphasis on YHWH's deliverance through ordinary means. (This last argument cuts both ways: an extraordinary Goliath equally well showcases YHWH's power.) Josephus (Antiquities 6.171) gives the height as "four cubits and a span," following the LXX tradition.

The textual-critical verdict. Most contemporary textual critics, including conservative evangelical ones, judge that the original reading is "four cubits and a span" and that the MT "six cubits and a span" is a scribal exaggeration — either accidental (a copyist's slip from 'arba' to šēš, both being short Hebrew words) or deliberate (a later scribe enhancing the heroic dimension of the story). The evidence: the LXX and the DSS (the two oldest witnesses) agree; the MT stands alone; the shorter reading is internally coherent and even narratively more pointed. Brotzman and Tully, Wegner, and the editorial committees of recent English translations have generally followed this judgement.

11.3 What This Variant Teaches

The Goliath variant is a teaching example for several reasons. It is famous enough that the student has heard of it. It is concrete enough that the textual question can be presented cleanly. It is real enough that the student cannot dismiss it as imagined. And it is theologically modest enough that the student can examine it without anxiety. Three lessons follow.

First, the textual evidence is sometimes complex and sometimes points away from the MT. The Reformed evangelical does not start from the position that "the MT is always right." We start from the position that the proto-Masoretic tradition is the dominant ancient tradition, that the MT has been providentially preserved, and that where the MT and other ancient witnesses converge we have very high confidence in the reading. Where the older witnesses (LXX, DSS) unite against the MT, we have to examine the evidence and form a judgement.

Second, the variants that look most dramatic to popular readers often have negligible doctrinal stakes. Goliath at 6' 9" or 9' 9": David still wins, YHWH still vindicates his anointed, the messianic-shepherd-king typology still holds. The reader who hears "the LXX disagrees with the Hebrew about Goliath" and panics has confused a textual question with a doctrinal one.

Third, the older witnesses sometimes preserve older readings. This is part of what providential preservation has provided for us. By multiplying witnesses (Hebrew MT, Greek LXX, Aramaic Targums, Syriac Peshitta, Samaritan Pentateuch, Hebrew DSS, Latin Vulgate) God has ensured that we can cross-check readings. The witnesses are independent enough that the truth can usually be discerned; they are interdependent enough that the great mass of the text is held in stable convergence.

The Reliability of the OT Text — A Reformed Confessional Position

"The words of the LORD are pure words, like silver refined in a furnace on the ground, purified seven times." — Psalm 12:6

12.1 The Empirical Case

The cumulative empirical case for the substantial reliability of the OT text rests on several converging lines of evidence.

The proto-Masoretic textual tradition is ancient. The DSS (third century BC to first century AD) show that the proto-Masoretic text was the dominant Hebrew textual tradition already in the late Second Temple period. Outside Qumran, at sites like Masada and Murabba'at (where the discoveries date to AD 70 and the Bar Kokhba revolt of AD 132–135), the proto-MT is essentially the only Hebrew text type attested. The Masoretes did not invent the text they transmitted; they inherited a tradition that was already over a thousand years old when they received it.

The Masoretic transmission was extraordinarily careful. The comparison of 1QIsa-a (c. 125 BC) with the MT of Isaiah (c. AD 1008) — a thousand-year span — yields agreement at the 95-percent level. Studies of Andersen and Forbes and others have refined the comparison statistically and confirmed the basic finding. This is unprecedented in the textual transmission of any pre-printing manuscript culture. The Sopherim, the Masoretes, and the medieval Jewish scribes preserved the text with a fidelity that would be remarkable for a modern photocopier.

The cross-witness confirms the substance. Where the MT is in doubt, the LXX, the SP, the DSS, the Targums, the Peshitta, and the Vulgate provide independent witnesses against which to check. The independence of these witnesses (especially LXX vs MT vs SP) means that their agreement on the great mass of readings is strong evidence for those readings. The places where the witnesses disagree are precisely where careful critical work is needed — and they are concentrated in a small percentage of the text.

No major doctrine is at stake. The catalogue of doctrinally significant textual variants in the OT is short. Most are minor. None overturns the Christian faith. The Reformed confessional doctrine of the perfection and sufficiency of Scripture is fully consistent with the textual data — provided we understand "perfection" properly (the autograph was perfect; the transmission has been providentially adequate) and "sufficiency" properly (every doctrine of the faith is multiply attested in the canonical text).

12.2 The Doctrinal Case

The Reformed confessional doctrine of Scripture (described more fully in Section 13) provides the dogmatic framework within which the empirical data is interpreted. The doctrine has two parts.

Inspiration. The Hebrew Scriptures were given by inspiration of God (2 Tim 3:16; 2 Pet 1:20–21); they are God's own Word, breathed out by his Spirit, free from error in the original autographs. This is the church's universal historic confession; it is the position of every Reformed confession; it is the necessary entailment of the Bible's self-attestation.

Providential preservation. God has not only inspired Scripture but providentially preserved it across the centuries, so that the church has always had ready access to the substantial content of the divine Word (WCF 1.8). Preservation is not a miraculous fixity of every individual manuscript; it is the providential ordering of the transmission process — multiple witnesses, scrupulous scribes, the manuscript record kept by Jewish and Christian communities — such that the textual content of the autographs has been substantially retained.

These two doctrines together generate the Reformed expectation: that the manuscript evidence, when carefully examined, will show a substantially stable, faithfully transmitted, reliably reconstructable Old Testament text. And that is precisely what the manuscript evidence shows. The empirical data does not vindicate the doctrine by accident; the doctrine is the theological reading of the historical-providential process by which the data came to be.

12.3 The Christian's Warranted Confidence

The Christian student, reader, pastor, or scholar who learns the science of OT textual criticism does not come out doubting Scripture; he or she comes out with greater confidence in it. The textual evidence is genuinely strong; the doctrinal commitments are genuinely warranted; the two converge in a single conviction: the Old Testament we have in our Bibles is, in the words of the WCF, "kept pure in all ages" and is therefore "authentical." We do not have the autograph of Genesis or Isaiah; but we have, through the providence of God and the careful labour of two thousand years of scribes, a text whose substantial content is what the inspired authors wrote.

לְעוֹלָם יְהוָה דְּבָרְךָ נִצָּב בַּשָּׁמָיִם "Forever, O LORD, your word is firmly fixed in the heavens. Your faithfulness endures to all generations; you have established the earth, and it stands fast." Psalm 119:89–90 — the dogmatic claim the textual evidence supports

The Reformed Confessions on the Preservation of Scripture

"All Scripture is breathed out by God and profitable for teaching, for reproof, for correction, and for training in righteousness." — 2 Timothy 3:16

13.1 Westminster Confession of Faith 1.8

The locus classicus of Reformed teaching on the preservation of Scripture is WCF 1.8 (1646), reproduced verbatim in the 1689 LBCF (the Reformed Baptist confession) and substantially in the Savoy Declaration (the Congregationalist confession). The text reads:

"The Old Testament in Hebrew (which was the native language of the people of God of old) and the New Testament in Greek (which at the time of the writing of it was most generally known to the nations), being immediately inspired by God, and by his singular care and providence kept pure in all ages, are therefore authentical; so as in all controversies of religion the church is finally to appeal unto them. But because these original tongues are not known to all the people of God, who have right unto, and interest in the Scriptures, and are commanded, in the fear of God, to read and search them; therefore they are to be translated into the vulgar language of every nation unto which they come, that the Word of God dwelling plentifully in all, they may worship him in an acceptable manner, and through patience and comfort of the Scriptures, may have hope." Westminster Confession of Faith 1.8

The careful student of WCF 1.8 will notice four claims, each one precisely formulated.

(1) The original language Scriptures are inspired. The confession identifies the inspired text as the OT in Hebrew and the NT in Greek. The autographs were inspired in their original languages; translations, however excellent, are translations of the inspired text, not themselves the immediately inspired text.

(2) The text has been "kept pure in all ages." The confession affirms divine preservation. But notice the phrase: not "kept perfect in every manuscript" — which would be empirically untenable and theologically unnecessary — but "kept pure in all ages." The Hebrew word for "pure" here translates the underlying theological concept of integrity and substantial reliability, not numerical identity of every copy.

(3) The preservation is by God's "singular care and providence." Preservation is providential, not miraculous in the sense of suspending ordinary scribal processes. God preserves the text through the means of careful Jewish and Christian scribal traditions, the multiplication of manuscripts, the cross-witness of versions, and the discipline of the church. The Reformed doctrine is a doctrine of providence, not magic.

(4) The Hebrew and Greek originals are "authentical." This is a precise technical term. To be "authentical" means to be the authoritative standard — the text against which translations are measured, the text to which the church appeals in controversies, the text from which doctrine is drawn. The Reformed tradition has consistently insisted on the priority of the original-language text and has resisted any elevation of a translation (such as the Vulgate in pre-Tridentine Catholicism or, in our day, the KJV in some quarters) to that authentical status.

13.2 Belgic Confession 3–7

The Belgic Confession (1561), a Reformed confession adopted by the continental Reformed churches and one of the Three Forms of Unity, treats Scripture in articles 3 to 7. Article 3 affirms the divine origin of Scripture ("we confess that this Word of God was not sent nor delivered by the will of man, but holy men of God spoke as they were moved by the Holy Spirit, as the apostle Peter says"). Article 4 lists the canonical books and excludes the Apocrypha (treated in Article 6). Article 5 affirms the divine authority of Scripture and that the Holy Spirit witnesses in our hearts that the canonical books are from God. Article 7 affirms the sufficiency of Scripture.

The Belgic Confession does not formulate the doctrine of preservation as precisely as WCF 1.8 does (the issue had not been forced to the same degree of formulation in 1561 as it had been by 1646). But the implicit doctrine is the same: the Scripture the church has received is the inspired Word of God, sufficient for faith and life, to be read in the original languages by those competent to do so and in translations by those who are not.

13.3 The 1689 London Baptist Confession 1.8

The 1689 LBCF reproduces WCF 1.8 word for word in its own chapter 1.8. The Particular Baptist signatories of the 1689 wanted to make clear that they affirmed the same doctrine of Scripture as the Westminster divines. The only Reformed Baptist distinctives in the 1689 are in chapter 7 (covenant theology, where 1689 Federalism develops a distinctive position) and chapters 26 and 29 (ecclesiology and baptism). On Scripture itself, the 1689 is verbally identical with the Westminster.

This matters for contemporary Reformed Baptists who sometimes meet King-James-Only voices claiming to speak for confessional Reformed Baptist theology. The 1689 itself does not endorse KJV-onlyism; it endorses the immediately inspired and providentially preserved Hebrew and Greek originals as authentical. The KJV is a fine and venerable translation; it is not the authentical text. The confession's authentical text is the underlying Hebrew of the OT and Greek of the NT, providentially preserved across the centuries and engaged through careful textual study.

13.4 Heidelberg Catechism, Helvetic Consensus

The Heidelberg Catechism (1563), the third of the Three Forms of Unity, does not treat the doctrine of Scripture as such (it is organised around the Apostles' Creed, the Decalogue, and the Lord's Prayer, with Scripture as the foundational authority throughout). The Heidelberg's doctrine of Scripture is therefore implicit but pervasive.

One Reformed confessional document does explicitly address the question of textual preservation: the Helvetic Consensus Formula (1675), a Swiss Reformed confession associated with the Buxtorf school in Basel. The Helvetic Consensus is the most rigorous Reformed statement on the preservation of the Hebrew text, going so far as to affirm the inspired status of the Hebrew vowel points and the consonantal text alike. This is a more aggressive claim than WCF 1.8 makes; the Helvetic Consensus reflects a specific polemical situation (against Roman Catholic appeals to the Vulgate and against the textual scepticism of Louis Cappel) and has been understood by most subsequent Reformed theologians as overshooting the actual claim of the confessional tradition. The standard contemporary Reformed view, following WCF 1.8 rather than the Helvetic Consensus, distinguishes the inspired consonantal text from the later (but providentially adequate) Masoretic vowel pointing.

13.5 Why the Confessional Position Is Not KJV-Onlyism

The confessional Reformed doctrine of providential preservation has sometimes been conflated with King-James-Only theology. The conflation is a mistake on at least three grounds.

First, the confessions identify the inspired text as the original-language text, not a translation. The Westminster, the 1689, and the Belgic all explicitly distinguish the inspired Hebrew and Greek from translations, however good. The KJV is a translation. To elevate it to authentical status is to do what the Reformed confessions explicitly refused to do for the Vulgate.

Second, the underlying texts of the KJV are not themselves the universally agreed Reformed standard. The KJV NT is translated from the textus receptus (the printed Greek text of Erasmus and his successors); the textus receptus is one printed edition of the Greek manuscript tradition, with its own textual decisions, some of which (e.g., the Comma Johanneum at 1 John 5:7) are virtually universally regarded as inauthentic. The KJV OT is translated from the Ben Chayyim Second Rabbinic Bible; this is a fine medieval rabbinic edition, but it is not the inspired autograph either. The "providentially preserved text" the confessions point to is not any single printed edition; it is the textual tradition as a whole.

Third, the doctrinal claim of preservation is consistent with the existence of small variants across manuscripts. The confessions do not claim that every manuscript is perfect; they claim that the textual content of the autographs has been substantially preserved across the ages. This is precisely what the manuscript evidence shows. The KJV-only argument has typically assumed a more rigid claim — that one printed text is verbally identical with the inspired autograph — and then tried to ground that claim in the Reformed confessions. The confessions do not actually make the claim.

The Reformed evangelical scholar can therefore affirm the Westminster confessional position on Scripture with full confidence, while using BHS, BHQ, and the modern English translations descended from them, without any inconsistency. The confession describes a textual situation that the modern critical editions in fact reflect: an inspired Hebrew and Greek original, providentially preserved across the manuscript tradition, accessible to the modern church through careful textual scholarship.

Common Misunderstandings

"Be diligent to present yourself approved to God, a worker who does not need to be ashamed, rightly handling the word of truth." — 2 Timothy 2:15

14.1 The "Evolving Bible" Claim

The popular and sometimes academic claim that the OT "evolved" through such heavy redaction and editing that the "original" is irrecoverable has roots in nineteenth-century Wellhausenian source criticism and continues in some twenty-first-century postmodern biblical scholarship. The claim usually appears in popular books and documentaries: the Bible has been changed so often by editors that we can never know what it originally said.

The kernel of truth: some OT books did go through stages of compilation and editing. Proverbs explicitly includes collections from different periods and editors (Prov 1:1; 10:1; 25:1; 30:1; 31:1). Jeremiah and Baruch produced and re-produced scrolls of Jeremiah's oracles (Jer 36). The historical books include sources (the Book of Jashar, the Book of the Wars of YHWH, the records of the kings of Judah and Israel). Recognising the literary and historical processes of OT composition is part of careful exegesis.

The fatal overreach: the manuscript evidence shows that whatever editorial processes lay behind the canonical OT books, the canonical text was substantially stable from at least the second century BC (the DSS evidence). The proto-Masoretic tradition dominates that evidence; the LXX, the SP, and the Targums confirm it; the Masoretic transmission of the next thousand years was extraordinarily faithful. The picture of an OT being "rewritten in every generation" is not supported by the documentary evidence. The OT we have is substantially the OT of the second century BC, which is substantially the OT of the apostles, which is substantially the OT the Hebrew authors wrote.

14.2 The Marcionite Use of Textual Issues

Marcion of Sinope (c. AD 85–160) was a second-century heretic who taught that the God of the OT was a different and inferior deity from the God of the NT, that the OT should be rejected entirely by Christians, and that the only legitimate Christian Bible was a shortened Luke and ten of Paul's letters (purged of OT references). The church rejected Marcion's teaching definitively; the canonical recognition of the four-Gospel + thirteen-Pauline + general-letters NT was in part a direct response to Marcion's pruning.

Marcionism has had a long afterlife. In every generation some Christian voice has wanted to demote the OT relative to the NT — sometimes by appealing to textual difficulties ("the OT is so corrupt that we can hardly trust it; let's focus on the NT"), sometimes by appealing to ethical difficulties (the conquest, the imprecatory psalms), sometimes by simply preferring the comfortable Jesus of selective NT reading to the demanding God of the whole canon. The contemporary form — sometimes called "soft Marcionism" — appears in popular preaching that treats the OT as a kind of cultural background to be tolerated rather than as the Christian Scripture it is.

The textual data this page surveys is the answer to soft Marcionism. The OT text is substantially what the apostles read and what Jesus identified as Scripture. The textual difficulties are real but modest; they do not warrant demoting the OT to background; they certainly do not warrant Marcion's stripped-down Christianity. The Reformed tradition has been emphatic on this point. Calvin's Institutes, book 2, chapter 10 ("The Similarity of the Old and New Testaments") and chapter 11 ("The Difference between the Two Testaments") is the classic Reformed statement: the OT and NT are one in substance, both centred on Christ, both authoritative for Christian faith and life.

14.3 The KJV-Only / Textus-Receptus-Only Claim

Some confessional Reformed brothers and sisters hold to a strong KJV-only or textus-receptus-only position, arguing that the Reformed doctrine of preservation entails the providential perfection of a particular printed text (the Ben Chayyim for the OT, the textus receptus for the NT). This page has already addressed this at length (Section 13.5). The summary points: the confessions identify originals, not printed editions; the underlying texts of the KJV are themselves the product of textual-critical decisions; the doctrine of preservation does not require numerical identity of every manuscript. The KJV is a fine translation and a venerable historical artefact; it is not the authentical text the confessions identify.

14.4 The "Maximalist" Denial of Variants

The opposite error to KJV-onlyism is the popular evangelical denial that any variants exist at all. The pastor who tells his congregation that "the Hebrew text has been perfectly preserved without a single variant for 3,500 years" has overstated the case. Variants exist; the textual apparatus catalogues them; the manuscript witnesses sometimes differ. The Reformed evangelical doctrine of preservation does not require denying the existence of variants; it interprets the variants within the larger picture of providential preservation.

This kind of denial typically arises from a defensive posture — a desire to protect the doctrine of Scripture from sceptical attack. But the defence is unnecessary and ultimately counterproductive. The textual evidence is genuinely strong; the doctrine is fully consistent with the evidence; the Christian student does not need to pretend that variants do not exist. He or she needs to understand what the variants actually show — which is precisely what this page has set out.

14.5 The "Apocrypha Was in the LXX" Argument

Roman Catholic and Eastern Orthodox apologists sometimes argue that because the LXX manuscripts include the deuterocanonical / Apocryphal books (Tobit, Judith, Wisdom, Sirach, 1–2 Maccabees, and so forth), and because the LXX was the apostles' Bible, the Apocrypha should be regarded as canonical. The argument is exegetically and historically defective.

The LXX as actually preserved is a Greek translation library, including translations of the canonical Hebrew Bible and additional Hellenistic Jewish writings. The boundaries of the library varied across manuscripts and editions; no single fixed LXX canon existed in the apostolic period. Jewish authorities (Josephus, Against Apion 1.37–43; the Babylonian Talmud, Bava Batra 14b) consistently distinguish the canonical Hebrew Bible (twenty-two or twenty-four books, identical with the Reformed thirty-nine) from "outside" books. The NT writers cite the canonical books as Scripture; they do not cite the Apocrypha as Scripture (Jude's reference to Enoch is allusion, not canonical citation, and the Apocrypha proper is never cited at all). The early Christian church largely followed Jewish canonical limits; Augustine and a few others broadened the canon, but Jerome (translating the Vulgate) explicitly distinguished the canonical Hebrew from the apocryphal Greek writings.

The Reformed position, articulated in WCF 1.3 and Belgic 6, is that the Apocrypha are not canonical. The canon is the Hebrew Bible — the twenty-four books of the Tanakh, divided into thirty-nine in Christian counting — the canon Jesus and the apostles received. This is consistent with the actual usage of the LXX in the NT and with the explicit testimony of the early Jewish canonical limits.

Practical Implications for the Preacher and Reader

"Until I come, devote yourself to the public reading of Scripture, to exhortation, to teaching." — 1 Timothy 4:13

15.1 Choosing an English Translation

The English-reading Christian's daily contact with the OT is through translation. The major contemporary translations — ESV, NASB, NIV, NKJV, CSB, NRSV — all rest on substantially the same Hebrew text (the Leningrad Codex of the MT, with attention to alternate readings recorded in the apparatus) and they all reflect responsible textual-critical decisions. The differences among them on OT textual variants are small — minor variations in which footnote alternatives are adopted, whether to follow the LXX or the MT at a few contested points, whether to restore a missing word (as at Gen 4:8 or 1 Sam 13:1) or to leave the MT difficulty visible.

For most Reformed evangelical readers, any of the major modern translations is a sound choice. The ESV and the NASB tend toward formal equivalence (closer to the Hebrew word order); the NIV and CSB tend toward more idiomatic English; the NKJV retains the Ben Chayyim Hebrew base for the OT and offers a more traditional reading experience. For the textually careful reader, the ESV with its full footnote apparatus is probably the best single translation; the NIV is more readable and equally responsible.

15.2 Preaching from a Text with a Textual Question

The preacher who comes to a passage with a significant textual question faces a pastoral judgement. Three principles are worth observing.

First, do not introduce textual issues unnecessarily. Most Sunday morning congregations are not equipped to follow a detailed discussion of the Hebrew Vorlage of Psalm 22:16, and the sermon's exegetical substance does not normally require it. The preacher uses the English text in front of his congregation and preaches what it says. If the congregation's translation differs from the preacher's preferred reading on a textual question, the preacher can briefly note the difference (a footnote in a Wednesday Bible study, perhaps) but should not allow the textual question to derail the sermon.

Second, address textual issues honestly when they cannot be avoided. Some passages — Goliath's height, 1 Samuel 13:1, the long ending of Mark, the Comma Johanneum, certain Christological prooftexts where the variant affects the apologetic — are sufficiently well known that congregations will eventually encounter them through internet sources, sceptical literature, or comparative reading of translations. When the issue cannot be ducked, the preacher should engage it directly, calmly, and confidently: the variants are real, the doctrine is unaffected, the text we have is the text God has providentially kept.

Third, never overstate the case. The preacher who claims that "every word of the Hebrew Bible has been perfectly preserved with no variation across three thousand years" has overstated; when a thoughtful congregant later encounters the actual variants, the preacher's overstatement undermines his credibility on the doctrine of Scripture as a whole. The honest, measured Reformed claim is the credible one: the OT text is substantially stable, providentially preserved, recognisably the text of the inspired authors, with a small percentage of genuine variants that the science of textual criticism engages.

15.3 The Pastor's Use of the Hebrew

Every Reformed pastor was once required to study Hebrew in seminary. Many have let that Hebrew rust. The minimum the Reformed pastor should retain is the ability to read a Hebrew word with the help of a lexicon, to look up an apparatus note in BHS, and to consult a careful Hebrew-aware commentary (Calvin, Keil & Delitzsch, the NICOT series, the Tyndale OT Commentary series, the Pillar OT, the New Studies in Biblical Theology series). A pastor who can do these three things has full access to the textual data when it is needed and can preach the Hebrew Bible faithfully without becoming a textual specialist.

The seminary student preparing for OT ministry should aim higher: a working command of Biblical Hebrew, a familiarity with the basic Aramaic of Daniel and Ezra, an ability to read BHS apparatus without constant reference to a key, and a working knowledge of LXX Greek for the major variants. The forthcoming Biblical Hebrew page in this OT pillar forthcoming will provide a structured pathway for the student starting from scratch.

15.4 For the New Believer

The new believer reading this page may be alarmed by the talk of variants and textual questions. Three things to remember.

(1) The OT in your English Bible is the OT Jesus had. The substantial content of your OT was the substantial content of the OT in the synagogues of first-century Palestine, where Jesus heard the Scriptures read, taught from them, and identified them as the Word of God. When you read Genesis or Exodus or Isaiah or the Psalms, you are reading what Jesus read.

(2) The variants are small and do not threaten the gospel. No major Christian doctrine depends on a disputed reading. The deity of Christ, the atonement, the resurrection, justification, sanctification, the second coming, the Trinity — all are taught across the canon, in passages whose text is secure. You can build your whole Christian theology on the OT text in any responsible modern translation, and you will arrive at orthodox Christianity.

(3) The doctrine of Scripture is well grounded. The inspiration of Scripture (the autograph was God-breathed) and the providential preservation of Scripture (God has kept the text substantially intact across the manuscript transmission) are doctrines fully consistent with the manuscript evidence. The OT is what the Reformed confessions say it is: the inspired Word of God, providentially preserved, sufficient for faith and life.

Bibliography & Further Reading

The works below represent the standard textual-critical scholarship on the Hebrew Bible together with the Reformed confessional and historical works that bear on the doctrine of preservation. Organised by category.

The Indispensable Reference Work

Tov, Emanuel. Textual Criticism of the Hebrew Bible. 4th rev. ed. Fortress Press, 2022. The standard reference. Tov is Jewish, not Christian, but his data, analysis, and judgements set the agenda of the field. Indispensable. The Reformed reader engages it critically on a few specific judgements but recognises its authority on the data.

Major Introductions for Students

Würthwein, Ernst, and Alexander Achilles Fischer. The Text of the Old Testament: An Introduction to the Biblia Hebraica. 3rd English ed. Eerdmans, 2014. The classic Protestant introduction, originally published in German in 1952 and substantially revised by Fischer. Written for students; clear, careful, and confessionally compatible.

Brotzman, Ellis R., and Eric J. Tully. Old Testament Textual Criticism: A Practical Introduction. 2nd ed. Baker Academic, 2016. The standard evangelical textbook. Tully's revision brings the work up to date with DSS and BHQ scholarship. Highly recommended for the Reformed seminarian.

Wegner, Paul D. A Student's Guide to Textual Criticism of the Bible: Its History, Methods, and Results. IVP Academic, 2006. The accessible entry-level introduction, treating both OT and NT. Useful for the lay reader or first-year seminarian.

McCarter, P. Kyle, Jr. Textual Criticism: Recovering the Text of the Hebrew Bible. Guides to Biblical Scholarship. Fortress, 1986. Older but still useful; very compact.

The Masoretic Tradition

Yeivin, Israel. Introduction to the Tiberian Masorah. Trans. and ed. E. J. Revell. Society of Biblical Literature, 1980. The standard scholarly introduction to the Masoretic apparatus, the masora magna and parva, vocalisation, and accentuation. Technical but essential for serious study.

Khan, Geoffrey. A Short Introduction to the Tiberian Masoretic Bible and its Reading Tradition. 2nd ed. Gorgias Press, 2013. A more accessible introduction to the Tiberian tradition than Yeivin.

Kelley, Page H., Daniel S. Mynatt, and Timothy G. Crawford. The Masorah of Biblia Hebraica Stuttgartensia: Introduction and Annotated Glossary. Eerdmans, 1998. A working guide to reading the Masora in BHS.

Andersen, Francis I., and A. Dean Forbes. Spelling in the Hebrew Bible. Biblica et Orientalia 41. Pontifical Biblical Institute, 1986. The major statistical analysis of orthographic patterns in the MT; one of several major Andersen–Forbes contributions.

The Dead Sea Scrolls

Tov, Emanuel. Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert. STDJ 54. Brill, 2004. Tov's detailed analysis of DSS scribal habits. Technical but invaluable.

Ulrich, Eugene. The Dead Sea Scrolls and the Origins of the Bible. Eerdmans, 1999. The major one-volume treatment of the biblical scrolls and their bearing on the formation of the OT text.

VanderKam, James, and Peter Flint. The Meaning of the Dead Sea Scrolls: Their Significance for Understanding the Bible, Judaism, Jesus, and Christianity. HarperOne, 2002. The accessible standard introduction.

Abegg, Martin, Peter Flint, and Eugene Ulrich, eds. The Dead Sea Scrolls Bible: The Oldest Known Bible Translated for the First Time into English. HarperOne, 1999. An English translation of the biblical DSS, with apparatus indicating divergences from the MT.

The Septuagint

Jobes, Karen H., and Moisés Silva. Invitation to the Septuagint. 2nd ed. Baker Academic, 2015. The standard evangelical introduction. Highly recommended.

Dines, Jennifer M. The Septuagint. T&T Clark, 2004. Compact academic introduction.

Pietersma, Albert, and Benjamin G. Wright, eds. A New English Translation of the Septuagint (NETS). Oxford, 2007. The standard English translation of the LXX, with introductions and notes. Available free online.

Aitken, James K., ed. T&T Clark Companion to the Septuagint. T&T Clark, 2015. The standard reference companion to LXX scholarship, by book.

Hengel, Martin. The Septuagint as Christian Scripture: Its Prehistory and the Problem of Its Canon. T&T Clark, 2002. Hengel's careful argument for taking the LXX seriously as Christian Scripture without overstating the case.

The Samaritan Pentateuch and the Other Witnesses

Anderson, Robert T., and Terry Giles. The Samaritan Pentateuch: An Introduction to its Origin, History, and Significance for Biblical Studies. SBL, 2012. The standard introduction.

Flesher, Paul V. M., and Bruce Chilton. The Targums: A Critical Introduction. Baylor University Press, 2011. The standard introduction to the Targums.

Weitzman, M. P. The Syriac Version of the Old Testament: An Introduction. Cambridge University Press, 1999. The major scholarly treatment of the Peshitta OT.

Kedar-Kopfstein, B. The Vocabulary of the Septuagint Translation of the Old Testament. (Various works; see bibliography in Würthwein-Fischer.)

Critical Editions

Biblia Hebraica Stuttgartensia (BHS). Ed. Karl Elliger and Wilhelm Rudolph. Deutsche Bibelgesellschaft, 1977. The standard scholarly Hebrew Bible. Multiple printings; available in compact, standard, and study formats.

Biblia Hebraica Quinta (BHQ). Ed. Adrian Schenker et al. Deutsche Bibelgesellschaft, 2004–. In-progress fascicles. As of 2025 the available books include Genesis (Ronald Hendel forthcoming), Deuteronomy, Judges, Ezra-Nehemiah, Proverbs, Ruth, Esther, Lamentations, Megilloth, the Twelve Minor Prophets, and others.

The Hebrew University Bible (HUB). Magnes Press, 1965–. Diplomatic edition of the Aleppo Codex. Isaiah, Jeremiah, Ezekiel published so far. Very full apparatus.

The Hebrew Bible: A Critical Edition (HBCE), formerly the Oxford Hebrew Bible. General editor Ronald Hendel. SBL Press, 2015–. Eclectic edition. Volumes appearing as available.

The Reformed Doctrine of Scripture

Westminster Confession of Faith (1646). Chapter 1, especially 1.8. The locus classicus. Available freely online.

Second London Baptist Confession (1689). Chapter 1, especially 1.8. Reproduces WCF 1 essentially verbatim.

Belgic Confession (1561). Articles 3–7.

Helvetic Consensus Formula (1675). The most aggressive Reformed statement on Hebrew preservation; engaged critically by most subsequent Reformed theologians.

Turretin, Francis. Institutes of Elenctic Theology. Vol. 1 (1679). Trans. George Musgrave Giger; ed. James T. Dennison Jr. P&R, 1992. Topic 2 ("The Holy Scriptures"), questions 4–11. The major post-Reformation Reformed treatment of textual preservation.

Owen, John. Of the Divine Original, Authority, Self-Evidencing Light, and Power of the Scriptures (1659) and Of the Integrity and Purity of the Hebrew and Greek Text of the Scripture (1659). In The Works of John Owen, vol. 16 (Banner of Truth ed.). Owen's polemic against Brian Walton's London Polyglot is the seventeenth-century Reformed engagement with textual criticism.

Muller, Richard A. Post-Reformation Reformed Dogmatics: Volume 2 — Holy Scripture: The Cognitive Foundation of Theology. 2nd ed. Baker Academic, 2003. The standard historical-theological study of the post-Reformation Reformed doctrine of Scripture, including its textual-critical engagements.

Contemporary Reformed Engagements

Beckwith, Roger T. The Old Testament Canon of the New Testament Church and its Background in Early Judaism. Eerdmans, 1985. The major Protestant treatment of the OT canon, with extensive bearing on the textual question.

Bruce, F. F. The Books and the Parchments: How We Got Our English Bible. Rev. ed. Revell, 1984 (originally 1950). Older but still a fine evangelical introduction.

Geisler, Norman L., and William E. Nix. A General Introduction to the Bible. Rev. ed. Moody, 1986. Substantial chapters on OT and NT textual criticism from an evangelical perspective.

Williams, Peter J. Can We Trust the Gospels? Crossway, 2018. A primarily NT-focused defence with relevant principles applicable to the OT side.

Carson, D. A. The King James Version Debate: A Plea for Realism. Baker, 1979. Engages the KJV-only controversy on the NT side, with principles transferable to the OT.

The Manuscripts Themselves (Facsimiles and Digital Resources)

The Aleppo Codex. Available online at aleppocodex.org. Full digital facsimile with searchable text.

The Leningrad Codex. Facsimile edited by Astrid Beck. Eerdmans, 1998. Also available digitally through the Westminster Leningrad Codex project.

The Dead Sea Scrolls. Digital images and transcriptions at www.deadseascrolls.org.il (Israel Museum) and the Leon Levy Dead Sea Scrolls Digital Library.

The Westminster Leningrad Codex. The digital scholarly edition of L used in most Bible software (Logos, Accordance, BibleWorks). Free at www.tanach.us.

Engaged but Not Fully Agreed With

Hendel, Ronald, ed. Hebrew Bible: A Critical Edition. SBL, 2015–. As above. The eclectic-edition methodology is more conjectural than the Reformed position prefers, but the volumes are important and worth engaging.

Schniedewind, William M. How the Bible Became a Book: The Textualization of Ancient Israel. Cambridge University Press, 2004. A historical-critical reconstruction of the OT's literary history. Read with discernment.

Carr, David M. The Formation of the Hebrew Bible: A New Reconstruction. Oxford, 2011. Another major historical-critical reconstruction. Read critically.