Which Bible Are You Reading

No autograph survives. The letter Tertius took down at Paul's dictation (Romans 16:22), the scroll that first carried Isaiah's words, the papyrus John's own hand may have touched: none of it exists. What exists are copies, and copies of copies, running from small early fragments to complete medieval volumes, conventionally grouped into four categories for the New Testament, sorted by writing material and script rather than by text type: papyri, uncials, minuscules, and lectionaries.

Papyri are the earliest and usually the most fragmentary, written on the reed paper of Egypt, where the climate happens to preserve organic material that would have rotted almost anywhere else. Papyrus 52, a scrap of John 18 the size of a credit card, is dated on the shape of its letters to around AD 125, making it, on that dating, among the earliest physical witnesses to any part of the New Testament, though some scholars argue the paleographic evidence allows a somewhat later range. Papyrus 66 and Papyrus 75, both containing large stretches of John and dated to around the late second or early third century, are far more substantial and among the most important papyri known.

Uncials are written in a formal capital-letter script on parchment, mostly produced from the fourth through the ninth century. Codex Sinaiticus and Codex Vaticanus, both fourth century, are the two most important nearly-complete Greek Bibles in existence, and both end Mark's Gospel at 16:8. Codex Alexandrinus, fifth century, is another major uncial witness.

Minuscules use a smaller, cursive lowercase script that came into common use from roughly the ninth century onward, and the great majority of the more than five thousand cataloged Greek New Testament manuscripts are minuscules copied well over a thousand years after the events they describe. The fourth category, lectionaries, are manuscripts that arrange scripture in the order it was read aloud in worship across the church year rather than continuously by book and chapter; several thousand are cataloged separately from continuous-text manuscripts, and while scholars weigh their readings with an eye to how a lectionary's format can itself introduce small changes, they still function as a further, independent check on the wording found elsewhere. Age and number are weighed together across all four categories rather than one substituting for the other: a reading found in one manuscript from AD 200 and a reading found in eight hundred manuscripts from AD 1200 are different kinds of evidence, not simply different amounts of the same kind.

The Old Testament has its own three witnesses. The Masoretic Text is the Hebrew text preserved and supplied with vowel points, since the Hebrew consonantal alphabet does not write vowels, by scribes called the Masoretes working roughly from the sixth to the tenth century AD; its two most important copies are the Aleppo Codex, tenth century, once complete but damaged in 1947 so that most of the Torah is now lost, and the Leningrad Codex, early eleventh century, the oldest complete copy, which underlies most printed Hebrew Bibles used today. The Dead Sea Scrolls, found near Qumran beginning in 1947, include copies of every Old Testament book except Esther and date from roughly the third century BC to the first century AD, centuries earlier than the Masoretic codices, and they mostly confirm the Masoretic tradition's care while also preserving some genuinely different readings. The Septuagint, a Greek translation of the Hebrew scriptures begun around the third century BC in Alexandria, is quoted by New Testament writers more often than the Hebrew text itself and occasionally preserves an older Hebrew reading than the one the Masoretes later fixed.

Fourteen centuries of copying by hand, before any printing press existed, guarantees variation. Given how many hands were involved, the real surprise is how stable the text is, not that it varies at all.

Most variants are accidents, and the accidents follow patterns a trained eye learns to spot. A scribe's eye can skip from one word to an identical word later in the line, dropping everything between (called homoioteleuton, 'similar ending'); a line can be copied twice by mistake (dittography) or a repeated word dropped once by mistake (haplography); a scribe copying from someone reading the text aloud can mishear one word for another that sounds alike; and in Hebrew, two letters that look nearly identical, such as the small hook of a yod and the slightly longer stroke of a vav, can be swapped, which may be the relationship between the Masoretic Text's 'like a lion' (ka'ari) at Psalm 22:16 and 'they pierced' or 'they dug' (ka'aru), the reading of the Septuagint and of 5/6HevPs, a Hebrew Psalms scroll from the Cave of Letters at Nahal Hever dated to the first century AD. That scroll is not from Qumran, and it is not the earliest Hebrew witness to this verse either; Qumran's own Psalms manuscripts run from the second to first century BC, and 4Q88 covers this same stretch of the psalm. Which direction the yod-vav swap ran here is itself an open question worth naming rather than settling by assertion: it is genuinely contested among Hebraists which reading is the original and which the corruption. The BSB, the translation this course quotes, takes a side: it prints 'they have pierced my hands and feet' and puts the Masoretic 'like a lion' in a footnote. Its footnote credits 'DSS', which here covers the Nahal Hever scroll, and says 'most MT', since a few Masoretic manuscripts also read ka'aru. That is a committee's judgement on a disputed point, not a settled fact, and you are free to weigh it differently.

Other variants are intentional, and not usually dishonest. A scribe harmonizing one Gospel's wording to match a familiar parallel, filling in a subject an ambiguous pronoun seemed to need, smoothing a theologically awkward statement, or copying a note a previous scribe had written in the margin as though it were part of the text itself, all produce real changes without any intent to deceive. Textual critics weigh exactly this pattern when they judge between two readings: a scribe is far more likely to turn a hard reading into an easier one than the reverse, so critics often give real weight to the more difficult reading as more likely original, a principle usually cited by its Latin tag, lectio difficilior potior, 'the harder reading is the stronger one.' Psalm 22:16 above is itself a live test of that rule rather than a settled example of it: the Masoretic 'like a lion at my hands and feet', as the BSB footnote renders it, is arguably the harder, stranger reading of the two, which is exactly why a number of Hebraists judge the Masoretic Text original here and the versions' 'pierced' the later, easier smoothing, against the more common assumption that runs the other way. The rule is a rule of thumb, not a law; it can point in more than one direction depending on which reading is judged harder, and it is weighed alongside the manuscript evidence rather than used alone.

Set against the number of manuscripts in existence, the number of places where wording is genuinely in question is small, and the number of those that affect anything more than spelling, word order, or a synonym is smaller still. No identified variant, once weighed, has been found to unsettle a doctrine that does not also stand on other, undisputed texts. That is a fair and checkable claim, not a reassurance offered instead of evidence, and the rest of this lesson tests it against the four most famous cases.

Four examples come up often enough that a seminary-level student should be able to state each one accurately rather than repeat a rumor about it in either direction.

Mark 16:9-20 is absent from Codex Sinaiticus and Codex Vaticanus, the two oldest complete Greek witnesses to Mark, and some early readers, including Eusebius and Jerome, note that the most accurate copies known to them lacked it. A different, much shorter ending appears in a handful of other manuscripts. The vocabulary and style of verses 9-20 differ measurably from the rest of Mark, which points toward a different hand finishing a Gospel that many scholars think originally ended, deliberately or not, on the women's fear in verse 8. Against that, the passage is very old in its own right, quoted already by Irenaeus in the late second century, and it is what the great majority of later Greek manuscripts print. What is genuinely at stake is whether Mark's Gospel first ended at verse 8 or continued with resurrection appearances, and whether the specific signs listed there, including picking up serpents and drinking poison without harm, carry apostolic authority; some fringe groups have built entire practices on those verses. What is not at stake is the resurrection itself, which the passage was never the only witness to.

John 7:53-8:11, the story of the woman caught in adultery, is missing from the earliest Greek manuscripts, including Sinaiticus, Vaticanus, and the early papyri that cover this section of John, and from the earliest commentaries on John's Gospel verse by verse. Where later manuscripts do include it, they do not agree on where: some place it here, some after John 7:36, some at the end of John, and a few attach it to Luke instead, and its vocabulary differs somewhat from John's usual style. All of that points away from its having stood originally at this place in John. And yet the story itself appears to be an old and independent tradition about Jesus, referred to as early as the third century in the Didascalia Apostolorum, and most textual critics today regard it as authentic material about Jesus even while doubting it was originally written here. What is at stake is whether this exact wording, in this exact place, is what John wrote. What is not at stake is anything in Christian teaching about grace and judgment, which does not depend on this passage alone and is taught elsewhere, for instance in Luke 7:36-50.

The Comma Johanneum, at 1 John 5:7-8, is the clause found in the King James Version: 'For there are three that bear record in heaven, the Father, the Word, and the Holy Ghost: and these three are one. And there are three that bear witness in earth,' against which the BSB reads simply, 'For there are three that testify: the Spirit, the water, and the blood - and these three are in agreement.' The added words naming the Father, the Word, and the Holy Spirit are absent from every known Greek manuscript before roughly the fourteenth century, and, tellingly, are never quoted by any Greek-speaking church father during the fourth-century Trinitarian controversies, where they would have made a decisive proof text against Arian opponents; an argument from silence, but a loud one. The clause is older in some Latin manuscripts and entered the Greek tradition later. Erasmus left it out of his first two printed editions of the Greek New Testament, in 1516 and 1519, then included it in his third edition of 1522. He is often said to have promised to include it if a single Greek manuscript could be produced containing it, and one duly appeared, Codex Montfortianus; but the study that traced this story most carefully, by H. J. de Jonge in 1980, found no evidence Erasmus ever made such a promise, so 'reportedly' is the right word for it, and the popular idea that Montfortianus was produced specifically to win a wager rests on that same unproven story. What is well supported independently is narrower: Montfortianus, also catalogued as minuscule 61, dates to around 1520, and its text of the comma reads like Greek translated out of the Latin Vulgate rather than an independent Greek witness. What is at stake is one proof text for the Trinity. What is not at stake is the doctrine of the Trinity, which never depended on this verse and is argued from texts including Matthew 28:19 and 2 Corinthians 13:14.

The Old Testament case worth knowing is Deuteronomy 32:8. The Masoretic Text reads that the Most High fixed the borders of the nations 'according to the number of the sons of Israel'; the Dead Sea Scrolls and the Septuagint read 'according to the number of the sons of God,' rendered in the Septuagint as 'angels of God.' The Qumran and Greek reading pictures God assigning the nations to members of his heavenly court while keeping Israel for himself in verse 9, a picture consistent with the assumption of a divine council found elsewhere in the Old Testament, for instance Psalm 82 and Job 1-2, and on the majority view a later scribe found it uncomfortable enough to revise toward something safer. The BSB, the translation this course quotes, takes that side: its main text reads 'sons of God' and the Masoretic wording goes into a footnote, as the ESV and several other modern translations also do. A committee makes that choice because the reading is now attested in Hebrew at Qumran as well as in Greek, and because it is the harder reading, the one a scribe would be likelier to change than to invent. The Masoretic reading still has serious defenders, and their case should be heard at full strength. It is the reading of the received Hebrew text, which the translations of a century ago printed without hesitation, and it makes coherent sense on its own: the seventy nations listed in Genesis 10 match the seventy members of Jacob's family who went down to Egypt (Genesis 46:27; Exodus 1:5), a correspondence Jewish interpreters drew, so that God laid out the world with Israel already in view. What is at stake is a detail of how one verse pictures divine governance of the nations. What is not at stake is monotheism, which this same book states without ambiguity elsewhere, including in the very next line: 'But the LORD's portion is His people, Jacob His allotted inheritance' (32:9).

The critical text position, sometimes called reasoned eclecticism, underlies editions such as the Nestle-Aland Greek New Testament and the United Bible Societies edition, which nearly all modern translations besides the King James and New King James use as their base text. It weighs each variant on its own evidence rather than defaulting to any one manuscript or family, giving real but not exclusive weight to the earliest and most geographically diverse manuscripts, on the reasoning that an error introduced early has had less time to spread uncorrected, and that agreement across independent regions is less likely to be coincidental than agreement within one copying stream. Westcott and Hort pioneered a version of this approach in the nineteenth century, leaning heavily on Vaticanus and Sinaiticus as a 'neutral text'; the field has revised a great many of their specific judgments since, so a modern critical edition is a distant descendant of their method rather than a repetition of it.

The majority text position, defended by textual critics including Zane Hodges, Arthur Farstad, Wilbur Pickering, and more recently Maurice Robinson, argues that the sheer numerical dominance of the Byzantine text type, roughly eighty to ninety percent of all surviving Greek New Testament manuscripts, is itself strong evidence rather than something to be explained away. Its defenders reason that a text in wide, repeated, geographically spread use across many centuries is more likely to preserve the original than one surviving mainly in a small number of early copies concentrated in Egypt, a survival pattern shaped as much by that region's dry climate as by anything about the quality of the text, and they note that scribes tended to conflate and expand readings rather than shorten them, which raises a real question about why the shortest reading should always be assumed earliest.

The critical text position has a standard answer to that numerical argument, and it deserves to be stated as fully as the argument it answers: Byzantine numerical dominance may simply reflect the size and continuity of the Greek-speaking scribal culture centered on Constantinople, which kept producing fresh copies of its own local text for roughly a thousand years after the Western and Egyptian churches had largely shifted to Latin and Coptic and stopped copying Greek manuscripts in the same volume. On this argument, a later manuscript count measures how long a text type had a living, copying community behind it, not how early or accurate its readings are. The majority position carries a named weakness of its own in return: the distinctively Byzantine form of the text is essentially absent from the papyri and from surviving patristic quotations before the fourth century, a gap its defenders have to explain if that text type were both original and merely under-represented in the earliest surviving evidence, rather than a later standardization that grew from earlier, more varied local texts.

The majority text position should be distinguished from King James Only advocacy, even though the two overlap in practice. Majority text scholars are working textual critics making an evidence-based argument about manuscripts; some King James Only advocates go further, treating the King James Version itself, or the Textus Receptus that underlies it, as specially and uniquely preserved by God in a way no other text or translation is, which is a theological claim about one sixteenth and seventeenth-century printed edition rather than a text-critical argument, and most majority text scholars explicitly reject it as a separate and further claim beyond their own.

A fair student should hold several things at once. Every mainstream position, majority or critical, agrees that the historic creeds and the consensus doctrines of the church are unaffected by how this debate is resolved. The two camps differ across roughly 6,500 variant units between the Nestle-Aland text and Robinson and Pierpont's Byzantine Textform, the large majority of them spelling or word-order differences invisible in translation, with the number that affect translatable meaning usually put in the low thousands rather than the single 'few hundred' a casual comparison might suggest. And neither side's current institutional standing should be mistaken for the argument being closed: the critical text dominates seminary training and modern translation committees, while credentialed, publishing textual critics continue to defend the majority position anyway, each side holding real evidence, continuous Byzantine transmission on one hand, early and geographically scattered papyri and uncials on the other, that the other side has to account for rather than dismiss.

Open any modern study or reference Bible and the small bracketed notes fall into a few recognizable types. A note beginning 'Some manuscripts' or 'Some manuscripts add' or 'Some manuscripts omit' is a text-critical footnote, reporting that the manuscripts themselves disagree at this point, which is the kind attached to Mark 16:9-20, John 7:53-8:11, and 1 John 5:7-8. A note beginning simply 'Or' is a translation footnote instead, reporting that the underlying text is not in dispute but that its wording honestly allows more than one rendering into English. Learning to tell these two kinds apart is most of the skill.

Deuteronomy 32:8 is worth walking through a second time as a model, because it does something unusual. The BSB footnote at verse 8 reads, in its own shorthand, 'DSS; LXX according to the number of the angels of God; MT according to the number of the sons of Israel.' Unpack the abbreviations: DSS is the Dead Sea Scrolls, LXX the Septuagint, MT the Masoretic Text. That single line tells you the committee followed a Hebrew scroll from Qumran, noted that the Greek tradition reads 'angels of God', weighed both against the Masoretic Text's usual authority, and printed the non-Masoretic reading in the main text anyway. Seeing a translation committee depart from its normal base text, and say so plainly in the footnote, is a better lesson in how this discipline actually works than any abstract description of method.

In practice: note the reference; read the footnote itself, since it will usually name which manuscripts or ancient versions read what, in brief scholarly shorthand; and, if you want to go further, open the same passage in Blue Letter Bible or STEP Bible, both free, and look at what either tool's textual notes or parallel manuscripts say. Comparing a King James footnote against an ESV or NASB footnote at the same verse is instructive on its own, since the King James, built on the Textus Receptus, will sometimes flag a different question entirely, or flag nothing where a critical-text Bible flags something significant.

A footnote is not a crisis. It is the text showing its own workings, in public, where any reader can check it against the same free tools the scholars use. A Bible that hid these facts from you would be less trustworthy for having hidden them, not more trustworthy for looking simpler on the page.