How the canon formed, and why Catholic and Orthodox Bibles have more books

"And regard the patience of our Lord as salvation, just as also our dear brother Paul wrote to you, according to the wisdom given to him, speaking of these things in all his letters. Some things in these letters are hard to understand, things the ignorant and unstable twist to their own destruction, as they also do to the rest of the scriptures." (2 Peter 3:15 to 16, NET)²⁵

Read that slowly. A New Testament writer is talking about Paul's letters, admits they're hard, and then ranks them with "the rest of the scriptures". Inside the New Testament itself, one set of Christian writings is already being treated as Scripture by another. Nobody voted. Nobody convened. A letter was read in a church, kept, copied, and read again, and at some point people were calling it Scripture and arguing about what it meant.

One caveat before we go further, because you'd find it out later and wonder why I hadn't said it. 2 Peter is itself one of the books that took longest to be received, and you'll meet it below in Eusebius's disputed column. Scholars argue about when it was written. If it's late, this passage shows a second-century church treating Paul's letters as Scripture rather than a first-century one, which is precisely what one of the two scholarly models in this lesson would predict. Either way it shows the process, which is what the lesson is about. I'd rather you knew that now than found it out at the end.

That's the whole story of the canon in miniature, and it's the story this lesson tells with dates attached. You'll need it, because you're going to meet two confident claims that both get it wrong: that the Emperor Constantine or the Council of Nicaea picked the books in 325, and that the Catholic Church "added" seven books to the Old Testament. By the end you'll be able to walk the actual evidence in order, say why Catholic and Orthodox Bibles hold more books than the one this course uses, give each side's case in words its own people would accept, and tell which parts of all this are history, which are argued, and which are faith.

What a canon is, and how one actually forms

"Canon" comes from a Greek word for a measuring rod, and it came to mean the list of books a community treats as its rule.

This is the model this course works with, and I'm calling it a model because it's one of three live ones and you'll meet the other two before this lesson is over. On this account a canon isn't made by a vote. It's the shape a community's worship has already taken, written down when someone asks. Churches read certain books aloud on Sunday, copied them, taught from them, and defended them when challenged. When a bishop finally wrote a list, he was describing what his churches did, the way a librarian's catalogue describes shelves that were already full.

That predicts what the evidence should look like, and you can check the prediction as we go. If canons formed by decree, you'd expect one document, one date, and silence before it. If they formed by use, you'd expect local lists that mostly agree, a handful of books argued over for a long time, and councils that confirm rather than choose.

The other models read the same dates differently, and I'll set them out properly once you have the evidence in hand. Keep all this in your head as we go, and each time I say the councils confirmed rather than chose, notice that I'm reading the evidence through the first one.

The Jewish canon by the first century

Start with the books Christians call the Old Testament and Jews call the Tanakh, because the first Christians inherited them rather than made them. Three witnesses, in order.

Some time after 132 BC, the year a Jewish translator records arriving in Egypt to work on his grandfather's book of wisdom, he wrote a preface to it. It survives as the Prologue to Sirach, and three times it refers to a collection in three parts: things "delivered unto us by the law and the prophets, and by others that have followed their steps"; his grandfather's reading of "the law, and the prophets, and other books of our fathers"; and the difficulty of translating "the law itself, and the prophets, and the rest of the books" (Brenton).¹ Two parts have names. The third is still just "the rest of the books". That is the earliest surviving description of the collection, and it's already three-part, with the third part's edges not yet named.

Josephus, writing Against Apion around AD 95, says the Jews have "only twenty two books: which contain the records of all the past times: which are justly believed to be divine", and adds that through all those ages "no one has been so bold, as either to add any thing to them; to take any thing from them; or to make any change in them."² Read that as a proud statement from an insider rather than a neutral survey. But it tells you the collection was closed enough by the 90s for a Jew to boast about its fixed size.

The third witness is Jesus. In Luke 24:44 he tells the disciples that "everything written about me in the law of Moses and the prophets and the psalms must be fulfilled" (NET).³ Law, Prophets, and a third part named by its biggest book: the same three-part shape the Sirach prologue described, now spoken in Jerusalem around AD 30.

One more, from inside Judaism. A passage in the Babylonian Talmud, Bava Batra 14b to 15a, records a tradition giving the order of the Prophets and the Writings.⁴ A list that gives an order presupposes a working list. It does not show the list was uncontested, and it wasn't: the rabbis elsewhere record arguments over whether Ecclesiastes, the Song of Songs, and Esther "defile the hands", which is their idiom for a book's sacred standing.²⁶

So be careful how you use this section, because it's where the Protestant argument later in this lesson gets its foundation. History establishes that a three-part collection was in use, and that its core was beyond dispute, by the first century. Scholars argue about whether its edges were closed then: John Barton and Timothy Lim read the same evidence and hold that the Writings were still open, and Lee Martin McDonald goes further.²⁶ A closed first-century Hebrew canon is the premise the Protestant case at the end of this lesson rests on, so I'm telling you now that the course's own strongest argument stands on a disputed foundation.

Say the rest in the terms a Jewish reader uses, because it's their Scripture first. The Tanakh has twenty-four books in three parts: Torah, Nevi'im, Ketuvim. It's a complete Bible, read on its own terms in synagogues today, and it ends with Chronicles and the call to go up to Jerusalem, not with Malachi looking forward. The Christian Old Testament contains those same books, rearranged into four sections. What Christians did next is where the disagreement starts, and we'll come to it.

The New Testament core by about 180

The new books come next, and the evidence for them arrives fast, early, and in an odd form: someone trying to cut the collection down.

Around 144, in Rome, a wealthy shipowner from Sinope named Marcion, and by the tradition the son of that town's bishop, produced a Bible of his own: an edited Luke and ten letters of Paul, with the whole Old Testament rejected.⁵ The church at Rome excommunicated him.

What his edit proves is argued, and the argument is worth having. Most conservative historians read it as Harnack did: you can't cut down a collection that doesn't exist, and Marcion's ten letters of Paul presuppose a Pauline corpus already circulating. Hans von Campenhausen read it the other way, holding that no normative collection of new writings is attested before Marcion and that the church's lists are a response to him. Harnack himself, who argued the first case, granted part of the second: the church didn't create a canon to answer Marcion, but it may have reflected for the first time on what it already had because he forced the question.²⁷ Either way, the churches of the 140s were reading more than Luke and Paul, and Marcion is why we can date that.

Around 180, Irenaeus, bishop of Lyon, wrote Against Heresies. In book three he names the four Gospels with their traditional authors and then argues that there can be neither more nor fewer than four. His reasoning is that "since there are four zones of the world in which we live, and four principal winds", the church spread across the world should have four pillars.⁶

You may find the argument odd, and two readings of it are available. On one, the argument from the four winds is the kind of reason people give for something already fixed rather than something still open. On the other, Irenaeus argues at all because the number was being pressed: Marcion had one gospel, Tatian had woven four into one, and other groups were reading others. Either way, the four he names are the four we have, a century and a half before Nicaea, and he names them as the church's rather than as his proposal.

The third witness is the one in the picture, and it comes with an honest caveat. The Muratorian Fragment is a damaged Latin list found in a codex in the Ambrosian Library in Milan and published by Ludovico Antonio Muratori in 1740.⁷ Its opening is lost, so it begins mid-sentence; the first words we have call Luke "the third book of the Gospel", which tells you two came before it.⁷

Then it lists John, Acts, thirteen letters of Paul, Jude, two letters of John, the Wisdom of Solomon (a book Protestant Bibles don't contain, sitting in a New Testament list), and the Apocalypse of John. It mentions the Apocalypse of Peter but says "some of us are not willing that the latter be read in church."⁷ And it deals with the Shepherd of Hermas, a widely read book, like this: it "ought indeed to be read", but not publicly in church among the prophets or apostles, because Hermas wrote it "very recently, in our times", while his brother Pius was bishop of Rome.⁷

What it doesn't mention is as telling as what it does: no Hebrews, no James, no 1 or 2 Peter, and no clear third letter of John.⁷

The caveat is the date. The traditional dating is late second century, at Rome, mostly because of that "in our times" line about Pius, who was bishop in the middle of the second century; Metzger's own headnote puts the list at about 170. In 1973 Albert Sundberg argued it was really fourth-century and Eastern, and Geoffrey Hahneman built the case out in 1992.

Their argument deserves to be heard at strength: the Fragment's shape has more in common with the fourth-century Eastern lists than with anything else surviving from the West around 200, no comparable Western list exists that early, and the "in our times" line may be a phrase the compiler inherited rather than a date he was writing in. Joseph Verheyden replied in 2003 that none of those points carries the weight put on it, and Eckhard Schnabel's 2014 survey judged that the traditional date still fits the evidence better. Clare Rothschild's 2022 critical edition keeps the fourth-century case in play.⁸

So the date is genuinely open, the field leans early, and what would settle it is a second witness to this list from a datable context, which nobody has. Either way, notice what the Fragment shows: a core of roughly twenty-two books recognised early, a fringe still under discussion, and a writer giving his reasons in the open.

The edges settle in the West and in Alexandria: 200 to 400

Between Irenaeus and the end of the fourth century, the argument was about the edges: Hebrews, James, 2 Peter, 2 and 3 John, Jude, and Revelation. Three documents let you watch it settle.

Eusebius's History 3.25 puts the four Gospels, Acts, Paul's letters, 1 John, and 1 Peter among the recognised books. Among the disputed, "which are nevertheless recognized by many", he lists James, Jude, 2 Peter, and 2 and 3 John. Then a third category, the spurious: the Acts of Paul, the Shepherd, the Apocalypse of Peter, the Epistle of Barnabas, and the Teachings of the Apostles. And a fourth, which matters later: the forgeries circulated by heretics, where he puts the Gospels of Peter, Thomas, and Matthias. Of Revelation he writes that it should be placed among the recognised "if it really seem proper", and he also names it among the rejected, which is his way of saying the church hadn't decided.⁹

A note on the date, because the argument below leans on it. Eusebius brought the History to its final form around 325, but the manuscripts show he had substantially written it earlier, in editions of about 313, 316, and 324. Book 3 most likely belongs to the earliest of those, a decade before the council. That makes the evidence better rather than worse, as you'll see.⁹

Every year the bishop of Alexandria sent a letter announcing the date of Easter. Athanasius's thirty-ninth, for Easter 367, took the chance to list the books. For the Old Testament he gives "twenty-two books in number" (and we'll come back to what's in that list, because it's a trap for both sides). For the New Testament: "the four Gospels, according to Matthew, Mark, Luke, and John", then Acts, seven Catholic Epistles, "fourteen Epistles of Paul" (which includes Hebrews), and "the Revelation of John". Then the two sentences that get quoted most: "These are fountains of salvation, that they who thirst may be satisfied with the living words they contain." And: "Let no man add to these, neither let him take ought from these."¹⁰

He isn't finished. He adds that there are "other books besides these not indeed included in the Canon, but appointed by the Fathers to be read by those who newly join us": the Wisdom of Solomon, Sirach, Esther, Judith, Tobit, the Teaching of the Apostles, and the Shepherd.¹⁰

Two things a lesson mustn't get wrong about this letter. First, Athanasius doesn't create the list. Read his words: he's reporting what the churches hold and telling them not to change it. The tone is a fence around a field that's already planted. Second, look at that second list. Esther is in it, not in his canon. So the first document with our exact New Testament doesn't have our exact Old Testament. Keep that for later.

Then the councils. In 393 a regional council at Hippo in North Africa, with Augustine present, listed a canon; the Third Council of Carthage repeated it in 397, and the African canons were compiled again in 419.¹¹ The Old Testament list included Tobit, Judith, two books of Maccabees, and Wisdom and Sirach among the five books of Solomon, close to the later Catholic canon. Note what the council then did. It asked that "the church across the sea", meaning Rome, be consulted to confirm the list. That clause survives in one manuscript strand rather than in every copy, so lean on it lightly; but if it's original, a body that thought it was choosing the books wouldn't have asked someone else to check its work, and a body that thought it was confirming common practice would.¹¹

One correction to "the edges settled by 400", because it's true of this course's own tradition and not of the whole church. The settlement you've just watched is a Latin and Alexandrian one. The Peshitta, the standard Syriac New Testament from the early fifth century, has twenty-two books and omits 2 Peter, 2 and 3 John, Jude, and Revelation; those five arrived in Syriac only with the Philoxenian version of 508 and the Harklean of 616, and the Church of the East has never received Revelation into its canon at all.²⁸ So when this lesson says the edges settled in the fourth century, hold the qualifier in your head: they settled where this course's ancestors were standing.

The dated witnesses in order, with the two the argument below turns on marked by shape as well as colour. The spacing is deliberately not proportional, so read the order and not the distances: the gap between Jerome and Florence is about a thousand years and the gap between Luther and Trent is twelve.

Look at where the entries sit. The ringed mark at 367 is the first list of exactly our twenty-seven. The square at 325 is the Council of Nicaea, and read what's next to it: nothing about the canon. Everything above the thousand-year gap is a witness describing what churches already did, spread across five centuries.

The tests the church used

Nobody in the second century wrote out a checklist. But when you read the reasons given in the Muratorian Fragment, Eusebius, and Athanasius, three tests keep recurring, and F. F. Bruce's standard history of the canon draws them out.⁵ Take each one with its mechanism, because a test you can't run is a test you've only memorised.

Apostolic origin. Written by an apostle, or by someone in the apostolic circle. In practice this was a question about generations rather than handwriting: could this book have come from the people who knew Jesus, or is it demonstrably later? The Muratorian Fragment's objection to the Shepherd is exactly this and nothing else, and it's dated by a family relationship, not a theological argument.

Orthodoxy. Agreement with the faith taught everywhere. This one ran by comparison rather than inspection. A church that had been taught by its founders measured a new book against what it had already received. This is why it worked against Marcion, whose god of the Old Testament was not the Father of Jesus Christ, and why it can look circular from outside. It is worth admitting that it partly is.

Continuous use in worship, across the churches. Not "is this book good" but "have congregations, in more than one place, been reading this aloud for a long time". Nobody could survey the churches, so in practice this ran on correspondence and travel, which is why the short letters took so long: 2 and 3 John are a page each, and not every church had a copy to have an opinion about.

"Constantine chose the books at Nicaea", walked date by date

You'll hear this from friends, in documentaries, and on the internet. The version most people have heard comes from Dan Brown's The Da Vinci Code (2003), where a character declares that the Bible "was collated by the pagan Roman emperor Constantine the Great".¹² Let's walk the dated evidence in order and see whether there's room for it.

Around 180, Irenaeus has four Gospels and argues there can be no other number.⁶ That's 145 years before the council. If Constantine picked the Gospels, he picked ones that had already been the only four for a century and a half.

Eusebius sat at Nicaea, and he sorted the books in a history he had substantially written before the council, in editions of about 313 and 316, and then revised as late as 324 or 325. He puts James, Jude, 2 Peter, and 2 and 3 John among the disputed and hedges on Revelation.⁹ Here is the force of the correction: he had every opportunity, in the revision, to add "as the great council decreed" to 3.25. He didn't. He left the disputed list standing.

In 367, Athanasius, forty-two years after Nicaea, lists our twenty-seven. He was at Nicaea too, as a young deacon. If the council had listed the books, the most obvious thing for him to write would be "as the great council decreed". He writes nothing of the kind. He tells the churches not to add or subtract, as if the list is something they already have.¹⁰

In 393 and 397, Hippo and Carthage list the books, nearly seventy years after Nicaea, and ask Rome to confirm.¹¹ If Nicaea had settled it, there'd be nothing to confirm.

And Nicaea itself? We have its creed and its twenty canons, and none of the surviving records of its business mention the canon of Scripture at all.⁵ The council was about whether the Son is of the same substance as the Father, and about the date of Easter and some church discipline.

So the Nicaea story fails on the evidence. But notice something better than that. It fails as the wrong kind of story. Once you've seen how the lists were made, "an emperor chose the books" is like saying a librarian decided which books people had been reading for two hundred years. It gets the direction of causation backwards. The lists exist because the reading came first.

Now sort the claims. History establishes that the four Gospels were fixed by about 180, that Nicaea's records contain no canon business, and that the fourth-century lists name books churches were already reading. Scholars argue about the date of the Muratorian Fragment, how early the fringe books were widely received, and whether those lists confirmed a collection or constituted one. We hold by faith that these books, and not others, are God's Word. The Nicaea story isn't wrong on the faith question; it's wrong on the history, which is the one part anyone can check.

What the disagreement is really about

Scholars agree on the tests and on the dates. What they argue about is what the process was, and there are three positions rather than two, because the two usually printed in Protestant textbooks are both Protestant-facing.

The recognition model. Bruce, and more recently Michael Kruger in Canon Revisited (2012), hold that the church recognised a canon it didn't create. Kruger's version is called self-authenticating: the books carry their own authority as God's word, God's providence guided their preservation and reception, and the Spirit brings believers to recognise them, with the historical evidence of early and wide use serving as corroboration rather than as the basis.¹³ On this view a canonical book was canonical the day it was written, and the fourth-century lists are the church catching up with a fact.

The community model. Lee Martin McDonald, in The Biblical Canon (2007) and his two-volume Formation of the Biblical Canon (2017), holds that the canon was community-determined: no early authority fixed a list, recognition was slow and local, different regions read different collections for centuries, and the fourth-century lists represent choices churches made under the pressures of their time.¹⁴ On this view "canon" in the strict sense of a closed list is a fourth-century thing, and the earlier evidence shows Scripture without yet showing canon.

The ecclesial model, which belongs to the two traditions the second half of this lesson is about. Catholic and Orthodox theology holds that a canon presupposes an authority competent to recognise one. On this account the question "which books?" can't be answered by the books themselves, because a reader has to know already which books to consult; the Church that received, read, and transmitted them is what makes the answer knowable, and a decree like Trent's, or a council's reception in the East, is that authority doing in public what it had been doing all along.

The sharp form of that is a question put to Protestants: if the canon is a list, and no inspired list appears anywhere in Scripture, then on your own principles you have a fallible collection of infallible books, and you got it from the very Church whose judgement you otherwise decline. Kruger's model is in large part an answer to that question, and you should meet the question before you're handed the answer. His reply is that God's providence and the Spirit's witness do the work an institution would otherwise have to do. Whether that reply succeeds is exactly where a thoughtful Catholic and a thoughtful Protestant stop agreeing.

All three models read the same evidence. All three reject the Nicaea story. All three see the core-and-fringe pattern. The disagreement is over whether the process discovered authority, conferred it, or exercised it, and that isn't a question a new manuscript could settle, because it's a question about what kind of thing a canon is.

This course holds the recognition model, and I'll say it where you can see it: we believe the church received these books because they were God's word, not that they became God's word by being received. That's a faith commitment stated as one. A critical scholar reading McDonald can look at almost every date on the timeline above and agree, the Muratorian Fragment being the one they'll want to move.

Kruger puts the recognition model to a church audience in the second video, so between the two you'll have heard both positions from their own advocates.

Why Catholic and Orthodox Bibles have more books

One thing before we start, about the question's own shape. "Why do they have more books" is how a Protestant asks it. A Catholic asks why the Reformers took seven out. An Orthodox reader would say the Church has read what she has always read, and that nobody in the East ever framed it as a count. I'll use the Protestant phrasing because it's how most readers of this course arrive at the question, and I want you to notice that the phrasing is already half an argument.

By the first century most Jews outside Palestine read their Scriptures in Greek, in the translation called the Septuagint. The Greek manuscripts that survive from the Christian centuries carry, alongside the books of the Hebrew Bible, a set of further books written mostly in the last centuries before Christ: Tobit, Judith, the Wisdom of Solomon, Sirach (also called Ecclesiasticus), Baruch with the Letter of Jeremiah, 1 and 2 Maccabees, and additions to Esther and Daniel. The early church read its Old Testament mostly in Greek, so these books came with it.⁵

Catholics call these books deuterocanonical, "second canon", meaning a second tier of recognition rather than a lesser authority. Protestants call them the Apocrypha, "hidden things", a word Jerome used. Orthodox Christians call them anagignoskomena, "things that are read". Each word is that tradition's own and this lesson uses each for the tradition that uses it.

Around 391, Jerome, translating the Old Testament afresh from Hebrew, wrote a preface that came to be called the Prologus Galeatus, the "helmeted prologue". In it he counts the twenty-two Hebrew books and says that Wisdom, Sirach, Judith, Tobit, and the Shepherd "are not in the canon."¹⁵ In his preface to the books of Solomon he adds that as the church reads Judith, Tobit, and Maccabees "but does not admit them among the canonical Scriptures", so it should read Wisdom and Sirach "for the edification of the people, not to give authority to doctrines of the Church."¹⁵

The Vulgate as it circulated kept the books anyway. Jerome translated Tobit and Judith quickly and didn't translate Wisdom, Sirach, Baruch, or Maccabees at all; those travelled with his text in their older Latin form. So the medieval West inherited both: a Bible with the books in it, and its greatest translator's note saying they were a different kind of book.

"Still open" needs a qualification a Catholic reader would insist on, because the Western church answered the question more than once before Trent. Pope Innocent I sent Exsuperius of Toulouse a canon list including Tobit, Maccabees, and Wisdom in 405. And the Council of Florence, in the bull Cantate Domino of 4 February 1442, listed the whole canon with the deuterocanonical books in it.²⁹ That one matters, because Florence was a general council and it sat seventy-five years before Luther had anything to react to. Trent's list in 1546 repeats Florence's. When a Catholic says Trent defined rather than added, this is the evidence they mean, and it's good evidence.

Martin Luther's 1534 German Bible printed the Apocrypha between the testaments under a heading, in the Gruber Collection's rendering, that they are books "not held equal to the Holy Scriptures, but are useful and good to read".¹⁶ He kept them and moved them.

He did the same thing to four New Testament books, and the point cuts both ways. In his September Testament of 1522 he moved Hebrews, James, Jude, and Revelation to the end and left them out of the numbered contents while the other twenty-three were numbered, and his general preface said those four "have from ancient times had a different reputation". The "epistle of straw" phrase people quote was in that general preface rather than in the preface to James, and he dropped it from every edition after 1537.³⁰ He didn't remove them, no Lutheran Bible since has, and he was applying Eusebius's own category of disputed books. But a Catholic reader will point out, fairly, that a man who reorders the New Testament on his own judgement has made himself the kind of authority he says the church shouldn't be.

The King James Bible of 1611 also placed the Apocrypha between the testaments, as English Bibles before it had.¹⁷ The British and Foreign Bible Society stopped funding Bibles containing the Apocrypha in 1826, and whether that decision was theological, financial, or both is itself debated; don't let anyone tell you one motive as fact.¹⁷ That, not Luther, is why the Protestant Bible on your shelf has no Apocrypha at all.

At the Council of Trent the Catholic Church restated the list, this time with a definition attached. On 8 April 1546 the fourth session issued a decree listing the Old Testament books, "Tobias, Judith, Esther, Job", "Wisdom, Ecclesiasticus", "Jeremias, with Baruch", and "two books of the Machabees, the first and the second" among them, and declared that whoever does not receive "the said books entire with all their parts, as they have been used to be read in the Catholic Church, and as they are contained in the old Latin vulgate edition", "let him be anathema".¹⁸ That gives the Catholic Bible its forty-six Old Testament books and seventy-three in total.

The Orthodox churches aren't a bigger version of Catholicism, and an Orthodox reader would want you to know it. They received the Greek Old Testament directly, and their Bibles carry books the Catholic canon doesn't: 1 Esdras, 3 Maccabees, the Prayer of Manasseh, and Psalm 151, with 4 Maccabees often in an appendix.¹⁹

An Orthodox reader makes the case in their own terms like this. The Church has read the Greek Old Testament since the apostles, and the anagignoskomena have been read in her services without interruption, so the question of a list is downstream of the question of what the Church reads. Wisdom is appointed at Vespers on feasts of hierarchs; the Song of the Three Holy Children is chanted at Holy Saturday. On this account the canon isn't a boundary a council draws but a description of the Church's life, which is why Orthodoxy has never issued anything like Trent's decree, and why Orthodox lists still differ a little between local churches without anyone treating that as a crisis. What binds is reception, not a defining act: the Quinisext Council of 692 ratified the earlier canons, and the Synod of Jerusalem of 1672 answered a Calvinist-leaning confession by calling Wisdom, Judith, Tobit, the additions to Daniel, Maccabees, and Sirach "genuine parts of Scripture" without claiming to settle anything the Church hadn't already been doing.²⁰ And an Orthodox reader would add, gently, that both Western sides are arguing inside a frame Orthodoxy doesn't use: "which books are in the canon" assumes the canon is the kind of thing with a legally defined edge.

Orthodox writers differ among themselves on the standing of these books. Timothy Ware's standard introduction says most Orthodox scholars today, following Athanasius and Jerome, hold them to be part of the Bible while standing on a lower footing than the rest of the Old Testament.²¹ The Ethiopian Orthodox Tewahedo Church has the largest canon of all, eighty-one books, including Enoch and Jubilees.¹⁹

Hebrews 11, 2 Maccabees 7, and what a quotation proves

The hard case is a quotation, and I want you to do part of it yourself.

Hebrews 11 is the roll-call of faith. Verse 35 says: "and women received back their dead raised to life. But others were tortured, not accepting release, to obtain resurrection to a better life" (Hebrews 11:35, NET).²² Now open 2 Maccabees 7, the story of seven brothers tortured to death under Antiochus for refusing to eat pork. The second brother tells the king that "the King of the world shall raise us up, who have died for his laws, unto everlasting life" (Brenton).²³ Nothing in the Hebrew Bible matches Hebrews 11:35 as closely. The writer of Hebrews, listing heroes of faith, seems to be reaching for a story from a book Protestant Bibles don't contain.

A Catholic reader makes this case, in their own terms: the apostolic church's Bible was the Greek Septuagint, with these books in it; the New Testament alludes to them, as here; Hippo and Carthage listed them; Florence listed them in 1442; the church read them as Scripture for well over a thousand years before Trent; Trent didn't add anything, it defined what was already practised; and it's the Protestant list, drawn up in the sixteenth century on Jerome's minority view, that is the innovation.

A Protestant reader makes this case, in their own terms: the canon Jesus and the apostles used was the Hebrew one, and Luke 24:44 describes it in its three parts; Josephus counted twenty-two; the New Testament never cites a deuterocanonical book as Scripture with a formula like "it is written", though it alludes to many books it doesn't treat as Scripture; Jerome, who knew the Hebrew, distinguished them; and Athanasius's list has them among books "to be read" but outside the canon.

Then there's Athanasius, whom both sides reach for. His Old Testament is twenty-two books; Wisdom, Sirach, Judith, and Tobit are outside the canon, "to be read" by new converts. That helps the Protestant. But his twenty-two leave out Esther, which every Protestant Bible has, and count Baruch and the Letter of Jeremiah inside Jeremiah, which no Protestant Bible does.¹⁰ Neither modern list matches his. He's a witness that the question was still being settled, not a referee who settled it.

So sort it. History establishes what Josephus, Athanasius, Jerome, Florence, Luther, and Trent said; both sides agree on every quotation above. Scholars argue about how much weight each witness should carry, and about whether the Greek-speaking Jews of Alexandria ever had a wider canon of their own. Most scholars now doubt that an "Alexandrian canon" existed as a defined thing, which takes one old argument away from the Catholic side. Be careful how you use that, though, because the argument that dissolved it is Albert Sundberg's, and Sundberg's conclusion was that in the first century neither community had a closed list: no Alexandrian canon, and no closed Palestinian one either.⁸ Taken whole it costs the Catholic side an argument and the Protestant side its premise, and what survives on the Catholic side is the argument from the church's continuous use, which was always the stronger one. What is held by faith, and by belonging to a church, is which list is Scripture.

This course uses the sixty-six books, for the Protestant reasons above, and we say so as our position. Two things belong alongside that. First, few Christian doctrines turn on the difference, and the clearest one that does is prayer for the dead: 2 Maccabees 12, where Judas Maccabeus collects money for a sin offering on behalf of fallen soldiers, and the narrator comments that "if he had not hoped that they that were slain should have risen again, it had been superfluous and vain to pray for the dead" (Brenton), which Catholic teaching cites for purgatory.²³ A Catholic would add that Wisdom, Sirach, and Tobit shape the liturgy and the tradition's moral vocabulary whether or not a doctrine hangs on a proof text, and that's fair. Second, the deuterocanonical books are worth reading whatever you hold. Wisdom and Sirach were read by the apostles' generation; 1 Maccabees is the history behind Hanukkah and the world Jesus was born into; and you can't read Hebrews 11 as its first readers did without 2 Maccabees 7 in your head.

What people get wrong

"Nicaea, or Constantine, chose the books." Walked above. Irenaeus 180, Eusebius listing the disputes in a history begun before the council, Athanasius 367, Hippo and Carthage in the 390s, and no canon business in Nicaea's records.^5691011

"The lost gospels were suppressed." Lesson 6 introduced Thomas, Peter, and Judas.

"Rome added the Apocrypha at Trent." They were in the Septuagint, the Vulgate, and every Western Bible for a millennium, and Florence listed them in 1442, before Luther was born. Trent defined; Luther set apart.^15161829

"Luther cut seven books out of the Bible." He moved them and headed them, and he did the same to four New Testament books. He removed nothing, and the empty shelf in your Bible is the Bible Society's decision of 1826.¹⁶¹⁷³⁰

"Protestants never had the Apocrypha." The King James Bible of 1611 printed it, and so did every major English Bible before it. If you own a Bible without it, that's a nineteenth-century publishing decision rather than a Reformation one.¹⁷

"Athanasius invented the list." He reports it and fences it. Read his sentence again: "Let no man add to these, neither let him take ought from these."¹⁰ That's the voice of someone guarding something, not inventing it.

"The Protestant canon is the original one." Say that to a Catholic and you'll hear, with reason, that the Protestant Old Testament is the sixteenth-century list and theirs is the one the church read for a thousand years. Say the reverse to a Protestant and you'll hear that the apostles' Bible was the Hebrew one. Both are stating their case; neither is stating a fact the other has missed.

"The criteria were arbitrary." They weren't a checklist and they weren't a whim. The Muratorian Fragment gives its reason for the Shepherd in the open, and Hebrews shows the tests being weighed against each other over two centuries rather than applied mechanically. A rule that can be argued about in public for two hundred years is not arbitrary; it's also not a formula.

"A New Testament allusion proves a book is Scripture." Jude quotes 1 Enoch and nobody outside Ethiopia concluded 1 Enoch was Scripture. Allusion proves familiarity. Both sides know this, which is why the Catholic case rests on church use rather than on Hebrews 11:35.

"The Muratorian Fragment proves the canon was closed by 200." It shows a core that was settled and a fringe that wasn't, and its own date is argued. Don't overstate your best evidence; the other side will notice.

"The four schools of thought about canon are levels of seriousness." Almost every reader arrives holding one account, usually inherited, and assumes it's the plain reading and the others are interpretation. All three are interpretation, each was worked out by people who took the evidence seriously, and the disagreement between them is about what kind of thing a canon is rather than about who is faithful.

Now do it yourself, on Athanasius and Eusebius, on Hebrews 11, and on six claims to sort

Connections

Lesson 1 gave you the counts and promised the reasons; you have them now. Lesson 2 used Luke 24:44 for the story's shape; here the same verse turned out to be the earliest Christian witness to the Jewish canon. Lesson 6 introduced the other gospels; now you have the tests that sorted them.

Lesson 9 asks the next question. Knowing which books tells you nothing about whether the words in them are the words their authors wrote. That's the question of manuscripts, copying, and the three famous passages you can find in your own Bible. Lesson 10 then asks what a Christian makes of all this: what "inspired" means, and how the three-way sort you used three times in this lesson becomes a tool for every argument about the Bible you'll ever have.

Go deeper

• F. F. Bruce, The Canon of Scripture (IVP, 1988), chapters 12 and 17. The standard one-volume history, with the Muratorian Fragment and Athanasius each given a chapter. Bruce is an evangelical Protestant and the recognition thesis is his, which is worth knowing before you use his tests.
• Michael Kruger, Canon Revisited (Crossway, 2012), part 1. The self-authenticating model argued at full length; his free lecture course is at BiblicalTraining.org.
• Lee Martin McDonald, The Biblical Canon, 3rd ed. (Hendrickson, 2007). The community model by its leading advocate; read the chapters on the fourth century alongside Kruger and decide for yourself.
• Dale Martin, RLST 152, lecture 2, "From Stories to Canon", Open Yale Courses. Free video and transcript; the critical position from its own teacher.
• Eckhard Schnabel, "The Muratorian Fragment: The State of Research", Journal of the Evangelical Theological Society 57 (2014). A short survey of the dating dispute, and a good look at how scholars actually argue about a document.

Sources

1. Prologue to Sirach (Ecclesiasticus), Brenton's Septuagint translation (1851), public domain, fetched at Stage 4 from the codexcommunion/bible-toolkit mirror. The three tripartite references quoted in the body; and "in the eight and thirtieth year coming into Egypt, when Euergetes was king", which is the basis for the conventional date of 132 BC and which dates his arrival, not the preface. The draft quoted the NRSVue here; that translation is copyrighted and is not among the four Editorial Standards 4.7 permits, so it was replaced.
2. Josephus, Against Apion 1.8 (1.37 to 43), c. AD 95, Whiston translation at penelope.uchicago.edu. "Only twenty two books ... justly believed to be divine"; "no one has been so bold, as either to add any thing to them; to take any thing from them; or to make any change in them."
3. Luke 24:44, NET. The law of Moses, the prophets, and the psalms.
4. Babylonian Talmud, Bava Batra 14b to 15a. The baraita giving the order of the Prophets and the Writings. As summarised in SOURCES.md from the standard editions; the manuscript was not consulted.
5. F. F. Bruce, The Canon of Scripture (IVP, 1988). Chapter 12 on the Muratorian Fragment, chapter 17 on Athanasius and after. SOURCES.md records Bruce's thesis, those two chapter titles, and the three working tests, and nothing else; the book was not opened for this lesson. So: the three tests and the recognition thesis are his, as recorded. Marcion's edited Luke and ten Pauline letters c. 144, the absence of canon business from Nicaea's surviving records, the early church's Greek Old Testament, and the doubt that an "Alexandrian canon" existed are given here as standardly reported, not on Bruce's pages, and the Alexandrian point is Sundberg's argument (note 8).
6. Irenaeus, Against Heresies 3.1.1 and 3.11.8, c. 180, ANF translation at New Advent. The four Gospels with their authors; "it is not possible that the Gospels can be either more or fewer in number than they are", argued from the "four zones of the world" and "four principal winds". That Marcion, Tatian's Diatessaron, and other groups were pressing the number in Irenaeus's own environment is from his own pages and the standard accounts.
7. The Muratorian Fragment, Metzger's translation at bible-researcher.com. Discovered by Muratori in the Ambrosian Library and published in 1740; the manuscript seventh or eighth century; "the third book of the Gospel is that according to Luke"; the Apocalypse of Peter, "some of us are not willing that the latter be read in church"; the Shepherd "ought indeed to be read" but not publicly, written "very recently, in our times", while Pius was bishop. Books not mentioned: Hebrews, James, 1 and 2 Peter, a clear 3 John. Metzger's headnote dates the list to about 170, which is why the body gives the traditional range as 170 to 200 rather than 180 to 200.
8. Dating of the Muratorian Fragment: A. C. Sundberg, "Canon Muratori: A Fourth-Century List", Harvard Theological Review 66 (1973); G. M. Hahneman, The Muratorian Fragment and the Development of the Canon (Oxford, 1992); J. Verheyden, "The Canon Muratori: A Matter of Dispute", in The Biblical Canons (Leuven, 2003); E. J. Schnabel, "The Muratorian Fragment: The State of Research", JETS 57 (2014); C. K. Rothschild, The Muratorian Fragment (Mohr Siebeck, 2022). Also Sundberg, The Old Testament of the Early Church (Harvard, 1964), for the argument against a distinct "Alexandrian canon" and for his own conclusion that neither community had a closed first-century list. Positions as summarised in SOURCES.md and from published summaries; none of these was read in full here.
9. Eusebius, Ecclesiastical History 3.25, NPNF translation at New Advent. All four categories: recognised, disputed "which are nevertheless recognized by many", spurious, and the forgeries circulated by heretics, where the Gospels of Peter, Thomas and Matthias sit. Revelation named in both the recognised list ("if it really seem proper") and the rejected. On the date: the manuscripts show editions of about 313, 316 and 324 to 325, with Book 3 most likely in the earliest; from Burgess's analysis as standardly reported, not read here.
10. Athanasius, Festal Letter 39, Easter 367, NPNF translation at New Advent. "Twenty-two books in number" for the Old Testament, with "Jeremiah with Baruch, Lamentations, and the epistle, one book" and Esther absent; the New Testament list; "These are fountains of salvation"; "Let no man add to these, neither let him take ought from these"; the books "not indeed included in the Canon, but appointed by the Fathers to be read by those who newly join us".
11. Councils of Hippo (393) and Carthage (397; canons compiled 419), text and discussion at bible-researcher.com. The Old Testament list including Tobit, Judith, Esther, two books of Esdras, two books of Maccabees, and five books of Solomon. The clause "For the confirmation of this canon the church across the sea shall be consulted" (transmarina ecclesia consulatur) is preserved in one manuscript strand rather than in every copy, which the body now says. That Augustine was present at Hippo is standardly reported.
12. Dan Brown, The Da Vinci Code (Doubleday, 2003), chapter 55. The character Teabing's claim that the Bible "was collated by the pagan Roman emperor Constantine the Great". Quoted from secondary summaries of the chapter; the novel was not opened. Named as the source of the popular story, not as a historical source.
13. Michael J. Kruger, Canon Revisited (Crossway, 2012). The self-authenticating model. Described from published summaries of the argument; not opened.
14. Lee Martin McDonald, The Biblical Canon: Its Origin, Transmission, and Authority, 3rd ed. (Hendrickson, 2007); The Formation of the Biblical Canon, 2 vols. (T&T Clark, 2017). The community-determined model. Described from published summaries; not opened. The earlier draft gave the publisher as Baker, which was wrong.
15. Jerome, Prologus Galeatus (preface to Samuel and Kings, c. 391) and preface to the books of Solomon, NPNF translations. "Wisdom ... and the book of Jesus the Son of Sirach, and Judith, and Tobias, and the Shepherd are not in the canon"; "As, then, the Church reads Judith, Tobit, and the books of Maccabees, but does not admit them among the canonical Scriptures, so let it read these two volumes for the edification of the people, not to give authority to doctrines of the Church." That Jerome translated Tobit and Judith hastily and did not translate Wisdom, Sirach, Baruch or Maccabees at all is standardly reported.
16. Luther's 1534 German Bible, contents page and Apocrypha heading, per the LSTC Gruber Collection description as recorded in SOURCES.md: books "not held equal to the Holy Scriptures, but are useful and good to read". Not verified against a facsimile in this environment.
17. King James Version (1611), with the Apocrypha between the testaments; the British and Foreign Bible Society's 1826 decision to stop funding Bibles containing the Apocrypha, with the motive (theological, financial, or both) debated. As recorded in SOURCES.md.
18. Council of Trent, Session IV, 8 April 1546, Decree Concerning the Canonical Scriptures, English text at papalencyclicals.net. The book list and the anathema as quoted. The vote tally sometimes reported for this session is via secondary sources and disputed, and is omitted here.
19. Orthodox and Ethiopian canons: 1 Esdras, 3 Maccabees, the Prayer of Manasseh, and Psalm 151 in Orthodox Bibles, with 4 Maccabees in an appendix; the Ethiopian Orthodox Tewahedo canon of eighty-one books. As recorded in SOURCES.md. The liturgical examples (Wisdom at Vespers on feasts of hierarchs, the Song of the Three Holy Children at Holy Saturday) and the Quinisext Council's ratification of the earlier canons are standardly reported and were not verified from a service book here.
20. Synod of Jerusalem (1672), Confession of Dositheus, Question 3, English text at ccel.org (Schaff, Creeds of Christendom). Wisdom, Judith, Tobit, Bel and the Dragon, Susanna, Maccabees, and Sirach as "genuine parts of Scripture"; a regional synod, binding where received.
21. Timothy Ware (Kallistos), The Orthodox Church (Penguin), chapter on Holy Tradition: "most Orthodox scholars at the present day, following the opinion of Athanasius and Jerome, consider that the Deutero-Canonical Books, although part of the Bible, stand on a lower footing than the rest of the Old Testament." Wording confirmed from the widely reproduced excerpt of that chapter; the book was not opened and the page in the current edition is not confirmed. The draft rendered this as "many ... though not all agree", which weakened a source it was citing.
22. Hebrews 11:35, NET.
23. 2 Maccabees 7:9 and 12:44, Brenton (1851), public domain, fetched at Stage 4. "The King of the world shall raise us up, who have died for his laws, unto everlasting life"; "for if he had not hoped that they that were slain should have risen again, it had been superfluous and vain to pray for the dead." Note the wording: Brenton reads "superfluous and vain", where the draft's NRSVue text read "superfluous and foolish".
24. Jude 14 to 15, NET. The quotation from 1 Enoch.
25. NET Bible, 2nd edition (2019), netbible.org. Text verified verbatim at this lesson's Stage 4 against the two mirrors recorded in SOURCES.md, edition confirmed NET2; bible.org and netbible.org are unreachable from the drafting environment, and the earlier note claiming an API fetch was wrong. Passages: 2 Peter 3:15 to 16, Luke 24:44, Hebrews 11:35, Jude 14 to 15. Scripture quoted by permission. Quotations designated (NET) are from the NET Bible® copyright ©1996, 2019 by Biblical Studies Press, L.L.C. All rights reserved. See Editorial Standards 4.7 for why this course uses the NET.
26. On whether the Jewish canon's edges were closed in the first century: Mishnah Yadayim 3:5 on whether Ecclesiastes and the Song of Songs "defile the hands"; John Barton, Oracles of God; Timothy Lim, The Formation of the Jewish Canon; Lee Martin McDonald, as above. SOURCES.md records this as a standing instruction for this course: a list that gives an order presupposes a working list and does not show the list was uncontested. None of these was opened here.
27. On what Marcion's edit proves: Adolf von Harnack, Marcion: The Gospel of the Alien God, for the case that the edit presupposes an existing collection, together with his own qualification that the church did not create a canon in response; Hans von Campenhausen, The Formation of the Christian Bible, for the case that no normative collection of new writings is attested before Marcion. A minority (BeDuhn, Klinghardt, Vinzent) argues Marcion's gospel is prior to canonical Luke. From published summaries; none opened.
28. The Peshitta's twenty-two-book New Testament, omitting 2 Peter, 2 and 3 John, Jude and Revelation; the Philoxenian (508) and Harklean (616) versions supplying the five; the Church of the East not receiving Revelation. Standardly reported in reference works; no manuscript or critical edition consulted here.
29. Council of Florence, bull Cantate Domino, 4 February 1442, listing the full canon with the deuterocanonical books; and Innocent I's letter to Exsuperius of Toulouse, 405, with a similar list. English text of Florence at papalencyclicals.net. Not fetched in this environment; standardly reported, and the point matters enough to the Catholic case that the fact-check should confirm the decree's text when egress allows.
30. Luther's treatment of Hebrews, James, Jude and Revelation in the September Testament of 1522: moved to the end, left out of the numbered contents, and described in the general preface as having "from ancient times had a different reputation"; the "epistle of straw" phrase in that general preface rather than in the preface to James, and dropped from editions after 1537. Standardly reported; the 1522 and 1534 volumes were not consulted.