Open any Bible and the first page you meet is a table of contents. It is also the one page no Christian believes is inspired. No verse in Scripture lists the books of Scripture; the New Testament nowhere names its own twenty-seven. So before a single doctrine can be argued from the Bible, a prior question has to have been answered — which writings are the Bible? — and it has to have been answered from outside the text. That is not a Catholic debating trick. It is a plain fact about the shape of the book, and every tradition that prints a table of contents has already relied on somebody’s answer.

This article tells the story of how that question was answered for the New Testament: the criteria, the contenders, the crisis that forced the issue, and the fourth-century moment when the twenty-seven became fixed. Two neighboring questions have their own articles and are only touched here: why the Catholic Old Testament has the seven deuterocanonical books Protestants removed, and what the canon’s history implies about the authority of the Church that discerned it. Here the subject is the history itself — because the best Protestant reply to that authority argument is a historical claim, and it deserves to be tested against the record.

Scripture itself shows canon-consciousness beginning — and shows why it could not finish the job alone. The most striking text is 2 Peter: “as also our most dear brother Paul, according to the wisdom given him, hath written to you: As also in all his epistles, speaking in them of these things; in which are certain things hard to be understood, which the unlearned and unstable wrest, as they do also the other scriptures, to their own destruction” (2 Peter 3:15–16). Inside the New Testament, Paul’s letters are already classed with “the other scriptures.” Remarkable — and double-edged, for 2 Peter is itself the most weakly attested and latest-accepted of the eventual twenty-seven. The very verse that shows Scripture calling Paul Scripture comes from a book whose own place had to be settled by somebody.

The second text must be handled honestly, because apologists routinely overclaim it. Paul writes: “For the scripture saith: Thou shalt not muzzle the ox that treadeth out the corn: and, The labourer is worthy of his reward” (1 Timothy 5:18). The first saying is Deuteronomy 25:4. The second matches, word for word in the Greek, a sentence found in written form only in Luke’s Gospel: “for the labourer is worthy of his hire” (Luke 10:7) — the Douay renders the same Greek phrase “reward” in one place and “hire” in the other. If Paul is calling both sayings “scripture,” this is the earliest attestation of a Gospel ranked with the Law. But “the scripture saith” may govern only the Deuteronomy clause, or Paul may cite the Lord’s saying from oral tradition rather than Luke’s finished text. The honest conclusion is modest: the New Testament shows the beginning of a canon — apostolic writings already received as Scripture — but nowhere a list, and nowhere a method for producing one.

Luke himself tells us why a list would be needed: “Forasmuch as many have taken in hand to set forth in order a narration of the things that have been accomplished among us” (Luke 1:1). Gospel-writing was a crowded field from the first generation; where many take pen in hand, someone must eventually judge which narrations the Church will read at the altar.

Here is the fact the “self-authenticating” theory must survive: for roughly three hundred years, devout, orthodox Christians did not agree on the contents of the New Testament. The core was never seriously in doubt — the four Gospels, Acts, and the letters of Paul were received early and nearly everywhere. But around that core stood a disputed fringe, and the dispute ran in both directions: books now in every Bible were doubted, and books now in no Bible were read in church.

We do not have to reconstruct this from scraps. Eusebius of Caesarea, in his Church History (c. A.D. 325), sorted the books and reported the state of the question. First the accepted: “First then must be put the holy quaternion of the Gospels; following them the Acts of the Apostles… the epistles of Paul… the extant former epistle of John, and likewise the epistle of Peter… After them is to be placed, if it really seem proper, the Apocalypse of John, concerning which we shall give the different opinions at the proper time.” Then the disputed — the antilegomena, the “spoken-against” books: “Among the disputed writings, which are nevertheless recognized by many, are extant the so-called epistle of James and that of Jude, also the second epistle of Peter, and those that are called the second and third of John, whether they belong to the evangelist or to another person of the same name.” And then the rejected: “Among the rejected writings must be reckoned also the Acts of Paul, and the so-called Shepherd, and the Apocalypse of Peter, and in addition to these the extant epistle of Barnabas, and the so-called Teachings of the Apostles; and besides, as I said, the Apocalypse of John, if it seem proper, which some, as I said, reject, but which others class with the accepted books” (Church History 3.25).

Read that catalogue slowly. In 325 — on the eve of Nicaea — James, Jude, 2 Peter, 2 and 3 John are still “spoken against,” and the Apocalypse appears in two of Eusebius’s columns at once, because the Church was that divided over it. Meanwhile the books that lost their bid were not obscure forgeries: the great fourth-century manuscript Codex Sinaiticus binds the Shepherd of Hermas and the Epistle of Barnabas after the Apocalypse, and these books were read in Christian assemblies for generations. Whatever discerned Scripture from near-Scripture, it was not a glow on the page every believer could see — believers demonstrably did not see it alike.

One class of books, however, was never in the running. Of the writings circulated by heretical sects under apostolic names — he names “the Gospels of Peter, of Thomas, of Matthias” among them — Eusebius writes that “they clearly show themselves to be the fictions of heretics. Wherefore they are not to be placed even among the rejected writings, but are all of them to be cast aside as absurd and impious.” The Gospel of Thomas was not narrowly edged out at some council; no church’s canon ever listed it. The real contest was between the apostolic books and edifying near-misses like Hermas — not between the Bible and the material of modern novels.

The man who made a canon list unavoidable was a heretic. Around A.D. 140 the shipowner Marcion came to Rome teaching that the God of the Old Testament was not the Father of Jesus Christ — and backed the doctrine with the first fixed scriptural list in Christian history: one Gospel (an edited Luke) and ten letters of Paul. Nothing else. Rome expelled him in 144, but once a heretic had published a closed list, the Church could no longer leave her own boundaries informal. Marcion did not create the Church’s Scriptures; he forced the Church to say out loud what they were.

Her answer took the form of criteria, applied by bishops across generations. Apostolicity: was the book written by an apostle or an apostle’s close companion — Mark preserving Peter’s preaching, Luke Paul’s? Orthodoxy: did it conform to the rule of faith handed down in the churches the apostles founded — the test on which Serapion of Antioch, around 200, withdrew the so-called Gospel of Peter once he read it against the apostolic doctrine (Eusebius, Church History 6.12)? And catholicity: was it received and read in the liturgy across the whole Church, not prized in one province only? Notice what all three criteria are: judgments the Church makes about a book — by her memory, her rule of faith, her worship. None is a property a reader extracts from the text alone in his study.

The earliest surviving canon list from the Church’s side is the Muratorian fragment, a damaged Latin catalogue naming the four Gospels, Acts, thirteen Pauline letters, Jude, two epistles of John, and the Apocalypse. Its reason for excluding the beloved Shepherd is the criteria in action: “The Pastor, moreover, did Hermas write very recently in our times in the city of Rome, while his brother bishop Pius sat in the chair of the Church of Rome. And therefore it also ought to be read; but it cannot be made public in the Church to the people, nor placed among the prophets, as their number is complete, nor among the apostles to the end of time.” Edifying, yes — read it. Apostolic, no — too recent. The fragment’s date must be stated honestly: the traditional c. 170 rests on that “very recently in our times” (Pius died around 155), but some scholars — Sundberg and Hahneman most prominently — argue for a fourth-century Eastern origin, and the question remains genuinely disputed. Nothing here hangs on the early date; if the fragment is late, the earliest orthodox list simply moves closer to Eusebius and the process looks longer still — which helps the Catholic argument, not the Protestant one.

In the year 367, in the routine Easter letter by which the bishop of Alexandria announced the date of the feast, Athanasius — the man who had all but carried the Nicene faith on his back — set down the books of Scripture: “Again it is not tedious to speak of the [books] of the New Testament,” he writes, and names four Gospels, Acts, seven Catholic epistles, fourteen epistles of Paul, and the Revelation of John (Festal Letter 39). It is the first surviving document in Christian history to list exactly the twenty-seven books — no more, no fewer — three hundred and thirty-odd years after the Resurrection.

Two honest footnotes belong to that milestone. Athanasius’s letter did not end the discussion by itself — it was one bishop’s norm for his flock, and the East went on hesitating over the Apocalypse for generations. And his Old Testament list has wrinkles of its own (it omits Esther from the canon proper and joins Baruch to Jeremiah) — a reminder that no single Father was the Church’s final organ of discernment. The settlement came the way the Church settles things: in council.

At Hippo Regius in 393, and at Carthage in 397 and again in 419 — councils in which Augustine took part — the bishops of North Africa enacted the canon in law: the twenty-seven books of the New Testament, and an Old Testament including the deuterocanonical books, with the provision that nothing else be read in church as divine Scripture and that “the church across the sea” — Rome — be consulted for the confirmation of the canon; the 419 text directs that it be sent to Boniface, bishop of Rome, and the other bishops, “that they may confirm this canon.” These were regional councils; their authority lay precisely in their reception by the wider Church and by Rome, which is how the identical list passed into the usage of the whole West and, eventually, into the solemn definitions of Florence and Trent. (Trent defined the same 73-book canon Carthage listed — a response to sixteenth-century subtraction, not a medieval addition; that story belongs to the deuterocanon article.) From 397 onward the contents of the New Testament were not an open question again in the Church — until the objector’s own tradition reopened four books, briefly, in 1522.

Now to the steelman — and the first thing to say is that half of it is simply Catholic doctrine. The Church has never taught that she conferred inspiration or made books canonical by fiat. The First Vatican Council teaches that the books are sacred and canonical not because the Church’s authority approved merely human compositions, but because, written under the inspiration of the Holy Ghost, they have God for their author — and were as such entrusted to the Church (Dei Filius, ch. 2). Hippo and Carthage made nothing inspired. “Recognized, not conferred” is not the Protestant escape hatch; it is the Catholic position. The real question was never whether the canon was recognized rather than created — both sides say so — but whether the recognition was trustworthy: whether the judgment that bound James in and shut Hermas out carries divine protection, or is a fallible opinion that happens, one hopes, to be right.

Put that question to the self-authentication model and it comes apart on its own evidence. If the divine qualities of the books, witnessed by the Spirit in believers’ hearts, suffice to identify the canon, the history of Sections III through V could not have happened: Spirit-filled, orthodox, martyr-producing churches doubted James, 2 Peter, and the Apocalypse for three centuries while binding Hermas and Barnabas into their Bible manuscripts. Either the divine qualities were not perceptible enough to settle the question — and cannot bear the weight the theory puts on them — or the Spirit’s witness was heard rightly by some churches and wrongly by others, in which case one still needs an organ competent to say which. Kruger’s model, to its credit, sees this and adds “corporate reception” as a confirming attribute. But the corporate reception that actually occurred was the process this article has narrated — bishops applying criteria, councils enacting lists, Rome confirming them. Call the discerning body “the covenant community” and the theory has not eliminated the Catholic Church from the story; it has renamed her.

The circularity, meanwhile, sits in plain view. To know that exactly twenty-seven books carry the divine qualities, one must already know which books to test — and in practice every Protestant receives the list from the community’s prior judgment. The Westminster Confession demonstrates the point: its first chapter argues that Scripture evidences itself, while its second section simply prints the sixty-six-book list — a list no verse of Scripture supplies. The Confession’s canon catalogue is, on its own terms, an unwritten tradition. And the honest wing of the Reformed world has conceded the consequence: R. C. Sproul famously summarized the Protestant position as holding a fallible collection of infallible books. Candid — and devastating. If the collection is fallible, the objector cannot be certain his infallible rule of faith includes any given disputed book; the doubt leaks upward from the table of contents into everything argued from it.

Last, the sheep and the voice. John 10 is true, and the Catholic gladly argues from it — but it describes a flock following its Shepherd, not a solitary reader adjudicating manuscripts. Ask how, concretely, the sheep corporately heard the Shepherd’s voice in the matter of the canon, and history returns one answer: through the discernment of the Church — her memory of the apostles, her rule of faith, her liturgy, her councils. The verse the objection quotes, followed to the end, walks the objector into the arms of the Church that did the hearing.