Canon formation

The word "canon" derives from the Greek kanon, meaning a reed or measuring stick, and by extension a rule or standard.¹ When applied to the Bible, "canon" refers to the collection of books that a religious community recognizes as authoritative scripture—the texts that serve as the rule and standard for faith and practice.^{1, 2} The process by which this collection was determined is called canon formation, and contrary to popular imagination, it was neither instantaneous nor uncontested. The biblical canon emerged over centuries through complex negotiations among diverse communities, shaped by theological controversy, liturgical practice, and the contingencies of history.^{2, 3}

Understanding canon formation is essential for evaluating claims about biblical authority. If the canon was self-evident or divinely revealed in a single moment, the Bible's authority might rest on a straightforward divine guarantee. But if the canon was shaped by human decisions over many centuries—with different communities reaching different conclusions—then the question of what counts as scripture becomes more complex. The historical evidence points decisively toward the latter scenario.^{2, 4}

The Council of Nicaea myth

A widespread popular belief holds that the Emperor Constantine and the bishops at the Council of Nicaea in 325 CE voted on which books would be included in the Bible. This claim gained currency through Dan Brown's 2003 novel The Da Vinci Code, which sold over 80 million copies, and in which a character declares that "more than eighty gospels were considered for the New Testament" and that Constantine "commissioned and financed a new Bible" that omitted certain gospels.^{48, 49} The claim is entirely without historical foundation. The Council of Nicaea addressed the Arian controversy about Christ's divine nature, producing the Nicene Creed with its affirmation that Christ was "of one substance (homoousios) with the Father." The council also established a uniform date for Easter, issued twenty canons governing church discipline, and addressed the Meletian schism in Egypt—but it made no decisions about the biblical canon.^{50, 51} Eusebius of Caesarea, who was present at the council and later wrote a biography of Constantine, recorded no discussion of Scripture's contents; his silence, combined with his detailed treatment of canonical debates elsewhere in his Ecclesiastical History, confirms the council did not address the question.⁵² Historians across the theological spectrum agree: Bart Ehrman calls the claim "absolutely false," Craig Blomberg notes that "the council of Nicaea in 325 did not discuss the biblical canon," and F. F. Bruce wrote that the canon "was not the result of any pronouncement by a Church Council; it was rather the general consensus of the churches."^{53, 54, 55}

The Hebrew Bible and Old Testament

The canon of the Hebrew Bible (the Jewish scriptures, also known as the Tanakh) developed over approximately a thousand years, from the earliest written traditions around the tenth century BCE to the final recognition of all twenty-four books by Jewish communities in the early centuries of the Common Era.^{5, 6} This process is not fully documented, but scholars have reconstructed its general outlines from internal biblical evidence, ancient Jewish sources, and archaeological discoveries.⁵

The Hebrew Bible is traditionally divided into three sections: Torah (the five books of Moses), Nevi'im (Prophets), and Ketuvim (Writings)—an arrangement reflected in the acronym Tanakh.⁶ These three sections appear to have achieved canonical status at different times. The Torah was likely the first section to be regarded as authoritative scripture, probably by the late fifth or fourth century BCE, following the reforms of Ezra and Nehemiah.^{5, 7} The Prophets seem to have been recognized as a closed collection by the second century BCE, as evidenced by the prologue to Ben Sira (Ecclesiasticus), which refers to "the Law and the Prophets and the other books" around 132 BCE.⁸ The Writings were the last section to be finalized, with some books (such as Ecclesiastes, the Song of Songs, and Esther) remaining disputed into the rabbinic period.⁵

The so-called "Council of Jamnia" (or Yavneh), once thought to have definitively closed the Jewish canon around 90 CE, is now regarded by scholars as largely legendary in this respect.⁹ There was a rabbinic academy at Yavneh after the destruction of the Temple in 70 CE, and the rabbis there discussed which books "defile the hands" (a technical term for canonical status), but there is no evidence of a council that formally voted on the canon.^{9, 10} Jack Lewis's influential 1964 study demonstrated that the idea of a canon-defining council at Jamnia was a scholarly construct without solid historical foundation.¹⁰ The Jewish canon appears to have been effectively closed by consensus rather than by decree, probably by the end of the second century CE.⁵

The Septuagint and the deuterocanonical books

The situation becomes more complex when we consider the Septuagint (LXX), the Greek translation of the Hebrew scriptures produced in Alexandria, Egypt, beginning around 280–270 BCE.¹¹ The Septuagint was the Bible of the early church and is quoted extensively in the New Testament.¹¹ Importantly, the Septuagint manuscripts include books not found in the Hebrew canon: Tobit, Judith, the Wisdom of Solomon, Sirach (Ecclesiasticus), Baruch, 1 and 2 Maccabees, and additions to Esther and Daniel.¹²

These additional books are called "deuterocanonical" by Catholics (meaning "second canon," recognized as canonical though not in the original Hebrew collection) and "Apocrypha" by Protestants (meaning "hidden" or, pejoratively, "of doubtful authorship").¹² The difference between the Protestant Old Testament (39 books, matching the 24 books of the Hebrew Bible when counted differently) and the Catholic Old Testament (46 books) stems from the Protestant Reformers' decision to follow the Hebrew canon rather than the longer Septuagint tradition that the Catholic Church had inherited.^{12, 13}

Martin Luther and the Protestant Reformers relegated the deuterocanonical books to an appendix, judging them "not equal to the Holy Scriptures, yet profitable and good to read."¹⁴ The Catholic Church responded at the Council of Trent (1546) by formally affirming the deuterocanonical books as fully canonical.¹⁵ This was not a new decision but a confirmation of the practice that had prevailed in the Western church since at least the councils of Hippo (393) and Carthage (397).^{15, 16}

Old Testament canons across traditions^{12, 17, 18}

The New Testament canon

The New Testament canon developed through a different but equally gradual process. The earliest Christians had no "New Testament"—their scriptures were the Jewish scriptures (in Hebrew or Greek translation), and their authoritative traditions were oral teachings about Jesus transmitted by apostles and eyewitnesses.^{2, 19} The writings that would become the New Testament were composed between approximately 50 and 120 CE, but their recognition as "scripture" equivalent to the Old Testament emerged only gradually over the following centuries.²

Paul's letters were the earliest Christian writings, composed in the 50s and early 60s CE.²⁰ Evidence suggests that Paul's letters were being collected and circulated as a corpus by the end of the first century; 2 Peter 3:15–16, likely written in the early second century, refers to Paul's letters as a collection and places them alongside "the other scriptures."²¹ The four Gospels (Matthew, Mark, Luke, and John) were composed between approximately 70 and 100 CE and were being widely used by churches throughout the Roman Empire by the mid-second century.²

By the late second century, a core New Testament was taking shape. Irenaeus of Lyon, writing around 180 CE, argued for the authority of precisely four Gospels—no more and no less—comparing them to the four zones of the world and the four principal winds.²² Yet many other Christian writings were also being read as scripture in various communities, including the Shepherd of Hermas, the Didache, the Epistle of Barnabas, the Gospel of Peter, and the Apocalypse of Peter.^{2, 23} The process of determining which books belonged in the canon and which did not would take several more centuries.²

Marcion and the impetus for definition

One significant impetus for defining the canon came from Marcion of Sinope (c. 85–160 CE), a wealthy shipowner who traveled to Rome and developed a theology that sharply distinguished the God of the Old Testament from the God revealed by Jesus.²⁴ Marcion rejected the Old Testament entirely and created his own canon consisting of a shortened version of Luke's Gospel (purged of Jewish elements) and ten of Paul's letters.^{24, 25} The church excommunicated Marcion around 144 CE, but his movement spread widely and persisted for centuries.²⁴

Marcion's influence on canon formation is debated among scholars. The traditional view, articulated by Adolf von Harnack in the early twentieth century, held that Marcion was the first to create a defined New Testament canon and that the church responded by developing its own counter-canon.²⁵ More recent scholarship, particularly John Barton's work, has moderated this view, noting that collections of Christian writings were already circulating before Marcion and that the church's response was not simply reactive.²⁶ Nevertheless, Marcion clearly highlighted the need for the church to articulate which writings were authoritative.²

Criteria for canonicity

Early church leaders developed several criteria for evaluating which books should be included in the canon. These criteria were not always applied consistently, and different communities weighted them differently, but they provide a framework for understanding how decisions were made.^{2, 27}

The most important criterion was apostolic origin or connection. A book was more likely to be accepted if it was believed to have been written by an apostle (such as Matthew, John, Paul, or Peter) or by someone closely associated with an apostle (such as Mark with Peter, or Luke with Paul).²⁷ This criterion explains why the four canonical Gospels bear the names of apostles or apostolic associates, while the Gospel of Thomas, despite its attribution to the apostle Thomas, was rejected partly because its content was judged incompatible with apostolic teaching.²

A second criterion was orthodoxy or theological consistency—whether a book's teaching aligned with the "rule of faith" (regula fidei), the core doctrines that churches across the Roman world held in common.²⁸ Books perceived as teaching Gnostic, docetic, or otherwise heretical doctrines were excluded. The Gospel of Peter was initially used by some churches but was later rejected after Bishop Serapion of Antioch (around 200 CE) read it carefully and found it contained docetic Christology (the view that Jesus only appeared to have a physical body).²⁹

A third criterion was widespread use (catholicity)—whether a book was recognized and read by churches throughout the Christian world, not just in one region.²⁷ This criterion favored books that had achieved broad distribution and acceptance over those that remained local. It also meant that the canon-formation process was inherently conservative: books that had been used for generations in worship and teaching had an advantage over newer or less familiar texts.²

Early canon lists and the Muratorian Fragment

The earliest surviving list of New Testament books is the Muratorian Fragment, a damaged Latin manuscript discovered by Ludovico Antonio Muratori in the eighteenth century.³⁰ The fragment is traditionally dated to the late second century (around 170–200 CE), though some scholars have argued for a fourth-century date.^{30, 31} It lists most of the books that would become the New Testament canon, including the four Gospels, Acts, thirteen letters of Paul, Jude, two letters of John (possibly including 1 John implicitly), and Revelation.³⁰

Notably, the Muratorian Fragment omits Hebrews, James, 1 and 2 Peter, and possibly 3 John—books that would eventually be included in the canon.³⁰ It also accepts the Wisdom of Solomon and the Apocalypse of Peter, noting that "some of us are not willing that the latter be read in church."³⁰ This ambivalence illustrates how fluid the boundaries of the canon remained even as a core collection was solidifying.²

Other early witnesses to the developing canon include Origen of Alexandria (c. 185–254 CE), who distinguished between books that were universally accepted (homologoumena), books that were disputed (antilegomena), and books that were rejected as heretical.³² Eusebius of Caesarea (c. 260–339 CE), in his Ecclesiastical History, provides the most detailed early discussion of canon, categorizing books as "recognized" (the four Gospels, Acts, Paul's letters, 1 Peter, 1 John, and conditionally Revelation), "disputed" (James, Jude, 2 Peter, 2 and 3 John), and "spurious" (Acts of Paul, Shepherd of Hermas, Apocalypse of Peter, Epistle of Barnabas, Didache, and conditionally Revelation).³³ Eusebius's inclusion of Revelation in both the "recognized" and "spurious" categories reflects the ongoing controversy over that book.³³

Athanasius and the church councils

The first list that exactly matches the twenty-seven books of the modern New Testament comes from Athanasius, bishop of Alexandria, in his Thirty-Ninth Festal Letter of 367 CE.³⁴ In this annual letter to the churches under his jurisdiction, Athanasius listed the canonical books and distinguished them from the "apocrypha," which he said were "an invention of heretics."³⁴ He also identified a third category of books that, while not canonical, were "appointed by the Fathers to be read," including the Wisdom of Solomon, Sirach, Esther, Judith, Tobit, the Didache, and the Shepherd of Hermas.³⁴

Athanasius's list was influential but not immediately universal. The churches in Syria continued to use the Diatessaron (a harmony of the four Gospels) and had a shorter canon that omitted 2 Peter, 2 and 3 John, Jude, and Revelation until the fifth or sixth century.³⁵ The Ethiopian church developed its own distinctive canon that includes books found in no other Christian tradition, such as 1 Enoch and Jubilees.¹⁸

The councils of Hippo (393 CE) and Carthage (397 and 419 CE), regional North African synods, formally listed the books of the biblical canon, matching Athanasius's list for the New Testament and including the deuterocanonical books for the Old Testament.^{16, 36} These councils were not ecumenical (representing all of Christendom) but regional, and their decisions reflect the practice of the Western church.¹⁶ The Eastern church's canonical lists varied, with some including books rejected in the West and others omitting books accepted there.³⁵

It is important to note that these councils did not create the canon by fiat; rather, they recognized and ratified a collection that had achieved broad (though not universal) acceptance through centuries of use.^{2, 27} As Bruce Metzger observed, the church "did not so much create the canon as recognize it."²⁷ However, this formulation can obscure the extent to which the recognition involved choices—choices to include certain books and exclude others, choices that were contested and could have gone differently.⁴

Timeline of New Testament canon development^{2, 27, 34}

Disputed and rejected texts

The canonical history of the New Testament reveals that several books now considered scripture were disputed for centuries, while other books once read as scripture were eventually excluded. Understanding this fluidity is essential for grasping how the canon was formed.^{2, 33}

The antilegomena: disputed books now in the canon

The term antilegomena (Greek for "spoken against") refers to books whose canonical status was disputed in the early church.³³ Eusebius listed five such books that are now in every New Testament: James, Jude, 2 Peter, 2 John, and 3 John.³³ To this list we can add Hebrews and Revelation, both of which faced significant opposition.²

Hebrews was accepted in the East from an early date, largely because it was attributed to Paul, but the Western church remained skeptical until the fourth century.³⁷ The letter's anonymous authorship (it does not name its author) and its statement that those who fall away cannot be renewed to repentance (Hebrews 6:4–6) raised concerns.³⁷ Origen famously wrote, "Who wrote the epistle, God knows the truth."³²

Revelation (the Apocalypse of John) had an especially tumultuous canonical history. It was widely accepted in the West but viewed with suspicion in the East.³⁸ Bishop Dionysius of Alexandria (mid-third century) questioned its apostolic authorship on stylistic grounds, noting that its Greek was quite different from the Gospel of John.³⁸ The book's vivid imagery was also controversial; some feared it encouraged excessive apocalyptic speculation.³⁸ Many Eastern churches excluded Revelation from their lectionaries (lists of readings for worship) until well into the medieval period, and the Syrian church did not accept it until the sixth century.³⁵

2 Peter is often identified as the latest book in the New Testament canon and the most disputed.³⁹ Scholars widely regard it as pseudonymous (written in Peter's name by a later author), composed perhaps as late as 120–150 CE.³⁹ Jerome noted in the fourth century that many doubted its authenticity because of its stylistic differences from 1 Peter.⁴⁰ Despite these doubts, 2 Peter was eventually included, illustrating how attribution to an apostle could override stylistic concerns.²

Books once read as scripture but now excluded

On the other side of the divide, several books that were read as scripture by early Christian communities were eventually excluded from the canon. These include the Shepherd of Hermas, the Didache, the Epistle of Barnabas, the Apocalypse of Peter, and 1 Clement.^{2, 23}

The Shepherd of Hermas, a Christian apocalyptic work composed in Rome around 100–160 CE, was immensely popular in the early church.⁴¹ It appears in Codex Sinaiticus (fourth century), one of the oldest and most important biblical manuscripts, following the New Testament books.⁴¹ Irenaeus, Clement of Alexandria, and Origen all cited it as scripture.⁴¹ The Muratorian Fragment, while acknowledging its value, excluded it from the canon because it was written "very recently, in our own times" and not by an apostle or prophet.³⁰

The Didache (Teaching of the Twelve Apostles), a manual of church order and ethics probably dating to the first or early second century, was listed by Athanasius among books "appointed by the Fathers to be read" for instruction of converts.³⁴ It provides valuable evidence for early Christian practice but was never widely regarded as canonical scripture.⁴²

The Apocalypse of Peter, a Christian apocalypse describing visions of heaven and hell, was accepted by some churches and appears in the Muratorian Fragment's list.^{30, 43} Clement of Alexandria wrote a commentary on it.⁴³ However, it was gradually excluded as the church standardized its apocalyptic literature around the Revelation of John.⁴³

The Epistle of Barnabas, an early second-century treatise on the relationship between Christianity and Judaism, was included in Codex Sinaiticus—one of the earliest complete Greek Bibles—alongside the canonical books, but was ultimately excluded from the canon.⁵⁶ 1 Clement, a letter from the church at Rome to the church at Corinth written around 96 CE, was highly regarded and read liturgically in some churches; it appears in some early canon lists and in Codex Alexandrinus, but was likewise judged edifying rather than canonical.⁵⁷ The Gospel of Peter, known from a fragmentary manuscript discovered in 1886, includes a docetic portrayal of Jesus's crucifixion and a dramatic resurrection narrative; Eusebius explicitly rejected it as "forged."⁵⁸

The Gospel of Thomas, a collection of 114 sayings attributed to Jesus discovered at Nag Hammadi in 1945, presents a more complex case.⁴⁴ While some scholars argue that it preserves early independent traditions about Jesus, possibly as old as the canonical Gospels, the Gospel of Thomas was associated with Gnostic Christianity and was explicitly condemned by early church fathers such as Hippolytus and Origen.^{44, 45} Its exclusion reflects both theological concerns about its content and the growing consensus around the four-Gospel canon.²

Different canons in different traditions

The canon-formation process did not produce a single, universally agreed-upon Bible. Different Christian traditions emerged with different canons, and these differences persist to the present day.^{17, 18}

The Protestant canon, following the decisions of the sixteenth-century Reformers, contains sixty-six books: thirty-nine in the Old Testament (matching the Jewish canon) and twenty-seven in the New Testament.¹⁴ The deuterocanonical books are sometimes included in Protestant Bibles as "Apocrypha" but are not considered canonical scripture.¹⁴

The Roman Catholic canon contains seventy-three books: forty-six in the Old Testament (including the seven deuterocanonical books and additions to Esther and Daniel) and twenty-seven in the New Testament.¹⁵ This canon was formally defined at the Council of Trent in 1546 in response to Protestant rejection of the deuterocanonical books.¹⁵

The Eastern Orthodox churches generally follow a canon similar to the Catholic one but with some variations.¹⁷ The Greek Orthodox Church includes 1 Esdras, the Prayer of Manasseh, Psalm 151, and 3 Maccabees, while the Russian Orthodox Church follows the Greek tradition.¹⁷ The Slavonic Bible used by some Orthodox churches includes 4 Maccabees as an appendix.¹⁷

The Ethiopian Orthodox Tewahedo Church has the largest canon of any Christian tradition, with eighty-one books in its "broader canon."¹⁸ In addition to the books found in the Catholic and Orthodox canons, the Ethiopian canon includes Enoch (1 Enoch), Jubilees, and other books unique to the Ethiopian tradition.¹⁸ The inclusion of 1 Enoch is particularly significant because this apocalyptic text is quoted as scripture in the New Testament epistle of Jude (Jude 14–15).¹⁸

The Syriac churches historically used a New Testament canon that differed from the Greek and Latin traditions.³⁵ The Peshitta, the standard Syriac Bible, originally contained only twenty-two New Testament books, omitting 2 Peter, 2 John, 3 John, Jude, and Revelation.³⁵ These books were added later in some Syriac traditions but are still not universally accepted.³⁵

Number of books in different biblical canons^{17, 18}

Scholarly understanding of canon formation

Modern scholars understand canon formation as a thoroughly human process, shaped by theological, political, liturgical, and historical factors.^{2, 4} This does not necessarily mean that the process was arbitrary or that the resulting canon lacks authority; many scholars who are themselves believers hold that the Spirit guided the process.²⁷ But the historical evidence makes clear that the canon emerged through human decisions made over centuries, not through a single divine decree or a moment of supernatural revelation about which books belonged in the Bible.^{2, 4}

Lee Martin McDonald, author of The Biblical Canon: Its Origin, Transmission, and Authority, emphasizes the contingency of the process: the church's recognition of certain books "was not achieved without considerable debate and even some confusion."⁴ Different communities made different decisions, and the boundaries of the canon remained fluid for centuries longer than is commonly recognized.⁴

Harry Gamble, in The New Testament Canon: Its Making and Meaning, notes that the criteria for canonicity were applied inconsistently and post hoc. The church did not first articulate criteria and then apply them rigorously; rather, the criteria were developed to justify decisions that had already been made on other grounds, primarily widespread liturgical use and perceived orthodoxy.⁴⁶

Bart Ehrman, in Lost Christianities, emphasizes that the canon reflects the triumph of one form of Christianity (what became "orthodoxy") over other forms (later labeled "heresies").²³ The books that were included were those used by the communities that ultimately prevailed in the theological debates of the second through fourth centuries; the books that were excluded were often associated with groups that lost those debates.²³ From this perspective, the canon is as much a record of which Christianity won as it is of which books were "truly" inspired.²³

F. F. Bruce, a conservative evangelical scholar, offered a somewhat different perspective in The Canon of Scripture. While acknowledging the human elements of the process, Bruce emphasized that the books that became canonical had an inherent authority that the church recognized rather than conferred: "What imposed them was the authority of the Lord [Christ] himself."⁴⁷ This view, while respecting the historical evidence, interprets it through the lens of faith in divine providence guiding the process.⁴⁷

Implications for biblical authority

The history of canon formation has significant implications for how one understands biblical authority. If the canon was not self-evident but required centuries of discernment, debate, and decision-making by fallible human beings, then the authority of the Bible cannot rest simply on a claim that God dictated which books belonged in it.⁴

For traditions that emphasize church authority (such as Roman Catholicism and Eastern Orthodoxy), this poses less of a problem: the church, guided by the Spirit, determined the canon, and the church's authority guarantees the canon's validity.¹⁵ For traditions that emphasize sola scriptura (scripture alone), the situation is more complex: the Bible's authority is said to be independent of the church, yet the church determined which books would be in the Bible in the first place.^{4, 14}

The diversity of canons across Christian traditions also raises questions. If the sixty-six-book Protestant canon is the true Bible, why did the church use a larger canon for most of its history? If the seventy-three-book Catholic canon is correct, how do we account for the Ethiopian church's even larger canon? The historical evidence does not point to a single, obvious answer; rather, it reveals a range of reasonable conclusions reached by different communities applying different criteria.^{17, 18}

Scholars such as John Barton have argued that the concept of a closed canon may itself be somewhat artificial.²⁶ In practice, the Bible has always been read alongside other authoritative texts—creeds, liturgies, theological treatises—and its interpretation has always been shaped by tradition.²⁶ The sharp line between "canonical" and "non-canonical" may obscure the more complex reality of how sacred texts actually function in religious communities.²⁶

What is clear from the historical record is that the biblical canon was not delivered ready-made from heaven. It emerged through a centuries-long process of use, debate, and gradual recognition, shaped by the theological concerns, liturgical practices, and institutional interests of the communities that transmitted and interpreted these texts.^{2, 4} Understanding this process enriches our reading of the Bible by placing it in its historical context—as a collection of ancient texts that communities of faith deemed authoritative, preserved, and transmitted across generations.²⁷