[This entry contains four subentries, an Overview, Hebrew Bible, Old Testament, and New Testament.]


The word “canon” (Gk. kanōn; cf. Eng “cane”) originally meant a staff or measuring stick. Later it could refer to a guide or rule (e.g., 4 Macc 7:21 “whole rule of philosophy;” Gal 6:16 “who walk by this rule;” 1 Clem. 7:2 “holy rule of our tradition”). In time the word was applied to authoritative writings: “No psalms of private authorship can be read in the church, nor uncanonical books, but only the canonical books of the Old and New Testaments” (art. §59, Council of Laodicea, 363 C.E.).

The emergence of the Jewish, Samaritan, and Christian biblical canons was a gradual and largely ad hoc process. Although there are general references to “law” and “prophets” in old traditions (cf. 2 Kgs 17:13), the emergence of the Jewish canon of scripture is a later development. In all probability it did not begin until the time of Judea's national and religious restoration following the exile to Babylon. Not until the end of the sixth century or the beginning of the fifth century B.C.E. was the Torah (“teaching” or “law”) of Moses recognized as authoritative, though the familiar five-book arrangement (Genesis–Deuteronomy) was probably a somewhat later development. Still later a number of prophetic books were recognized, so that in time people could speak of “the Law and the Prophets” (2 Macc 15:9; 4 Macc 18:10; T. Levi 16:2; Matt 5:17; Rom 3:21).

The tripartite form of the Jewish canon of scripture (i.e., Law, Prophets, Writings) seems to have its beginnings in the second century B.C.E. Writing circa 180 the sage Jesus Ben Sira spoke of the “study of the law of the Most High” and “prophecies” (Sir 39:1), but his grandson, who translated his Hebrew work into Greek circa 132 B.C.E., spoke of “the law and the prophets and the other books of our fathers” (Prologue). It is not clear from the expression “other books” precisely which books are meant. Indeed, we cannot be sure which “prophets” were in view. Some scholars think that the tripartite form of the canon is alluded to in the Halakic Letter from Qumran, which speaks of “the book of Moses, the books of the Prophets, and David” (4QMMT C10; cf. “The law of Moses, the prophets, and the psalms,” Luke 24:44), but that is far from clear.

In the first century C.E. Jewish writers began to make reference to the number of books that make up authoritative scripture. According to Josephus (37–c. 100 C.E.), Jews recognized twenty-two books: five books of Moses, thirteen prophets, and four books that contain hymns and instructions for living (Ag, Ap, 1.37–41). These twenty-two books are probably the equivalent of the twenty-four mentioned by others (see Origen apud Eusebius, Hist. eccl. 6.26; Jerome, Preface to Latin translation of Samuel and Kings). The twenty-two tradition is echoed in some versions of Jubilees, in a typology based upon the number of letters in the Hebrew alphabet: “As there were two and twenty letters and two and twenty books and two and twenty heads of humankind from Adam to Jacob…” (2:23–24).

At the end of the first century the author of 4 Ezra ( = 2 Esdras) refers to twenty-four “public” books and seventy books “written last” that are to be held back from the public and read only by scholars (14:44–48). Here the twenty-four books refer to the Jewish canon of scripture, while the seventy refer to the books of the Apocrypha and Pseudepigrapha, though exactly which ones we do not know. The twenty-four books of scripture are listed, with commentary, in the Babylonian Talmud (cf. b. B. Bat. 14b–15a). The twenty-four books are (1) the five books of Moses, (2) the eight books of prophecy (i.e., Joshua, Judges, Samuel, and Kings—the last two counted as single not double volumes—Isaiah, Jeremiah, Ezekiel, and the Twelve), and (3) the eleven holy books or writings (i.e., Psalms, Job, Proverbs, Ruth, Canticles, Ecclesiastes, Lamentations, Esther, Daniel, Ezra-Nehemiah, and Chronicles—counted as single not double). Counted separately, these books number thirty-nine.

In contrast to the Jewish canon, the Samaritan canon of scripture is limited to the five books of Moses (i.e., the so-called Samaritan Pentateuch), whose Hebrew text is similar to the Masoretic Text, with a large number of mostly minor variants. However, there are other works of great historical and religious significance to the Samaritans. These include a work called the Memar Marqah (“teaching of Marqe”), which offers poetical commentary on the Pentateuch, and various chronicles (such as the Book of Joshua), which provide accounts of Samaritan origins.

The beginnings of the Christian canon of scripture may be traced to the second century, as disputes arose over matters of teaching and apostolic authority. The appearance of a number of gospels and gospel-like writings in the second century prompted leading Christian scholars (e.g., Irenaeus) to insist on the four first-century Gospels of Matthew, Mark, Luke, and John. Marcion's call for the acceptance of only some of Paul's letters, a truncated version of Luke, and none of the books of Jewish scripture provoked debate but was rejected.

Although most of the books that eventually would be included in the New Testament were read by most churches by the end of the second century, the canon of Christian scripture was not formally recognized until much later. The Jewish canon of scripture was accepted, under the rubric “Old Testament,” but many of the books of the Apocrypha retained a quasi-canonical status and were included in some of the oldest Greek Bibles of the church. Jerome himself, reluctantly, included these books as an appendix in the Latin translation of scripture, which he oversaw.

Why were some books regarded as authoritative and, eventually, received as canonical scripture? A number of criteria have been suggested. One criterion seems to have been very important, in both Jewish and Christian circles, and that is prophecy. Besides the prophets themselves, Moses and David were also regarded as prophets. For Christians, Jesus and his apostles possessed prophetic authority. Because prophecy was understood to be prompted by the Spirit of God, prophetic material was understood to convey the very word of God.

Today the Catholic canon of scripture includes the twenty-seven books of the New Testament, the thirty-nine books of the Old Testament, and twelve books of the Old Testament Apocrypha. Because some of the latter books are additions to books of the thirty-nine (e.g., Esther and Daniel) and because two of them are combined as one (Baruch and Epistle of Jeremiah) the total count comes to forty-six.

The canons of scripture of the Syrian, Greek Orthodox, and Coptic churches are somewhat longer still, for they include additional books of the Apocrypha and in the case of the Coptic church the book of Enoch. Most Protestant denominations do not recognize any of the books of the Apocrypha or Pseudepigrapha.



  • Barton, J. How the Bible Came to Be. Louisville, Ky.: Westminster John Knox, 1997.
  • Elliott, J. K. “Manuscripts, the Codex and the Canon.” Journal for the Study of the New Testament 63 (1996): 105–123.
  • McDonald, L. M. The Biblical Canon: Its Origin, Transmission, and Authority. Peabody, Mass.: Hendrickson, 2007.
  • McDonald, L. M. Forgotten Scriptures: The Selection and Rejection of Early Religious Writings. Louisville, Ky.: Westminster John Knox, 2009.
  • Metzger, B. M. The Canon of the New Testament: Its Origin, Development, and Significance. Oxford: Clarendon, 1987.
  • Pelikan, J. Whose Bible Is It? A History of the Scriptures through the Ages. New York: Viking, 2005.

C. A. Evans

Hebrew Bible

In order to properly investigate the concept of the canon of the Hebrew Bible, we must first define several terms. “Hebrew Bible” is a modern designation given to the literature of the Jewish scriptures and the first portion of the Christian canon, ordinarily referred to by Christians as the Old Testament. The medieval designation for scripture in Judaism is Tanak, an acronym taken from its three primary sections, or in earlier rabbinic literature “the holy writings” (kitvei ha-Kodesh) or “that which is read” (Mikra')—both expressions that distinguish the biblical books from all other works. The term “Hebrew Bible,” then, is not a historical designation used by either Jewish or Christian communities; it is an attempt to offer a descriptive word that is more neutral than “Old Testament,” a term with an overtly supersessionist nuance.

Hebrew Bible (or Hebrew Scripture[s]) is not altogether without difficulties, however. The corpus described by the designation “Hebrew Bible” includes some sections in Aramaic (parts of the books of Daniel and Ezra as well as very short passages in Genesis and Jeremiah), so this designation is not strictly accurate. Additionally, some scholars think that the term distances the canon from both the Jewish and the Christian communities that consider it to be authoritative (Seitz 1998). Nevertheless, the term Hebrew Bible has become a relatively standard way of referring to the twenty-four books considered to be authoritative in Jewish tradition.

The word “canon” is from the Greek word kanōn, meaning a “rule” or “standard,” which itself may have originated from a Semitic root, one that has cognates in Hebrew, Akkadian, and Ugaritic; for example, the Hebrew word qāneh in its most basic sense means “reed.” In several passages of the Hebrew Bible reeds are used as measuring sticks, thus giving the Hebrew word a broader semantic range of possibilities. In one of Ezekiel's visions, for example, he describes a reed being used to determine distance: “There was a wall all around the outside of the temple area, and the length of the measuring reed (qāneh) in the man's hand was six long cubits” (Ezek 40:5; see also 1 Kgs 14:15; Job 40:21). The Greek use of the word was extended metaphorically to indicate any kind of guide by which to evaluate the quality of other things. Thus, “canon” could mean something like a touchstone or litmus test.

Early Christians adopted the Greek sense of the word in different contexts to indicate various types of guidelines. For example, Paul uses “canon” in this way when referring to his own teaching about circumcision, offering a blessing to those who adhere to it: “As for those who will follow this rule (kanōn)—peace be upon them, and mercy, and upon the Israel of God” (Gal 6:16). A related meaning of the word is found in 2 Corinthians where Paul speaks of “canon” as a limit to boasting (2 Cor 10:13). Origen also refers to “canon” in a similar fashion to indicate a rule of faith.

One use of the word came to designate a standard group of literary works judged or “ruled” to be especially significant. The guideline provided by this use of canon did not indicate that the literature in question was a complete and closed list, but rather a selected body that was of highest quality. Eventually the term “canon” also came be used to distinguish such literature from texts that were categorically excluded, a use that was first employed by Christian authors such as Athanasius in the fourth century C.E. Thus historically the term canon has come to mean a group of texts that are religiously authoritative, or holy, to the exclusion of other texts. It is used anachronistically when discussing the Hebrew Bible, since it originated in discussions of the Christian canon, and is never used in rabbinic literature. Nevertheless, it is a convenient term to retain.

Principal Divisions of the Canon.

According to rabbinic tradition, the Hebrew Bible has a tripartite arrangement—Torah (“Instruction or Law”), Neviʾim (“Prophets”), and Ketuvim (“Writings”)—but it is not clear how this came to be. One position advanced by scholars (e.g., Barton) in recent decades is that the canon started out as a bipartite arrangement—Torah and everything else—the latter of which was subsequently subdivided to include two distinct sections, leaving us with a tripartite arrangement. Certainly the designation, “The Law and the Prophets” is used in a number of instances in the New Testament, reflecting Jewish practices of the first century C.E. For example, in Matthew, Jesus is quoted as saying, “Do not think that I have come to abolish the Law and the Prophets” (Matt 5:17; see also 2 Macc 15:9; 4 Macc 18:10; Luke 16:16; and Rom 3:21).

Other scholars, however, would point to several texts that seem to challenge this argument, including references in the New Testament and other literature to three distinct sections. For example, in Luke 24:44 Jesus says, “These are my words that I spoke to you while I was still with you—that everything written about me in the law of Moses, the prophets, and the psalms must be fulfilled”; this may be paralleled in a Qumran text, which apparently refers to “[the book of Moses] and the books of the prophets and Davi[d]” (4QMMTc 10). In the next verse, we are told that Jesus then “opened their minds to understand the scriptures” (v. 45), an indication that by the time of Luke's gospel, likely written around 85 C.E., scripture is probably described as having three primary parts. It is not only Christian writers, however, who refer to a tripartite scripture. We find similar language in the late second-century B.C.E. prologue to the translation of Sirach, which states, “Many great teachings have been given to us through the Law and the Prophets and the others that followed them.” Similarly, Josephus indicates that twenty-two books are arranged in a three-part order (C. Ap. 1.8.38–39). It is clear from this evidence that there were at least two groups, Law and Prophets, which were considered authoritative, while a final third category remained somewhat undefined (Evans 2002).


The Torah comprises the first five books of the Bible: Genesis, Exodus, Leviticus, Numbers, and Deuteronomy, a group that both Jews and Christians agree comprise a unit, always in this order. The word “Torah” is connected to the verbal root yārāh, which means “to hit a target.” The noun formed from the root means “instruction,” “law,” or “direction.” The word Torah is used in both a general sense, to refer to the entire body of sacred scripture, and also in a specific way to refer just to the first five the books, the Pentateuch.

Most scholars agree that the Torah was the first group of texts to receive authoritative status in the community. Many have associated this development with the person of Ezra. The book of Ezra describes him as a scribe “skilled in the law of Moses that the LORD the God of Israel had given” (Ezra 7:6). Later we are told that the law that Ezra brought with him on his return is a physical document, when it is referred to as “the law of your God which is in your hand” (Ezra 7:14). These hints in the text suggested to scholars that Ezra played a central role in the canonization of the Torah. There is good evidence to suggest that the Torah became an important part of life in the Persian period but to definitively tie canonization to the particular moment of Ezra's return certainly goes beyond what the biblical text can tell us.


Prophets, or “Neviʾim,” comprises two parts: the Former Prophets and the Latter Prophets. The first, the Former Prophets (“Neviʾim Rishonim”), the books of Joshua, Judges, Samuel, and Kings (always in that order, which is chronological), are counted among the “historical books” in the Christian canon. These are primarily narrative accounts detailing the history from the period of the conquest through the fall of the Judah to the Babylonians, namely the Iron Age (both Iron Age I and Iron Age II). These accounts contain stories about important prophetic figures, such as Deborah, Samuel, Nathan, and Elijah, who play a role in the history of Israel—these may have evoked the title “Former Prophets.” Unlike the prose third-person accounts recorded in the Former Prophets, the Latter Prophets contain a great deal of poetry as well as some prose sections, and are named according to the prophet with whom they are associated. The Latter Prophets (“Neviʾim Aharonim”), sometimes also known as the Writing Prophets, are grouped into two sections. The first of these contains the three “Major Prophets”: Isaiah, Jeremiah, and Ezekiel, in chronological order; in the Babylonian Talmud (b. B. Bat. 14b) the order is Jeremiah, Ezekiel, and Isaiah. They are “major” prophets because of the length of the books. The Book of the Twelve is the name given to the twelve shorter prophetic books, the “Minor Prophets,” traditionally written on one scroll, a practice attested among the Dead Sea Scrolls. The Twelve are arranged in a loosely chronological order from Hosea to Malachi, spanning a period from the eighth century B.C.E. through the early Persian period.

Based on a variety of designations in the Second Temple period referring to the “Law and the Prophets,” the dominant scholarly opinion is that this group of texts was the second to be considered an authoritative text. This is further supported by the tradition in Judaism that the period of prophecy extended only until the early Persian period (b. Soṭah 48b), an era after which there would be no further prophetic revelation.


Within the category of Writings (“Ketuvim”), there are a variety of genres, a fact which has led many scholars to conclude that this final grouping is a kind of “catch-all” category into which all the remaining books were gathered. The title for this final group of books did not stabilize until the first century C.E. Prior to that, there are a variety of designations, including “the Psalms” (Luke 24:44) and “the rest” (Prologue to Sirach). The various ways in which this category is designated, the relatively late date of many of the books in it, and the fluidity in the order of the books, has led to the conclusion that this category was the last to be considered authoritative. The books included in this group are Psalms, Proverbs, Job, the Five Scrolls—Song of Solomon, Ruth, Lamentations, Ecclesiastes, and Esther—Daniel, Ezra-Nehemiah, and Chronicles. In the Hebrew Bible, Ezra-Nehemiah is one book, although in the Christian Bible it is separated into two. The Five Scrolls (Megilloth) are grouped together because each of these five books is read at a major festival: the Song of Songs at Passover; Ruth at the festival of Weeks; Lamentations on Tisha B'Ab; Ecclesiastes at the festival of Booths; and Esther at Purim. (This grouping is relatively late; in earlier sources the five scrolls are not always together.) Daniel is the latest book found in the Writings. This is further evidence that the Jewish tripartite division of the canon is probably older than the Christian arrangement discussed below. Had Daniel been written earlier, it would have been included among the Prophets, especially since he is considered a prophet in rabbinic literature. However, since Daniel was written in the second century B.C.E., the book was not included in the “Prophets” section because it was already closed. Similarly, Ezra-Nehemiah and Chronicles, most likely written in the fourth century B.C.E., are similar in type to some of the books found in the Former Prophets, a category to which they could not be added due to the late date of their composition.

Different Canons.

Jews and Christians disagree both on the number of books in the Hebrew Bible canon and on their arrangement. In Jewish tradition, the Bible has twenty-four books arranged into three parts: the Torah (“Law”), Neviʾim (“Prophets”) and Ketuvim (“Writings”). The Christian arrangement of these same books, however, reflects a very different order and is divided in such a way to bring the total number of books to thirty-nine. Each of the Minor Prophets is counted as a separate book, and the books of Samuel, Kings, Chronicles, and Ezra-Nehemiah are each divided into two separate books. Although the New Testament references the traditional Jewish three-part order, this is not how the Christian Old Testament canon is arranged. Rather it follows a four-part division: the Pentateuch, the Historical Books, the Poetical and Wisdom Books, and the Prophetic Books. Both Jews and Christians agree concerning the Torah/Pentateuch as an initial unit—this is not surprising since it narrates the earliest events of the Bible, and is typically seen as first among equals in content. The second group of books for Christians, the Historical Books, includes the books of Joshua–2 Kings, the Former Prophets of the second part of the Jewish canon; to these are added the books of Ruth (between Judges and 1 Samuel because of its narrative chronology), Chronicles, Ezra, Nehemiah, and Esther, all of which occur in the Writings, the third part of the Jewish canon. The Poetical and Wisdom Books (Job, Psalms, Proverbs, Ecclesiastes, and Song of Solomon) also come from the Writings. The last section in the Christian arrangement is the Prophetic Books, which included all the Latter Prophets of the Jewish canon, to which were added, also from the Writings, the book of Lamentations, placed after Jeremiah because he was thought to be its author, and the book of Daniel, because the New Testament explicitly names him as a prophet (Matt 24:15; see also Mark 13:14).

The Christian arrangement of these books, although perhaps originally an alternative order in some Jewish traditions, seems to reflect a specific theological stance, one which orients readers at the close of the Old Testament to anticipate the New Testament. Thus, the first group of books relate to events in the past, including the Pentateuch and the Historical Books. The Poetical and Wisdom Books deal roughly with the present; they provide material for everyday life, by offering such resources as advice on how to live and language that can be used in prayer. Finally, the Prophetic Books, placed in the final position of the canon before the New Testament, anticipate the life of Jesus. The gospel of Matthew explicitly quotes several prophetic books as fulfilled in the life of Jesus, including Malachi, the last book of the Old Testament canon, which ends with the promise of a messenger who will come to prepare the way (Mal 3:1; see Matt 11:10 and cf. Mark 1:2; Luke 7:27).

Greek culture shaped Christianity in many important ways, influencing the literature that Christians produced, most of which was written in Greek. Christians adopted the Jewish scriptures—books that were frequently referenced and quoted throughout the New Testament—as the first part of their canon, although they often used the Septuagint, the ancient Greek translation of the Hebrew Bible, as their authoritative text. (When the Old Testament is quoted in the New Testament, it is the Septuagint version that is usually quoted.) Along with the books that were canonical in Judaism, however, Christians also included a number of other important Jewish texts that had been written or at least preserved in Greek, including Tobit, Judith, 1 and 2 Maccabees, the Wisdom of Solomon, Sirach, Baruch, and longer versions than the Hebrew of both Esther and Daniel. For the Jewish community, these books (with the possible exception of Sirach among some groups) did not have the same status as those that later became canonical. In the fifth century, Jerome recognized that these books were not a part of the Hebrew Bible. Thus, this group is often referred to as the “apocryphal” (“hidden”) or “deuterocanonical,” (“second canon”) books.

For most Christians, the Old Testament canon was mostly consistent up until the time of the Reformation, except for the additional inclusion of several texts in the Orthodox canon (Psalm 151, the Prayer of Manasseh, 1 and 2 Esdras, and 3 Maccabees). Martin Luther, who could read the Bible in the original languages, initiated a return of the Hebrew books of the Old Testament, a decision that has been affirmed by most Protestant denominations. Thus, currently there are four primary canons: the Jewish tripartite canon of twenty-four books; the Protestant canon including the same material but divided into thirty-nine books; the Roman Catholic canon which includes the deuterocanonical books; and the Orthodox canon which includes both the deuterocanonical books and the additional texts listed above.

Process of Canonization.

Nowhere in any of our sources is there a clear account that explicates the process of canonization. This must be reconstructed from various texts, most of which antedate the process considerably, that hint at the historical developments that finally produced a closed list of books. Problems of terminology accentuate the problem—because the word canon is not used in the particular sense we have mentioned until the fourth century C.E., it is often difficult to determine what is meant by various sources. Despite these limitations, it is clear that there was a long trajectory that eventually concluded with the application of the term canon to a fixed body of sacred texts, and that the Hebrew Bible was not canonized at one single time.

However, already within the Hebrew Bible there is evidence that certain words and writings were very highly regarded. Deuteronomy 24:16 is quoted as an authoritative text in 2 Kings 14:6. Moreover, according to 2 Kings 22–23, during the reign of Josiah, a scroll—later associated by scholars with the book of Deuteronomy—was found in the Temple. We are told that Josiah considered this “book of the law” to be of such high status that he had it read aloud in the midst of all the people and made a number of reforms after having read it. Within the book of Deuteronomy itself, the high priority of Moses' words is evident throughout. For example, Moses admonishes the people assembled before him, saying, “You must neither add anything to what I command you nor take away anything from it, but keep the commandments of the LORD your God with which I am charging you” (Deut 4:2; cf. 12:32; Prov 30:6). These examples cannot provide evidence for any sort of fixed canon, but they do indicate that certain words and texts had a great deal of authority. The writers considered them to be of such high value that they held a central place in the life of the community, and felt that they should not be altered.

With the destruction of the Temple in 586 B.C.E., a number of institutions—including kingship and the Temple rituals—were abolished. In Psalm 137, the psalmist posed the anguished question, “How can we sing the LORD's song in a foreign land?” The question asked on behalf of those living in exile indicates a very real tension that was felt by the community as they recognized the need to find new ways to worship God in the absence of the material realities in Jerusalem. As followers of Yahweh looked to develop a new sense of religious identity, the role of texts likely took on a more important communal function. The need for scribes intensified as gathering and preserving texts for study and liturgical use took the place of the Temple rituals. Even though some exiles were able to return, there were many who continued to live in Diaspora communities.

The book of Ezra, most likely reflecting events of the fifth century B.C.E., describes a number of the changes that had transpired. We are told that Ezra is “a scribe skilled in the law of Moses” (Ezra 7:6), one who “had set his heart to study the law of the LORD.… and to teach the statutes and ordinances in Israel” (Ezra 7:10). According to the text, the Persian king authorized this scribe from a priestly family to return to Jerusalem and to bring with him the law. The king's edict dictated that those still living in the land would be bound by both “the law of God and the law of the king” under penalty of death (Ezra 7:26). Several key elements are evident in the book of Ezra: the central role of a scribe in the life of the community; the value placed on the law—either referred to as the “law of Moses” or the “law of God”; and the necessity of communicating the law to the community.

The collection of books by the writing prophets also indicates the great degree of authority accorded to their words. We have explicit accounts about the words of Jeremiah being recorded by a scribe, brought to the king, read aloud, and, when found to be upsetting, the text was burned (Jer 36). Furthermore, the prophets continually refer to the fact that the words they speak were not their own but from God, using phrases such as “thus says the LORD” (i.e., Isa 50:1), “the word of the LORD came to me” (Hos 2:21), and, “this is what the Sovereign LORD says” (Ezek 3:11). Sometimes prophets did not want to speak the words but felt under divine duress to deliver a message, a motif authenticating the divine origin of the words (see esp. Jer 20:9).

Other hints about the movement toward a closed canon are found in our sources. According to later Jewish tradition, prophecy was a phenomenon limited to a set period of time in Israel's history, one which came to a close around 400 B.C.E., after the time of the prophet Malachi (b. Soṭah 48b). In the first-century B.C.E. book of 1 Maccabees, the distress that is felt after the death of Judas is compared to another great source of sorrow in Israel's history—the time when prophets “ceased to appear among them” (1 Macc 9:27). The period of revelation, then, had a terminal point beyond which further prophetic speech did not continue. According to some scholars, the warning in Ecclesiastes regarding any texts beyond those mentioned explicitly because “of making many books there is no end” (Eccl 12:12) may mean that the scriptural books are already sufficient and do not need anything added to them.

There are several places that refer to groups of books which appear to be religiously authoritative, not merely important literary works. Mishnah Yadayim (late second century C.E.), for example, discusses whether the books of Song of Songs and Ecclesiastes “defile the hands” (m. Yad. 3.5). The origin of this exact phrase is debated, though it likely is synonymous with the current use of the word “canon,” or at least works that have a sacred status, separate and apart from more profane use. It is unclear, however, if this passage, and the ones noted below, recount actual historical debates reflecting on the canonicity of particular books, or if they are later scholastic recreations meant to explain why certain books, despite their problems, are part of the canon.

Another critical piece of evidence comes from the Babylonian Talmud (approx. fifth century C.E.), Baba Batra 14b–15a. This text discusses various books, most especially, the order of the books and who authored each text. Implicit in this document is an assumption that the status of a book is directly correlated to who authored it. That is, it carries a certain authority because it was written by a prophet. The logic of the arrangement of the books seems to be based mostly on a chronological scheme, with the possible exception of Job.

Septuagint (LXX).

According to the apocryphal Letter of Aristeas (probably late second century B.C.E.), seventy-two Jewish elders were charged by King Ptolemy II Philadelphus of Egypt (285–246 B.C.E.) with the translation of the Hebrew text of the Torah into Greek, which, functioning as a committee, they completed in seventy-two days. According to later retellings of this account, each one translated the text separately, but all of the translations agreed perfectly with one another. The agreement between seventy-two separate translations indicated the divine inspiration of the translation. Because seventy-two different scribes were responsible for the translating the law, the translation was given the name Septuagint, meaning “seventy” (LXX). Although few consider the story a historical account of the translation process, the letter serves an etiological function, providing the putative origin for the Greek translation. The purpose and function of the letter is not clear since it has been proven to be a fictive account that never functioned as an actual letter (Sundberg 2002, pp. 74–75). It does, however, appear to demonstrate a concern for the legitimacy of a translated text. Authority in this case is contingent upon the Greek text exactly replicating an authentic original Hebrew text. Although historical evidence indicates that there were a variety of textual traditions, including various Septuagint translations, there was at least an idealized notion that the text should not be altered. (See further Wasserstein and Wasserstein 2006.)


Evidence from the Dead Sea Scrolls found at Qumran and elsewhere both complicates and clarifies the picture. On the one hand, there is evidence that many of the same books of the current Hebrew Bible canon were in use in the community and were viewed as authoritative and sacred. On the other hand, the wide variety within the contents of the books and their arrangement has illustrated a greater degree of textual fluidity than was previously known. There is no known copy of the book of Esther at Qumran; however, multiple copies of Jubilees were found there, and its use in sectarian literature suggests that it was what would later be called canonical. Furthermore, there were multiple versions of the psalter, with variation both in number and arrangement of the psalms. These pieces of evidence indicate that although there may have been movement toward fixing the canon, there was still a great deal of flexibility and variation. It is precisely on the basis of the new information provided by the Dead Sea Scrolls that some scholars have tried to argue that there is nothing which can be called a canon during the period of Second Temple Judaism, and certainly not for the people of Qumran (VanderKam 2002).

History of Scholarship.

We have seen that the term canon is a late designation and that there are much earlier hints about the historical developments by which the books of the Hebrew Bible came to be a fixed and exclusive group of books—a corpus to which other authors, however important, could not add. Because much is not known about canonization, scholars have advanced different theories to account for the evidence that we do have. Most agree, however, that the canon developed gradually, as the result of a process, and was not the product of a single moment.

By the mid-twentieth century, a broad census had emerged regarding a three-stage development of the canon, a conclusion that was based on the work of nineteenth-century scholars (e.g., Graetz 1886; Buhl, 1892). Based on rabbinic discussions about various biblical books (esp. m. Yad. 3.5), scholars postulated that about 90 C.E. a rabbinic council at Jabneh (Greek Jamnia) established the final, authoritative form of the canon. According to this theory, the canon was closed in three stages: the Torah by 400 B.C.E., the Prophets by 200 B.C.E., and the final group, the Writings, was decided upon at the council in question in 90 C.E. The proposal about a council where key decisions regarding the canon were made was a hypothetical reconstruction based on an analogy to later Christian councils where definitive decisions about church doctrine and canon took place; scholars now generally recognize that this analogy is inappropriate. A second proposal, paralleling the notion of a canon fixed at Jabneh, saw a second Greek canon emerging out of the Egyptian exilic community. This theory postulated that there was one canon in Hebrew that represented the Judaism of Palestine; the other was written in Greek and represented a Hellenized Judaism. In 1958, the position that held to an Alexandrian canon was challenged (Sundberg 1958) and subsequently largely abandoned.

In the second half of the twentieth century, however, the consensus regarding Jabneh had been weakened on a number of different fronts. In 1964, an influential work demonstrated some of the problematic assumptions of the Jabneh consensus (Lewis 1964; see also Lewis 2002). It showed that there was no conclusive evidence for any sort of council, and although m. Yadayim does discuss a gathering, it is never called a council. In particular there is no evidence that the purpose of the meeting was to definitively delimit an authoritative canon. Furthermore, Josephus lacks any awareness of a great council at which an important decision was made. Finally, study of the Dead Sea Scrolls further undermined many of the earlier conclusions about canonization; see further above. The overturning of the Jabneh consensus has been nearly universal, though no widely accepted alternative hypothesis has developed. Some have suggested that the canon emerged as a result of the destruction of the Temple in 70 C.E. (e.g., Sanders 1972) while others place it earlier (e.g., Leiman 1976). It is likely that the Jewish canon was largely fixed before the rise of Christianity, which inherited it from the Jews as a more or less complete collection.

Attempts have been made to clarify the language used about canon. Sundberg questioned the use of canon as a synonym with “Scripture,” suggesting that they are not the same in the early sources (Sundberg 1958), and many texts from the end of the first century B.C.E. to the first several centuries B.C.E. do employ the term “scripture.” Current scholarship on canon continues to stress the need for specificity and clear definitions of terms in order for the field to move forward productively (Ulrich 2002). These developments in the study of canon had corollaries in other aspects of biblical studies. More attention has been given to the complex social and historical realities of Hellenized Judaism. There has also been increased interest in the study of noncanonical texts, including both pseudepigraphical and apocryphal works. Both reflect a history-of-religion perspective rather just than a theological one. Viewed in this light, both canonical and noncanonical texts are the result of historical realities, and absent a theological priority for the former, the priority of the canon has been called into question (Childs 2005). Another direction was undertaken beginning in the 1970s. Moving in a different direction from the history-of-religions approach to canon, a canonical approach to biblical studies was proposed (Childs 1979). This sought to consider the question of canon alongside of theological concerns and to place emphasis on the canonical (or full literary context) of scripture as the primary location for exegesis. Thus, the need to understand what the canon is and how it came to be cannot be underestimated.

Other scholars have looked to internal evidence within the Hebrew Bible to discern clues about the relative status of its various parts. Two relevant texts are Deuteronomy 34 and Malachi 4:4–6 (Heb. 3:22–24). One argument saw evidence within the final part of Deuteronomy (Deut 34:10–12) for a claim that any revelation that came after Moses (that is, the Prophets) was not on equal standing with the Law. The text states, “[Moses] was unequaled for all the signs and wonders that the LORD sent him to perform” (Deut 34:11A). According to this view, it was only later that the Prophets took on a status of canonical standing with the Pentateuch, a perspective that is affirmed in the conclusion of Malachi (Blenkinsopp 1977). Against this argument, it has been suggested that Deuteronomy and Malachi do not represent two opposite poles but rather that both texts affirm the centrality of both the Torah and the Prophets (Chapman 2000). Discussions such as this, when viewed along with extrabiblical evidence, provide a rich resource for further investigation of the process of canon formation.

Despite this variety in approaches, there has been a significant amount of agreement among scholars that the various parts of the canon came to be viewed as central and authoritative at different times. These are not, as scholars had thought in the past, associated with one particular historical event such as Ezra's return to Jerusalem or a putative council at a coastal town in Judea. Nor is it useful to casually apply the term canon to these groups of books without acknowledging that this specific use of the word canon is a rather late development. Scholars have thus come to view the canon as the result of a process of recognition by the community that the text held a central role. It is a process that is both deeply connected to cultural influences and dependent on authority (Davies 2002). Most scholars would agree on a terminus ad quem by which each of the three parts came to occupy a central place in Judaism: circa 400 B.C.E. for the Torah; circa 200 B.C.E. for the Prophets; and sometime in the first century C.E. for the Writings.

Nevertheless, in the past decade there have been several continued attempts to sort through the various aspects of the problem of canon, some of which have implicitly questioned the three-part arrangement on new grounds. Schniedewind (2004) focuses on questions about the process of writing and the means by which traditions are written down and become a book. Ultimately he still holds to three stages in the development of canon formation, but they occurred earlier, associated respectively with Hezekiah, the exile, and the Persian period (for a critique of this proposal, see Van Seters 2007). Carr (2005) also focuses on the literary aspects, but sees the role of scribes and their education that depended on a specific body of texts as the clue to understanding canonization. Certainly the role of scribal education is significant for canon formation, but Carr's thesis suffers from a lack of evidence about the role of specific texts in educating scribes (Van Seters 2007).

The result of the various questions and the variety of new evidence that has been brought to bear on the study of canon stresses the need for a greater sophistication in approaches. Terminology must be defined and agreed upon in order for conversations about canon to proceed productively. Work on canon formation must consider a variety of kinds of evidence and proceed with caution in the way that sources are interpreted, taking care not to overextend the limitations of the information it can provide. The role of the community, geography, the physical aspects of writing, material realities, social dynamics, and authority must all be considered together for the field to progress.



  • Alexander, Philip S., and Jean-Daniel Kaestli, eds. The Canon of Scripture in Jewish and Christian Tradition. Lausanne: Zèbre, 2007.
  • Barr, James. Holy Scripture: Canon, Authority, Criticism. Philadelphia: Westminster, 1983.
  • Barrera, Julio C. Trebolle. “Origins of a Tripartite Old Testament Canon.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 129–145. Peabody, Mass.: Hendrickson, 2002.
  • Barton, John. Holy Writings, Sacred Text: The Canon in Early Christianity. Louisville, Ky.: Westminster John Knox, 1997.
  • Barton, John. Oracles of God: Perceptions of Ancient Prophecy in Israel after the Exile. London: Barton, Longman, and Todd, 1986.
  • Barton, John. The Spirit and the Letter: Studies in Biblical Canon. London: SPCK, 1997.
  • Blenkinsopp, Joseph. “The Formation of the Hebrew Bible: Isaiah as a Test Case.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 53–67. Peabody, Mass.: Hendrickson, 2002.
  • Blenkinsopp, Joseph. Prophecy and Canon: A Contribution to the Study of Jewish Origins. Notre Dame, Ind.: University of Notre Dame Press, 1977.
  • Brettler, Marc Zvi. “The Canonization of the Bible.” In The Jewish Study Bible, edited by Adele Berlin and Marc Zvi Brettler, pp. 2072–2077. New York: Oxford University Press, 2003.
  • Buhl, Frants. Canon and the Text of the Old Testament. Edinburgh: T & T Clark, 1892.
  • Carr, David M. Writing on the Tablet of the Heart: Origins of Scripture and Literature. Oxford: Oxford University Press, 2005.
  • Chapman, S. B. “No Prophet Like Moses: Canonical Conclusions as Hermeneutical Clues.” In The Law and the Prophets: A Study in Old Testament Canon Formation, pp. 111–149. Tübingen: Mohr Siebeck, 2000.
  • Childs, Brevard S. “The Canon in Recent Biblical Studies: Reflections on an Era.” Pro Ecclesia 14 (2005): 26–45.
  • Childs, Brevard S. Introduction to the Old Testament as Scripture. Phildadelphia: Fortress, 1979.
  • Cross, Frank M. “The History of the Biblical Text in Light of the Discoveries of the Judean Desert.” Harvard Theological Review 57 (1964): 281–299.
  • Davies, Philip R. “The Jewish Scriptural Canon in Cultural Perspective.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 36–52. Peabody, Mass.: Hendrickson, 2002.
  • Evans, Craig A. “The Scriptures of Jesus and His Earliest Followers.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 185–195. Peabody, Mass.: Hendrickson, 2002.
  • Evans, Craig A., and Emanuel Tov, eds. Exploring the Origins of the Bible: Canon Formation in Historical, Literary, and Theological Perspective. Acadia Studies in Bible and Theology. Grand Rapids, Mich.: Baker Academic, 2008.
  • Evans, Craig A., and H. Daniel Zacharias, eds. Jewish and Christian Scripture as Artifact and Canon. Studies in Scripture and Early Judaism and Christianity 13. Library of Second Temple Studies 70. New York: T & T Clark, 2009.
  • Finley, Thomas J. “The Book of Daniel in the Canon of Scripture.” Bibliotheca Sacra 165 (2008): 195–208.
  • Graetz, Heinrich. “Der Abschluss des Kanons des Alten Testaments.” Monatsschrift fur Geschichte und Wissenschaft des Judentums 35 (1886): 281–298.
  • Leiman, S. Z. The Canonization of the Hebrew Scripture: The Talmudic and Midrashic Evidence. Hamden, Conn.: Archon, 1976.
  • Lewis, Jack P. “Jamnia Revisited.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 146–162. Peabody, Mass.: Hendrickson, 2002.
  • Lewis, Jack P. “What Do We Mean by Jabneh?” Journal of Bible and Religion 32 (1964): 125–132.
  • Mason, Steve. “Josephus and His Twenty-Two Book Canon.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 110–127. Peabody, Mass.: Hendrickson, 2002.
  • McDonald, Lee Martin. The Biblical Canon: Its Origin, Transmission, and Authority. Peabody, Mass.: Hendrickson, 2007.
  • McDonald, Lee Martin, and James A. Sanders, eds. The Canon Debate. Peabody, Mass.: Hendrickson, 2002.
  • Metzger, Bruce. The Canon of the New Testament: Its Origin, Development, and Significance. Oxford: Oxford University Press, 1987.
  • Sanders, James. “Adaptable for Life: The Nature and Function of Canon.” In Magnalia Dei: The Mighty Acts of God; Essays on the Bible and Archaeology in Memory of G. E. Wright, pp. 531–560. New York: Doubleday, 1976; reprinted in From Sacred Story to Sacred Text: Canon as Paradigm. Philadelphia: Fortress, 1987.
  • Sanders, James. From Sacred Story to Sacred Text: Canon as Paradigm. Philadelphia: Fortress, 1987.
  • Sanders, James. “The Issue of Closure in the Canonical Process.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 252–263. Peabody, Mass.: Hendrickson, 2002.
  • Sanders, James. Torah and Canon. Philadelphia: Fortress, 1972.
  • Schniedewind, William M. How the Bible Became a Book: The Textualization of Ancient Israel. Cambridge, U.K.: Cambridge University Press, 2004.
  • Seitz, Christopher R. “Old Testament or Hebrew Bible? Some Theological Considerations.” In Word Without End: The Old Testament as Abiding Theological Witness, pp. 61–74. Grand Rapids, Mich.: Eerdmans, 1998.
  • Stone, Timothy. “The Biblical Canon According to Lee McDonald: An Evaluation.” European Journal of Theology 18 (2009): 55–64.
  • Sundberg, Albert C., Jr. “The Old Testament of the Early Church: A Study in Canon.” Harvard Theological Review 51 (1958): 205–226.
  • Sundberg, Albert C., Jr. “The Septuagint: The Bible of Hellenistic Judaism.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 68–90. Peabody, Mass.: Hendrickson, 2002.
  • Toorn, Karel van der. Scribal Culture and the Making of the Hebrew Bible. Cambridge, Mass.: Harvard University Press, 2007.
  • Ulrich, Eugene. “The Notion and Definition of Canon.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 21–35. Peabody, Mass.: Hendrickson, 2002.
  • VanderKam, James C. “Questions of Canon Viewed through the Dead Sea Scrolls.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 91–109. Peabody, Mass.: Hendrickson, 2002.
  • Van Seters, John. “The Origins of the Hebrew Bible: Some New Answers to Some Old Questions.” Journal of Ancient Near Eastern Religions 7 (2007): 87–108.
  • Van Seters, John. “The Origins of the Hebrew Bible: Some New Answers to Some Old Questions, Part Two.” Journal of Ancient Near Eastern Religions 7 (2007): 219–237.
  • Wasserstein, Abraham, and David J. Wasserstein. The Legend of the Septuagint: From Classical Antiquity to Today. New York: Cambridge University Press, 2006.
  • Watts, James W., ed. Persia and Torah: The Theory of Imperial Authorization of the Pentateuch. Atlanta: Society of Biblical Literature, 2001.
  • Wegner, Paul. The Journey from Texts to Translations: The Origin and Development of the Bible. Grand Rapids, Mich.: Baker Academic, 1999.

Rebecca S. Hancock

Old Testament

Israel's scriptures were preserved, combined, and organized in order to provide a touchstone collection of religious writings for the ongoing life of the Jewish community. The traditional term for this collection of scriptures is “canon.” By setting aside some books as the written core of their tradition, Jews designated a historical standard of conviction and action that could be used to assess their present devotion and inspire future generations to greater faithfulness.

Rather than constituting a legalistic imposition upon creativity, the establishment of a biblical canon spurred Jewish religious imagination in the form of commentary, leading in time to the production of subsequent collections of rabbinic teaching (e.g., the Mishnah, the Talmud), all of which engaged, supplemented, and interpreted the tradition's written core. The earliest Christians, who were Jews, shared a fundamental allegiance to this same written deposit of religious truth. As increasing numbers of non-Jews (or gentiles) joined Christian communities, however, the status and sufficiency of Jewish scripture eventually came into question, with Christians ultimately producing an additional body of religious instruction (i.e., the New Testament) that likewise engaged, supplemented, and interpreted Judaism's core canon.

Religious History.

Little was known, even in antiquity, about the precise details of the historical process that led to the establishment of the canon. For the Jewish historian Josephus (37–100 C.E.) the collection of sacred Jewish books had already existed for “long ages,” with no one venturing “either to add, or to remove, or to alter a syllable.” It was instead “an instinct with every Jew, from the day of his birth, to regard them as the decrees of God, to abide by them, and, if need be, cheerfully to die for them” (Ag. Ap., 1:37–43). Josephus and others did relate various ideas about the authorship of the books within the canon, but the manner of the books' collection and organization went for the most part unexamined. Josephus himself attributed the faithful transmission of the biblical writings to an unbroken succession of prophets, from the time of Moses (thirteenth century B.C.E.?) to that of Artaxerxes (465–423 B.C.E.).

Jewish Tradition.

According to the rabbinic writing Baba Batra 14b–15a (ca. 100–150 C.E.?), Israel's scripture took a tripartite form in which the five-book Torah (or Pentateuch) was followed by an eight-book subset of Prophets (Joshua, Judges, Samuel, Kings, Jeremiah, Ezekiel, Isaiah, and the book of the Twelve [or Minor Prophets], in the order Hosea, Joel, Amos, Obadiah, Jonah, Micah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi), and an eleven-book subcollection of Writings (Ruth, Psalms, Job, Proverbs, Ecclesiastes, Song of Songs, Lamentations, Daniel, Esther, Ezra, Chronicles). Baba Batra also posits that these twenty-four books had been determined and reliably conserved by a Great Assembly of rabbinic scholars operating between the time of the prophets and the beginning of rabbinic Judaism (ʾAbot 1:1), and by the continuously meticulous work of scribes. The theory of a Great Assembly was given fresh expression in the sixteenth century by the Jewish scholars David Kimchi (1160–1235) and Elias Levita (1469–1544), and exercised a major influence on Christian writers in turn until Abraham Kuenen (1828–1891) demonstrated that rabbinic references to the Assembly were all derived from Nehemiah 8–10 and possessed no independent basis in fact.

The order in which the books appeared did not remain constant within Judaism over time, although the five-book Pentateuch always appeared first and almost always in the same sequence. Another, slightly different, sequence of the prophetic books later became standard, with the Major Prophets in the order Isaiah, Jeremiah, and Ezekiel. The contents of the Writings exhibited the greatest fluidity. Sometimes Chronicles appeared at the beginning of the subcollection (e.g., in the eleventh-century C.E. Leningrad Codex) and often the five smaller festival scrolls, known as the Megilloth, were grouped together in order of their usage within the liturgical year (i.e., Song of Songs, Ruth, Lamentations, Ecclesiastes, Esther). In later Jewish Bibles, the Megilloth can even appear as a separate subcollection, placed between the Pentateuch and the Prophets, thus yielding a fourfold organization to the Hebrew Bible. Although the threefold structure (i.e., Law-Prophets-Writings) predominated from early on, evidence also exists in antiquity of alternative formats (e.g., the order Law-Writings-Prophets is implied in Sirach 38:34B—39:1, a Jewish work dating to ca. 180 B.C.E.; no known order can be recognized in the apparent allusion to scripture in 2 Maccabees 2:13–15, dating from the early first century B.C.E.). Sometimes the overall count was given as twenty-two books instead, but this number was reached simply by including Ruth with Judges and Lamentations with Jeremiah. Twenty-two symbolized completion, since it was also the number of letters in the Hebrew alphabet.

Christian Tradition.

Christians were always aware that the first part of their Bible had come from the Jews. Already in the second century Bishop Melito of Sardis (d. 180 C.E.) tells in a letter how he journeyed “to the east… where these things are preached and done” in order to determine, in response to a request for information, “the accurate facts about the ancient writings, how many they are in number and what is their order” (Eusebius, Hist. eccl. 4.25.13–14). Melito then lists these books, which appear largely as in Jewish tradition—but possibly including the Wisdom of Solomon (if Proverbs is not meant), without Esther (and possibly Nehemiah if a separate book from Ezra), and with the usual positions of Leviticus and Numbers reversed. As this example also indicates, Christians knew the Old Testament scriptures primarily in Greek translation and in varying formats. Building on the tradition related in 4 Ezra 14:44–46, Irenaeus (d. 202 C.E.) and Tertullian (160–220 C.E.) express the view that Ezra had been the one to recopy the biblical books in the fifth century B.C.E. (i.e., after Jerusalem's destruction and the Babylonian exile), organize them, and authorize them for use within postexilic Jewish life. Yet further questions about the development of the Old Testament canon went mostly unasked.

Early Christian Bibles.

Of the three great fourth-century C.E. codices, Alexandrinus, Sinaiticus, and Vaticanus (the first large Christian Bibles containing the entirety of both the Old and New Testaments in Greek), two—Alexandrinus and Sinaiticus—exhibit orders similar to the prevailing tripartite structure of Jewish scripture, with the Writings at the end. However, various additional books also appear within their contents. Alexandrinus includes Tobit, Judith, 1–2 Esdras, and 1–4 Maccabees prior to the Psalms, and Wisdom and Sirach at the end of the Writings; Sinaiticus places Tobit, Judith, and 4 Maccabees prior to the Prophets, and Wisdom and Sirach at the end of the Writings but prior to Job, which appears last. One codex—Vaticanus—presents something like the quadripartite form that later became more or less standard for the Christian Old Testament, that is, with the Latter Prophets at the conclusion of the canon (although in Vaticanus the Minor Prophets precede the Major Prophets, and the Major Prophets are followed by Daniel, which then appears in final position rather than Malachi). Each of these three codices thus contains different books (i.e., they vary in their canonical scope) in differing arrangements (i.e., they vary in their canonical order).

Nevertheless, all three begin with the Pentateuch and apparently proceed directly to the Former Prophets (Sinaiticus is incomplete in this section), which they conclude with Chronicles (and then Esdras in Sinaiticus and Vaticanus). They all list the Major Prophets in the order Isaiah, Jeremiah, Ezekiel (although Ezekiel is missing in Sinaiticus); arrange the first half of the Minor Prophets differently (i.e., in the order Hosea, Amos, Micah, Joel, Obadiah, Jonah [Hosea, Amos and Micah are absent in Sinaiticus]); and begin the poetical books with the Psalms. These additional books are all of Jewish origin, and a few (e.g., Sirach) were treated for centuries as authoritative scripture by at least some streams in Jewish tradition (e.g., several rabbinic passages cite Sirach as an authoritative source). Eventually, however, these books lost their foothold within Judaism and survived almost exclusively in Christian Bibles. There they were distributed throughout the rest of the biblical books in a seemingly ad hoc fashion and never achieved the same acceptance or stability as the rest.

Catholic Tradition.

The later Latin translation of the Bible (i.e., the Vulgate), especially as codified by the Council of Trent (1546) and printed in the official Clementine edition (1592), standardized the scope and order of the Old Testament canon for Catholic Bibles. Within this familiar fourfold organization, the Pentateuch is followed by a fourteen-book grouping of Histories (Joshua, Judges, Ruth, 1–2 Samuel, 1–2 Kings, 1–2 Chronicles, Ezra [or 1 Esdras], Nehemiah [or 2 Esdras], Tobit, Judith, and Esther [with additions]), a seven-book unit of Poetry (Job, Psalms, Proverbs, Ecclesiastes, Song of Songs, Wisdom, and Sirach [or Ecclesiasticus]), and a twenty-book array of Prophets (Isaiah, Jeremiah, Lamentations, Baruch [including the Epistle of Jeremiah], Ezekiel, Daniel [with additions], Hosea, Joel, Amos, Obadiah, Jonah, Micah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi, and 1–2 Maccabees). Moreover, the Prayer of Manasseh, 3–4 Esdras (known as 1–2 Esdras when Ezra and Nehemiah do not already appear as 1–2 Esdras), and Psalm 151 appear as a four-work appendix, designated as extra-canonical or “apocryphal.” The result is the Catholic Old Testament canon of forty-six books, arranged in the order Pentateuch-Historical Books-Poetic Books-Prophets, with a supplementary four-book collection of Apocrypha. It should be noted that the books of 3–4 Maccabees were not included within the Vulgate, although they had appeared in many Greek Bibles for centuries, and that three books (Esther, Baruch, and Daniel) are presented in an expanded form at variance from the Hebrew biblical tradition.

Apocryphal/Deuterocanonical Books.

Today the Roman Catholic and Greek Orthodox traditions have agreed to designate as “deuterocanonical” (rather than “canonical” or “apocryphal”) the works known as Tobit, Judith, Additions to Esther, Wisdom, Sirach, Baruch, Epistle of Jeremiah, Additions to Daniel, and 1–2 Maccabees. In addition, Greek and Russian


Standardizing the Old Testament Canon.

Seventeenth-century engraving of the Council of Trent (1545–1563).


view larger image

Orthodox churches recognize as “deuterocanonical” 3 Esdras (although numbered as 1 Esdras), the Prayer of Manasseh, Psalm 151, and 3 Maccabees. (4 Maccabees is also then appended as an “apocryphal” book in the Greek canon.) The Slavonic church recognizes the book of 4 Esdras (numbered as 3 Esdras) as part of its canon. The Coptic Orthodox canon does not include the Prayer of Manasseh or 3–4 Esdras but does include 3 Maccabees. The Syrian Orthodox canon, in its East Syriac form, also contains 3 Maccabees but omits Tobit. The Armenian Apostolic Church treats 3 Maccabees and 3–4 Esdras as extra-canonical. The Ethiopic Bible exists in two forms. The narrower version is actually the printed Christian Bible with the widest scope, and includes all of these books plus Jubilees and Enoch—except for 4 Maccabees (and featuring significantly different versions of 1–3 Maccabees). An even broader Ethiopian canon exists in theory but is loosely defined and unpublished.

Protestant Tradition.

Protestantism has been more resistant to the books that did not survive within Judaism. It terms all of them “apocryphal,” drawing upon learned tradition within Catholicism. For example, Jerome (347–420 C.E.) coined the term “apocryphal” and did not include the deuterocanonical books in his original version of the Vulgate—however, when pressed he finally did produce translations of the Additions to Esther, the Additions to Daniel, Tobit, and Judith. Doubts about the full canonicity of the deuterocanonical books were also expressed by Julius Africanus, Athanasius, Cyril of Jerusalem, Rufinus, John of Damascus, Nicholas of Lyra, Ximenes, and Cajetan, among others. This scholarly position emphasized the character of the Greek Bible (or Septuagint) as a translation from the “Hebrew truth” (Latin: hebraica veritas). By the Renaissance, controversy had intensified over the apocryphal books because commentators found it increasingly problematic that the books' Hebrew versions were not extant. In his 1519 debate with John Eck, Martin Luther (1483–1546) rejected the canonical authority of 2 Maccabees because of its apparent support for the sale of indulgences and the doctrine of purgatory (see 2 Macc 12:39–45; 15:12–14). Andreas Karlstadt (1486–1541) broadened this exclusion, further classifying 1–2 Maccabees, Wisdom, Sirach, Judith, and Tobit as uncanonical yet holy, and 3–4 Esdras (numbered as 1–2 Esdras), Baruch, the Prayer of Manasseh, and the Additions to Daniel as wholly uncanonical. Karlstadt allowed that Christians might still read the first group with profit, but insisted that the books which were instead “canonical beyond all controversy” should be studied first and foremost. Without the apocryphal books, the narrower Protestant Old Testament contains the same twenty-four books as within Judaism but counted as thirty-nine and grouped by and large as within the Vulgate: the Pentateuch, followed by a twelve-book collection of Former Prophets (Joshua, Judges, Ruth, 1–2 Samuel, 1–2 Kings, 1–2 Chronicles, Ezra, Nehemiah, Esther), a five-book set of Poetic Books (Job, Psalms, Proverbs, Ecclesiastes, Song of Solomon), and a seventeen-book unit of Latter Prophets (Isaiah, Jeremiah, Lamentations, Ezekiel, Daniel, Hosea, Joel, Amos, Obadiah, Jonah, Micah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi).

Protestants and the Apocrypha.

Although Luther did not write commentaries or preach on the books of the Apocrypha (except for two sermons on Sirach), he continued to believe that they were “useful and good to read.” Early English Bibles, beginning with the Coverdale Bible (1535), were strongly influenced by Luther's Bible of 1534, which included a separate Apocrypha between the Old and New Testaments. Reformed groups, however, were even more critical of the apocryphal books and at least one version (1640) of the Geneva Bible was published in Amsterdam without them. According to the Westminster Confession (1646), “The Books commonly called Apocrypha, not being of Divine inspiration, are no part of the Canon of Scripture; and therefore are of no authority in the Church of God, nor to be any otherwise approved or made use of than other human writings” (Article 3). The highly influential Authorized Version (or King James Bible) of 1611 was originally published containing the Apocrypha and this practice continued routinely until in 1827 the British and Foreign Bible Society decided only to print Bibles without the Apocrypha. Opposition persisted in the later U.S. revision of the Authorized Version known as the American Standard Version (1901). Not until the Revised Standard Version (1952) did the Apocrypha return to a major U.S. translation of the Bible (added in 1957). Moreover, the Revised Standard Version eventually included 3–4 Maccabees and Psalm 151 in its Apocrypha. That broadened scope has been maintained in the New Revised Standard Version (1989), although this translation is currently available in two formats: one with the Apocrypha/Deuterocanonical Books, and one without. As a point of contrast, the New International Version (1978) continues on principle to omit the Apocrypha.


From this description it can be seen that the Christian Old Testament canon is singular in conception but manifold in format. Its precise boundaries have always been blurry, and slightly different understandings of its contents and internal structure have developed within the various streams of global Christianity. Most branches of Christendom have viewed the “deuterocanonical” or “apocryphal” books as suitable for “example of life and instruction of manners” (39 Articles of Religion, Church of England [1563], Art. 5), but some uncertainty has persisted regarding their doctrinal merit (even among Roman Catholics and the Orthodox). The precise scope and order of the Old Testament canon thus remain undetermined for Christians worldwide, and the shared Christian commitment to a single biblical canon depends more upon custom and usage than on any explicit ecumenical decision.

Critical Scholarship.

Early modern scholars quickly targeted the biblical canon as a symbol of ecclesial obscurantism and a barrier to free inquiry into the scriptures. The rise of historical-critical scholarship thus went hand-in-hand with an attack on the historical reliability of the traditional canon and its trustworthiness for the adjudication of theological questions. This attack on the received canon was combined with open distaste for the Old Testament's contents, which were perceived as antithetical to the values of modernity.

Johann Salomo Semler (1725–1791).

More than anyone else, the German Enlightenment scholar Semler began a new phase of canonical study by insisting that the Bible was not the unchanging “Word of God” but a product of human understanding and activity, historically conditioned over the course of centuries. He had gotten this basic notion from Richard Simon (1638–1712), a French Catholic whose ultimate goal was to demonstrate the decisive role of religious tradition within the formation of the Old Testament, and thus the inadequacy of the Protestant approach to the Bible. Rather than scripture preceding tradition, he maintained, tradition had historically preceded scripture—a finding that stood Reformation teaching on its head. However, Catholic officials responded to Simon's skepticism with regard to Moses' authorship of the Pentateuch by expelling him from his order and destroying copies of his Histoire critique du Vieux Testament (A Critical History of the Old Testament, 1678). Such opposition lessened, but did not eliminate, the impact of Simon's ideas. For his part, Semler criticized the view of inspiration held within Lutheran orthodoxy by calling attention to problematic features within the biblical literature itself. Scripture could not be directly and naively appropriated by the Christian believer, he argued, but needed to be sifted critically—especially the Old Testament, whose “coarse Jewish prejudices” had to be corrected by recourse to the New Testament. Historically, Semler detailed how there had been varieties of the canon in antiquity, even among the Jews (e.g., he pointed to the Samaritans and the Sadducees as having alternate canons). In other words, Semler insisted forcefully that the biblical canon had a history, and he worked toward charting that history's contours. However, the task of reconstructing the canon's full history fell to others.

Modern Scholarship.

By the beginning of the nineteenth century the “progressive formation” of the Old Testament had become a standard topic for scholarly investigation and was increasingly being explained in association with the tripartite division of the canon within Jewish tradition. More specifically, the Old Testament was envisioned as having formed in three stages corresponding to the tripartite literary arrangement of Law-Prophets-Writings. The rationale for this correspondence came partly from renewed attention to rabbinic sources, in which an awareness of the three subdivisions was a prominent feature. But the impulse was also partly chronological. In the highly regarded Old Testament introduction of Wilhelm Martin Leberecht de Wette (1780–1849), for example, the roots of the Pentateuch are traced to Moses (who, de Wette thought, likely wrote just a few laws and thus made only a “feeble commencement” of it), and the Pentateuch's completion is judged probably to have occurred during Josiah's reign (649–609 B.C.E.), with the oracles of the Prophets and the wisdom teachings of the Writings generally following after the Pentateuch in succession. Finally, the appeal of this impulse lay as well in its ability to do greater justice to certain peculiarities of the Hebrew canon's scope and order: “The reception of historical and of some prophetical writings into the Hagiographa [or Writings] can be explained only on the hypothesis that both the former collections were closed when this was begun” (Introduction, § 12b–13). The idea that the canon's subdivisions were closed in succession thus helped to explain why the books were ordered in ways that seemed at odds with their contents or genre.

Three-stage Theory.

By the end of the nineteenth century a scholarly consensus had formed: the Old Testament had developed in a three-stage process corresponding to the tripartite division of the Hebrew canon. First, the Pentateuch was completed and invested with authority in the time of Ezra (mid-fifth century B.C.E.). Echoes of this transformative event were thought to exist in Nehemiah 8–10, and in the presentation of Ezra as “a scribe skilled in the law of Moses” (Ezra 7:6). This initial canonization of the Pentateuch alone, as something like Israel's first Bible, helped to explain how the Samaritan canon could include the Pentateuch but not the Prophets. The Samaritans had separated from Judean Judaism after the Pentateuch had become canonical but before the Prophets were added to the canon. Moreover, inner-biblical allusions within the later writings of the Old Testament appeared to support the written, canonical status of the Pentateuch beginning in the Persian period (539–333 B.C.E.)—and to vitiate any similar claims for the Prophets or the Writings. Finally, the new approach to the formation of the Pentateuch and the history of Israelite religion, as presented above all by Julius Wellhausen (1844–1918), also dated the final pentateuchal source (i.e., the Priestly Source or “P”) to the Persian period, thus providing a tight weave of interlocking critical theories. It was further believed, however, that a crucial preliminary step toward the canonization of the Pentateuch had been taken during the preexilic reign of Josiah, when an early form of Deuteronomy (consisting more or less of Deut 12–26) was discovered during repair work on the Jerusalem Temple and made the basis of a wide-ranging religious reform (2 Kgs 22–23). For many scholars, Deuteronomy's self-referential rhetoric (phrases such as “all the words which are found in this book” occur throughout) and apparent religious impact indicated the origin of scripture within ancient Judaism. Yet Deuteronomy's legal rulings (= “D”) were sometimes countered by competing rulings of “P” in the final form of the Pentateuch, and both “D” and “P” had seemingly been harmonized in Chronicles and Ezra-Nehemiah. For this reason, the scripturalization of Deuteronomy looked to be both authoritative and provisional, with a truly delimited and absolute notion of “canon” only beginning to function when the entire Pentateuch was set apart as the “law of Moses” by Ezra.

The prophetic writings were thought to have existed in some form since the eighth century B.C.E. But critical scholarship suggested that they had experienced additions and editorial revision down through the early Hellenistic period (332–54 B.C.E.). So the Prophets were not added to the pentateuchal canon until they were literarily complete. These writings had always been controversial anyway; it took centuries for their religious merits to be fully appreciated. However, the destruction of Jerusalem, as the preexilic prophets had predicted, would also have lent substantial validation to the prophetic message. The book of Sirach provided an important external witness for the canonization of the Prophets. The body of the book was composed by Jesus Ben Sirach in Hebrew circa 180 B.C.E. Later, a prologue was added in Greek circa 130 B.C.E., in which Jesus' grandson tells how his grandfather came to write the book and why he then decided to translate his grandfather's work into Greek. The prologue specifically refers to “the law and the prophets and the other books of our ancestors,” suggesting that by this time a broader canon was in existence—indeed, that its contents were available in Greek as well as Hebrew (since the grandson explicitly draws a comparison between his own undertaking and the translation of Israel's scriptures). In the body of the book, a recitation of Israelite heroes (Sir 44–50) exhibits a striking familiarity with biblical narrative, even including a citation of Malachi 4:5–6 (Heb. 3:23–24), which is often considered one of the last additions to the prophetic corpus, in its description of Elijah (Sir 48:10). Sirach also refers to the twelve minor prophets as a group (Sir 49:10); they were transmitted as a single book in antiquity. Moreover, the Hebrew canon counts the book of Daniel, widely dated to approximately 165 B.C.E., among the Writings rather than the Prophets. This classification could be explained by supposing that the prophetic corpus had been “closed” before this time, thus excluding Daniel.

So the prophetic collection as such was often dated circa 200 B.C.E. With this addition, it was theorized, Israel's pentateuchal canon had become a canon of Law and Prophets. A similar process of addition, revision, gradual acceptance, and ultimate approval led to the development of yet another canonical subcollection, that of the Writings. Although largely complete by the turn of the millennium, the Writings had not been absolutely closed until the end of the first century C.E., creating a new (third) canon with three canonical subdivisions: Law, Prophets, and Writings. Rabbinic references to certain discussions in Jamnia (or Yavneh) about the status of some of the biblical books were invoked as evidence that a Jewish “council” had ultimately delimited the full tripartite canon circa 90 C.E. In sum, this model of Old Testament canon formation became the majority view in scholarship and was reproduced in one form or another in most of the biblical handbooks and introductions published throughout the twentieth century.

Criticism of the Three-stage Theory.

Beginning in the final quarter of the twentieth century, however, the three-stage view encountered increasing objections and criticism. It turns out that while not everything about it was wrong, even those things it had right were in need of significant revision and restatement.

In recent scholarship, the portrait of the Persian period has changed from one of a wooden religious orthodoxy to one of intense religious creativity and ferment. Correspondingly, the notion of an absolutely complete Pentateuch, enshrined under Ezra, carries much less conviction than it once did. Questions also continue to be raised about the Samaritans and their canon. Cross (1998, p. 208) has traced the distinctive Samaritan script to a Hasmonean examplar, which not only indicates a date for the origin of the Samaritan Pentateuch in the second century B.C.E. but also suggests that the absence of the Prophets in the Samaritan canon is due to their conscious rejection and not their early lack of canonicity. Moreover, the precise contents of the Samaritan canon prior to the turn of the millennium actually remain open to question (e.g., it seems that the earliest historical witness to a Pentateuch-only Samaritan canon is Origen, 185–254 C.E.). In fact, even the much later Samaritan Chronicles (whether they are a forgery or not) contain material from the biblical prophetic books, thereby ruling out any total opposition to the Prophets within Samaritan tradition. While exegetical work on the prophetic books continues to emphasize their gradual formation and relatively late completion, scholarship has also continued to uphold their fairly early origins and the real possibility of a nascent prophetic collection prior to the Babylonian exile (587–539 B.C.E.). For example, Gottwald (1985, p. 465) envisions a sixth-century prophetic corpus consisting of portions of Isaiah, Jeremiah, Hosea, Amos, Micah, Nahum, Habakkuk, and Zephaniah. Indeed, although Gottwald largely holds to the three-stage theory, he adopts the position that “despite the divisions between the two collections [of Law and Prophets], it is nonetheless evident that these two sets of traditions interacted intimately within the institutional life of Israel” (p. 458). Such a proposal responds implicitly to what had always been a problematic feature of the majority view: namely, that the books later canonized as the Prophets could have been remembered, transmitted, and cherished without possessing any canonical authority prior to the time of their conclusion. It was especially difficult to explain how an uncanonical assortment of prophetic writings could have gained canonical stature in precisely the same period in which Israel supposedly acknowledged a Pentateuch-only canon.

An Alternative Approach.

The idea of sustained interaction between the two traditions later represented by Law and Prophets is more appealing than the three-stage theory. Such interaction is bolstered by internal biblical evidence in which the prophetic traditions are increasingly understood corporately (e.g., 2 Kgs 17:13; cf. also the combination of the Minor Prophets within a single scroll), treated as authoritative precedents (e.g., Jer 26:16–18), and perceived as complementary to the legal tradition (e.g., Ezra 9:10; Dan 9:9–10), with the result that “the law of Moses” and “the words of [God's] servants the prophets” ultimately become twin criteria of right belief (e.g., Jer 26:4–6; Zech 7:12), a merism for Jewish faith. Josephus's theory of the canon's prophetic transmission is therefore not idiosyncratic but simply a historicized version of the unity between law and prophets already found in the canon. This unity is also expressed within the Bible by the idea of the prophets as biblical authors (see 1 Chr 29:29; 2 Chr 9:29; 12:15; 13:22; 26:22; 32:32; 33:19).

A similar dynamic was apparently at work in the transmission of the Writings, which cannot be restricted in their origins and literary development to the period after the theorized Law and Prophets canon was supposedly promulgated (ca. 200 B.C.E.). Some of the books in this subdivision are widely recognized to have undergone a lengthy process of collection, addition, and redaction (e.g., Psalms); their very preservation and even their adaptation are increasingly seen as markers of canonicity instead of non-canonicity. Even if some of these compositions were in fact authored during the Hellenistic period, it is difficult to imagine that their substantial canonicity was uncertain by the end of the first century B.C.E. To be sure, there may have been questions raised about particular books; even today the precise boundaries of the canon are fuzzy (see above).

Dead Sea Scrolls.

The ancient manuscripts discovered at Qumran (and in other locations near the Dead Sea) may not be perfectly representative of Hellenistic Judaism. Nevertheless, these documents exhibit a pronounced regard for scripture. Every biblical book later found in the Jewish canon has been discovered in some form at Qumran (except for Esther, a small book, and Nehemiah, if a separate book from Ezra at this time). Other books, however, have been found there as well, including books later known from Greek Bibles, and sectarian writings unique to the Qumran community. In this last group, frequent reference is made to the biblical books, which are known collectively as “[that which is] commanded by the hand of Moses and by the hand of all his servants the prophets” (1QS 1:2–3) or “the books of the law…the words of the prophets” (CD 7:15–17). On this basis, the Qumran finds testify to an emerging canon that did not, to be sure, possess absolute boundaries or a fixed text, but which nevertheless constituted—and this is no insignificant matter—a recognizable body of religious teaching with reliable contours and undisputed authority.

One reference to an emerging canon at Qumran is seemingly tripartite (or perhaps quadripartite, but this possibility is increasingly doubtful). The document known as the Halakic Letter (4QMMT or 4Q394–399) refers to “the book of Moses and the words of the prophets and David…the generations.” Here “David” may well stand for the Psalms or even for an early collection of the Writings. The document has been convincingly dated to ca. 150 B.C.E. If that date holds, then the document might provide the first external witness to something resembling the later tripartite canon. With such a witness, this letter would antecede Sirach's tripartite reference to “the law and the prophets and the other books of our ancestors” (ca. 130 B.C.E.) and the later tripartite allusion in the New Testament Gospel of Luke to “the law of Moses, the prophets, and the psalms” (Luke 24:44).

A First-century C.E. Canon.

A further indication of an expanded, perhaps tripartite, canon exists in the New Testament motif of a succession of earlier martyrs, “from the blood of Abel to the blood of Zechariah” (Matt 23:35; Luke 11:51). In this hendiadys, Abel is considered the first martyr because he is the first figure slain for the faith (Gen 4:8), and Zechariah is thought to be the last such martyr because he likewise appears at the conclusion of some sort of historical or canonical sequence. Since Matthew 23:35 specifies his identity as “Zechariah son of Barachiah, whom you murdered between the sanctuary and the altar,” and Luke 11:51 describes him as the one “who perished between the altar and the sanctuary,” this Zechariah is apparently the same one mentioned in 2 Chronicles 24:20–22. The resultant parallel between Abel and Zechariah could have been understood as purely historical, but it is highly unlikely anyone actually thought that the final Jewish martyr was killed during the reign of Joash (835–796 B.C.E.). More persuasive is the idea that Chronicles' canonical placement, either coming last in the historical books or at the end of the Writings, invited the sense of “from first to last,” on which the “from Abel to Zechariah” motif depended for its intelligibility and force. Yet there is finally no way to know if there was a customary order to the emerging Hebrew canon at this point or, if so, what it was.

Information about the biblical canon remains sketchy for the first century C.E. But both the New Testament and early rabbinic traditions give every impression of working with a stabilized biblical corpus—or at least treating the biblical corpus as stable even while it may have continued to undergo some change. Cross (1998, p. 217) dates the Pharisaic recension of the Hebrew biblical text, the direct ancestor to the later rabbinic Bible, to the early first century C.E. The well-known rabbi Hillel (d. 10 C.E.?), among others, engaged in textual debates and promulgated interpretive rules that can function only on the basis of a stable biblical text.

The Sadducees.

Moreover, there are no explicit debates about the canonicity of particular biblical books or teachings during this period, either in the rabbinic literature or in the New Testament. It is especially striking that the disputes between Jesus and the Pharisees and Sadducees never turn on which books are held to be canonical, or are more canonical than others. Some scholars believe that the Sadducees, like the Samaritans, may have possessed a Pentateuch-only canon, because both the New Testament book of Acts (23:8; cf. Matt 22:23 // Mark 12:18 // Luke 20:27) and Josephus (Ant. 18:16–17) report that the Sadducees did not subscribe to a belief in the resurrection of the dead. Such a belief is alleged to appear only in non-pentateuchal scripture, and so the inference is drawn that the Sadducees would only have acknowledged the Pentateuch as canonical. Yet this inference is faulty on five counts: first, rabbinic arguments for the resurrection of the dead were easily mounted on the basis of pentateuchal as well as non-pentateuchal scripture (e.g., Sanh, 90b; Pesaḥ. 68a; cf. Sipre 306); second, Acts 23:8 also characterizes the Sadducees as denying “angel or spirit,” but both figure prominently in pentateuchal scripture; third, in another passage Josephus explains the differences between the Pharisees and Sadducees by saying that the Pharisees “had passed on to the people certain regulations not recorded in the laws of Moses,” while the Sadducees “hold that only those regulations should be considered valid which were written down” (Ant. 13:297)—therefore the distinction between the two groups turned on the authority of oral tradition rather than the authority of non-pentateuchal scripture; fourth, prophetic scripture was well-known and a normal part of the synagogue service (Matt 2:4–6; Luke 4:16–20); and fifth, the gospel disputation scenes involving Jesus and the Sadducees cannot be satisfactorily interpreted according to the inference that they observed different canons.

For example, Jesus responds to a request for a sign from the Pharisees and Sadducees by referring to the “sign of Jonah” (Matt 16:4) without any indication that the Sadducees would have dismissed such an appeal outright. In Matthew 22, Jesus' scriptural proof for resurrection is the pentateuchal phrase “I am the God of Abraham, the God of Isaac, and the God of Jacob” (Exod 3:6); the point is the present tense verb used in reference to the long-dead ancestors, implying their continued existence. The argument of those who think the Sadducees only acknowledged the Pentateuch is that here Jesus accommodates his argument to their presuppositions. But the Sadducees' question actually assumes resurrection (Matt 22:28; perhaps tongue in cheek) and Jesus' answer confronts their presuppositions regardless of the possibility that they might be offended. Both Origen (Against Celsus, 1.49) and Jerome (Commentary on Matthew, 22:31) voice the view that the Sadducees' canon consisted of the Pentateuch alone, but they wrote much later and were likely dependent upon Josephus.

Alexandrian Canon Hypothesis.

It was also at one time broadly accepted that different biblical canons had developed in Palestine and Alexandria. As initially advanced by Francis Lee (1661–1719), the Palestinian canon was the narrower one that later became normative within Judaism. The Alexandrian canon was wider, and when it was translated into Greek it became the Septuagint, with a different order (i.e., quadripartite) and scope (i.e., including the deuterocanonical books). Especially as a result of Sundberg's work (1964), this theory has been abandoned, primarily because it is now evident that there was considerably more awareness of Greek culture in Palestine than once thought, and that Judaism and Hellenism intermingled thoroughly in the eastern Mediterranean. For example, Greek biblical texts were also discovered at Qumran. Moreover, some ancient Greek biblical texts give evidence of being revised in the direction of a Hebrew text (e.g., the Minor Prophets scroll from Wadi Murabbaʿat). This recensional activity not only indicates that the Greek and Hebrew biblical traditions interacted with each other and were linked in the editorial work they underwent; it also implies a certain privileging of the Hebrew text tradition, over the Greek, as primary and ultimately controlling.

“Council” of Jamnia Myth.

Yet recent scholarship has been most at odds with the three-stage theory in its rejection of a canonical “council” in Jamnia circa 90 C.E.. First suggested by Heinrich Graetz (1817–1891), the Jamnia hypothesis became a crucial linchpin for the older approach, which viewed it as not only providing the historical context for the official inclusion of the Writings within the Hebrew canon but also the date of the finalization of the Hebrew Bible as a whole (thus Ryle 1895, pp. 182–183). However, the references to any such gathering are meager; only a single rabbinic text relates that discussions took place, and then only about Song of Songs and Ecclesiastes (m. Yad. 3.5). But there are other rabbinic mentions of such discussions regarding various books even later. Moreover, the presenting question regarding Song of Songs and Ecclesiastes in m. Yadayim 3.5 is not whether they are to be made canonical, but whether their traditional scriptural standing is still to be affirmed. There appears to have been an academy or school rather than a council in Jamnia, but the idea that this academy fixed the Jewish canon in formal proceedings cannot be sustained. There is in fact no evidence that Judaism has ever made an official, binding decision on the scope and order of its biblical canon.


Biblical canon formation is increasingly being characterized as an organic process of gradual growth and subdivision, arising from religious practice, rather than a linear development of successive stages determined by official decision(s). The term “stabilization” describes this process better than “closure.” Several scholars (Beckwith 1985; Chapman 2000; Ellis 1991; Lightstone 1979; Steinmann 1999; Trebolle Barrera 1998) have rejected the three-stage theory of Old Testament canon formation in favor of a model in which the works now found in all three subdivisions were developing simultaneously and even mutually influencing each other. However, stalwart defenders of the three-stage theory persist (e.g., McDonald 2007) and debate continues.

Terms and Criteria.

As both of the preceding sections have illustrated, the biblical canon is characterized by unity and diversity. For this reason, care must be taken with the term “canon” so that inaccurate or misleading assumptions do not prejudice investigation unfairly in advance. For example, some scholars have advocated making an absolute terminological distinction between “canon” and “scripture” (e.g., Sundberg 1964: Swanson 1970). According to this view, a document's scriptural status has to do with its authority, but its canonical standing is instead a matter of its fixity. In other words, a document might be religiously authoritative and still undergo change; to speak of canonicity as the equivalent of religious authority, they argue, obscures the further nuance of delimitation that the term “canon” properly expresses. Such a distinction, however, is itself anachronistic.

“Canon 1” and “Canon 2.”

Similarly, McDonald employs the terminology of “canon 1” and “canon 2,” with “canon 1” referring to a religiously authoritative writing or collection of writings and “canon 2” referring to a delimited (or “closed”) writing or collection. For McDonald, “canon 2” is the fullest expression of canonical standing. However, the language of “canon 1” and “canon 2” has been appropriated from Sheppard (2005), who uses these terms differently. For Sheppard, authority (“canon 1”) and fixity (“canon 2”) are two poles encompassing a spectrum rather than successive phases. Sheppard acknowledges that usually the movement does appear to go from less fixity to more as scriptures and scriptural collections form. Yet he also cites the example of the scriptures of the Manicheans, which were produced by their founder Mani (216–276 C.E.) with a high degree of fixity from the outset. In this particular case, fixity seems to have preceded authority, or at least they functioned simultaneously. With respect to the biblical corpus, Sheppard's point is precisely to undercut the frequent assumption that canonization is only a late extrinsic application of religiously associative meaning upon discrete and unrelated texts. Instead, Sheppard makes the case that the gradual compiling of the biblical collection exercised an increasing impact on the literary form of the individual books themselves in the ongoing process of their editorial reworking and structural combination. For example, Sheppard's own treatment of the wisdom literature (1980) illustrates how the pentateuchal tradition of Mosaic law, in the process of becoming more literarily fixed, had a transformative effect on Israel's wisdom tradition, as evident from certain books in the Writings (e.g., Eccl 12:13–14). He also describes the influence of the wisdom tradition upon other traditions and genres in Old Testament scripture. In this way a canonical impulse is observable within the literary development of the biblical books and cannot be restricted to a final extrinsic decree.

Scripture, Bible, Canon.

The effort to divide cleanly between “scripture” and “canon” also does not do justice to the category of “scripture,” which involves many nuances rather than only one (i.e., “authority”) and can connote unity just as strongly as the term “canon.” It is certainly true that Israel did not possess a “Bible” in the same way that people think about the Bible today. (For one thing, there was no physical book-format like the codex until the first century C.E.; prior to that time, scrolls were used.) However, Israel's scriptures were not the equivalent of a “library” or an “anthology” either, since the individual writings within a library or anthology are not usually imagined to have influenced each other in their composition and history of transmission (although much recent literary criticism has in fact illustrated that all writings do and have had exactly this kind of influence on each other, to varying degrees). In this sense the term “canon” continues to be an illuminating word to use in describing the distinctive literary dynamic of Israel's biblical tradition, which characteristically exhibits a wide variety of internal cross-referencing, numerous examples of mutual influence, and thus a high degree of intertextuality.

To this extent the old debate about whether “canon” refers properly to a standard or a list misses what is crucial about the nature of the literature it seeks to describe: its internally associative character. Rather than a standard or a list, Sheppard characterizes the biblical canon as an “intertext.” Yet attempts to date the canon still often speak past each other. As Barton (1998) has amply illustrated, when canon is understood as a fixed list, the dates offered for its establishment tend to be late. So too when canon is taken to be an authoritative standard, the dates advanced for it are correspondingly early. This conflict arises because the biblical canon, to judge by both its history and its contemporary formats, functions by its very nature with a stable core and permeable boundaries. It also arises because what is being dated in either case is not arguably the most significant aspect of the canon. More helpful, because more precise, is the recent move by several scholars to stress the operation of a “core canon” by at least the second century B.C.E. This approach corrects reconstructions that grant too much literary fixity too early by envisioning a completely finished Bible in the second century or even earlier (Beckwith; Freedman 1992; Leiman 1976). It also corrects just as much those accounts that postpone the reality of the canon until well after the composition of the books in the New Testament (Gese 1974) or even after the fourth century C.E. (McDonald).

By the first century C.E. the biblical canon of Judaism certainly represented a known and unquestioned authority, a normative written expression of Jewish religious heritage. It was inherited by the Christian church as the faithful record of what God had done on Israel's behalf in preparation for the advent of Israel's messiah, whom the church confessed to recognize in Jesus Christ. The Christian confession of the messiahship of Jesus thus necessarily entails an affirmation of Israel's role within God's activity in the world and an endorsement, as well as a rereading, of the Old Testament canon.



Historical Sources

  • De Wette, Wilhelm Martin Leberecht. A Critical and Historical Introduction to the Canonical Scriptures of the Old Testament. Translated by Theodore Parker from the 5th German ed. (1840). 3d ed. 2 vols. Boston: Rufus Leighton, 1859; 1st German ed. 1817. Influential nineteenth-century German textbook on the Old Testament.
  • Eusebius. The Ecclesiastical History. Translated by Kirsopp Lake. 2 vols. Loeb Classical Library 153, 265. Cambridge, Mass.: Harvard University Press, 1926. The earliest history of the Christian church.
  • Graetz, Heinrich. “Der alttestamentliche Kanon und sein Abschluss.” In Kohelet, oder, Der salomonische Prediger, pp. 147–174, Leipzig: C. F. Winter, 1871.
  • Jerome. Commentary on Matthew. Translated by Thomas P. Scheck. The Fathers of the Church 117. Washington, D.C.: Catholic University of America Press, 2008. Provides evidence that the Samaritan canon consisted of the Pentateuch only.
  • Josephus. Against Apion. Translated by H. St. J. Thackeray. Loeb Classical Library 186. Cambridge, Mass.: Harvard University Press, 1926. The most important first century C.E. witness for a stable and authoritative Jewish canon.
  • Josephus. Jewish Antiquities, Books XII–XIII. Translated by Ralph Marcus. Loeb Classical Library 365. Cambridge, Mass.: Harvard University Press, 1943. Suggests the Sadducees were opposed to the oral teachings of the Pharisees rather than prophetic scripture.
  • Josephus. Jewish Antiquities, Books XVIII–XIX. Translated by Louis H. Feldman. Loeb Classical Library 433. Cambridge, Mass.: Harvard University Press, 1965. Characterizes early debates between Pharisees and Sadducees.
  • Kuenen, Abraham. Over de mannen der Groote Synagogue. Amsterdam, n. p.: 1876. Demonstrates that the notion of a Great Assembly is a myth.
  • Origen. Against Celsus. Translated by H. Chadwick. Cambridge, U.K.: Cambridge University Press, 1965. Provides evidence that the Samaritan canon consisted of the Pentateuch only.
  • Ryle, H. E. The Canon of the Old Testament. 2d ed. London: Macmillan, 1895. The classic English-language presentation of the three-stage theory of Old Testament canon formation.
  • Semler, Johann Salomo. Abhandlung von freier Untersuchung des Canon. Texte zur Kirchen-und Theologiegeschichte 5. Gütersloh: Mohn, 1967; 1st German ed. 1771–1775. Early modern work focusing on the history of the biblical canon's development.
  • Simon, Richard. Histoire critique du Vieux Testament. Rotterdam: Chez R. Leers, 1685. Early historical-critical appraisal of the Old Testament.
  • Westcott, Brooke Foss. The Bible in the Church. 2d ed. London: Macmillan, 1913. A history of the Bible's role and function throughout church history.

Modern Literature

  • Barton, John. Holy Writings, Sacred Text: The Canon in Early Christianity. Louisville, Ky.: Westminster John Knox, 1998. An investigation into New Testament canon formation, emphasizing both its early stability and its blurred boundaries.
  • Beckwith, Roger T. The Old Testament Canon of the New Testament Church and its Background in Early Judaism. London: SPCK; Grand Rapids, Mich.: Eerdmans, 1985. Major treatment that dates the conclusion of the Old Testament canon to the Maccabean period.
  • Chapman, Stephen B. The Law and the Prophets: A Study in Old Testament Canon Formation. Forschungen zum Alten Testament 27. Tübingen: Mohr Siebeck, 2000. A comprehensive critique of the three-stage theory and a reconstruction of a more organic process of Old Testament canon formation.
  • Cross, Frank Moore. From Epic to Canon: History and Literature in Ancient Israel. Baltimore: Johns Hopkins University Press, 1998. An exploration of canon formation centering on text-critical issues and dating the canon to the first century C.E.
  • Elliott, J. K. “Manuscripts, the Codex and the Canon,” Journal for the Study of the New Testament 63 S (1996): 105–123. Detailed exposition concerning the scope and order of Old Testament books in early Christian manuscripts and codices.
  • Ellis, E. Earle. The Old Testament in Early Christianity: Canon and Interpretation in the Light of Modern Research. Wissenschaftliche Untersuchungen zum Neuen Testament 54. Tübingen: Mohr Siebeck, 1991. An exploration of the character and function of Old Testament scripture in the early church; rejects the three-stage theory of canon formation.
  • Freedman, David Noel. “The Symmetry of the Hebrew Bible.” Studia theologica 46, no. 2 (1992): 83–108. An argument for a high degree of editorial balance among the three subdivisions of the Hebrew canon, and for very early canonization.
  • Gese, Hartmut. Vom Sinai zum Zion: Alttestamentliche Beiträge zur biblischen Theologie. Munich: Christian Kaiser, 1974. An argument for the reception of an open Old Testament canon within early Christianity.
  • Gottwald, Norman K. The Hebrew Bible: A Socio-Literary Introduction. Philadelphia: Fortress, 1985. A standard textbook suggesting the likelihood of a preexilic prophetic corpus.
  • Hayes, John H. “Historical Criticism of the Old Testament Canon.” In Hebrew Bible, Old Testament: The History of Its Interpretation; Volume 2: From the Renaissance to the Enlightenment, edited by Magne Sæbø, pp. 985–1005. Göttingen: Vandenhoeck & Ruprecht, 2008. A learned discussion of canonical scholarship in the eighteenth century, particularly that of Semler.
  • Leiman, S. Z. The Canonization of Hebrew Scripture. Hamden, Conn.: Archon Books, 1976. Like Beckwith, Leiman argues for a conclusion to canon formation in the Maccabean period, but primarily on the basis of rabbinic sources.
  • Lewis, Jack P. “Jamnia Revisited.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 146–162. Peabody, Mass.: Hendrickson, 2002. A recent summary of scholarship on the Jamnia question and a reaffirmation of the author's skeptical position.
  • Lewis, Jack P. “What Do We Mean by Jabneh?” Journal of Bible and Religion 32, no. 2 (1964): 125–132. The initial attack on the Jamnia hypothesis.
  • Lightstone, Jack N. “The Formation of the Biblical Canon in Judaism of Late Antiquity: Prolegomenon to a General Assessment.” Studies in Religion/Sciences religieuses 8, no. 2 (1979): 135–142. An argument against the three-stage theory, based on rabbinic sources.
  • McDonald, Lee Martin. The Biblical Canon: Its Origin, Transmission, and Authority. 3d ed. Peabody, Mass.: Hendrickson, 2007. The standard work in English, it upholds the three-stage theory, works with an overly narrow understanding of “canon” and therefore adopts very late dates.
  • Rogerson, J. W. An Introduction to the Bible. London: Penguin, 1999. A popular introduction in which the modern plurality of Bibles and canons is highlighted.
  • Rüger, H. P. “The Extent of the Old Testament Canon.” Bible Translator 40, no. 3 (1989): 301–308. Contains detailed information about the contents of the canon within contemporary global Christianity.
  • Sheppard, Gerald T. “Canon.” In Encyclopedia of Religion, edited by Lindsay Jones, vol. 3, pp. 1405–1411. 2d ed. Detroit: Macmillan, 2005. Treats “canon” as a socioliterary phenomenon holding a place along a continuum between authority and fixity.
  • Sheppard, Gerald T. Wisdom as a Hermeneutical Construct: A Study in the Sapientializing of the Old Testament. Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft 151. Berlin: de Gruyter, 1980. An examination of the relationship between canon formation and the wisdom tradition.
  • Steinmann, Andrew E. The Oracles of God: The Old Testament Canon. St. Louis: Concordia Academic Press, 1999. An investigation of Old Testament canon formation rejecting the three-stage theory.
  • Sundberg, A. C., Jr. The Old Testament of the Early Church. Harvard Theological Studies 20. Cambridge, Mass.: Harvard University Press, 1964. An argument against the Alexandrian canon hypothesis; also distinguishes sharply between “scripture” and “canon.”
  • Swanson, T. N. The Closing of the Collection of Holy Scripture. PhD diss., Vanderbilt University 1970. Advocates a thorough-going distinction between “scripture” and “canon.”
  • Trebolle Barrera, Julio. The Jewish Bible and the Christian Bible: An Introduction to the History of the Bible. Leiden, Netherlands: Brill; Grand Rapids, Mich.: Eerdmans, 1998. Comprehensive overview of canon formation rejecting the three-stage theory.

Stephen B. Chapman

New Testament

Discussion of the canon of the New Testament has been revitalized in recent scholarly research because of several factors. These include a renewed interest in extracanonical documents, reevaluation of the historical processes surrounding the development of early Christianity, and reassessment of many of the sources that are important in canonical discussion.

Definition of Canon.

The term “canon” is used with reference to the New Testament as that fixed and definable group of books that is recognized as being uniquely authoritative within the Christian church. The word “canon” is derived from the Greek word kanōn, which was a reed used for measuring and hence indicates the standard by which something is judged. The term canon was used in early Christianity in such phrases as “the canon [or rule] of faith” to indicate the standards for Christian belief or orthodoxy, but was not used in a technical sense of a body of writings until the fourth century, when Athanasius used it of a list or group of authoritative Christian writings (Thirty-ninth Festal Letter). In earlier discussions, the terms canon or canonical were sometimes used simply to describe the status of writings that were considered to have a measure of authority. The term canon today, however, is used in a more restricted sense to describe the result of a lengthy process of development and evaluation, and is distinguished here from the notion of scripture. Scripture indicates individual writings that had authority within early Christian communities, some but by no means all of which later came to be acknowledged as canonical or part of the canon of uniquely authoritative writings. The term “New Testament” (or new covenant; cf. Luke 22:20; 1 Cor 11:25) was first used in distinction to the Old Testament/Covenant in the late second century, probably by Marcion, and then later of the body of writings that were recognized as canonical.

In the most thorough study of the canon to date, Theodor Zahn (1888–1892) concluded that the core of the New Testament was established by the late first and early second century, as indicated by its liturgical use. He was opposed by Adolf Harnack (1925), who argued that the canon was not formed until the late second century. His view has tended to prevail until recently, when scholars have disputed the date for New Testament canonical formation, depending upon degrees of fixity of the various books and thoughts regarding authorship. Many think that the canon was formed in its major parts sometime in the second century, while others would place the date of canonical formation later, up to and including the fourth century or possibly beyond.

Historical Development and Transmission of Collections of Early Christian Writings.

The process of canonical formation extends from the time of the writing of the original documents until the New Testament canon was recognized as uniquely authoritative for the church, whether that occurred in the second, third, or fourth/fifth centuries. Virtually all scholars agree that what came to be recognized as the canon of the New Testament emerged over the course of time and in several different groupings, including the Gospels, the Pauline letters, and the remaining writings.


The Gospels (and Acts for the sake of discussion here, although Acts has a varied history of its own) form one of the major divisions in the formation of the New Testament canon. During the second and third centuries, there were many extracanonical Christian works written, which many scholars see as vying for authoritative status with the canonical documents (see below for discussion of some of the most important of these extracanonical works). As a result, most scholars take a gradual developmental view of the establishment of this part of the canon of the New Testament. However, there are signposts along the way that give insight into this process of development and the status of the various documents.

The Gospels reflect a complex process of inscripturation, beginning with probably the earliest Gospel, Mark's, purportedly written sometime around 65 C.E. Luke's and Matthew's Gospels, probably using Mark's Gospel and other sources, were written later, possibly as late as around 85 C.E., followed by John's Gospel in around 90 C.E. On the basis of statements in Eusebius citing Papias (Hist. eccl. 3.39.4, 15–16), scholars often draw attention to the emphasis upon oral tradition in the early church. However, the Gospels were written relatively early in relation to the events they purport to represent—whatever the status and influence of oral tradition.

The next landmark in development is the long ending of Mark (16:9–20), which though not original appears to be fairly widely known by the mid- to late second century (see Justin Martyr, 1 Apol. 45; Tatian's Diatessaron in the Persian form, and Irenaeus, Against Heresies 3.10.6), indicating that the longer ending was written in the first half of the century. The significance of this is that the ending appears to be a pastiche of the other gospels and Acts, thus indicating knowledge of and reliance upon the four gospels and Acts at this early stage.

At around the same time, Marcion was forming his corpus of writings. In around the middle of the second century, according to statements written later by Tertullian (Against Marcion 4.2, 5), Marcion chose from among the apostles, which included the four we know today, to include only one unnamed gospel, what appears to be Luke's. If these authors are correct, then Marcion had a choice of at least the four canonical gospels.

Tatian's Diatessaron, though no longer extant in Greek, was a harmony of the Gospels created by Tatian, a disciple of Justin Martyr, in around 150–172 C.E. Some contend that other gospels are cited in the Diatessaron, but the overwhelming majority of the text clearly comes from the four canonical gospels, and indicates their increasing authority and use as a body of texts. An early parchment manuscript (0212) was once thought to be part of Tatian's Diatessaron, but is now considered to be simply its own harmony of the Gospels (parts of the four gospels are contained in this fragmentary manuscript) dated to the latter second century. There were also other harmonies written about this time that relied upon the four gospels.

Three New Testament Greek papyri have been thought to indicate the process of gathering the canonical gospels into a single document. Upon their discoveries, the dates assigned to P4 (fragment of Luke), P64 (fragment of Matthew), and P67 (fragment of Matthew) varied greatly. However, more recent study has suggested that they may have belonged to the same manuscript from the late second century. T. C. Skeat suggests that they belonged to the first single-quire codex. Differences in the manuscripts and codicological discrepancies have called some of these conclusions into question, although most of the objections can be answered. If they do belong together—whatever the shape of the codex—they may well attest to an early gathering of the four gospels. Similarly, Skeat thinks that P75, a manuscript that contains portions of Luke and John (dated to 175–225 C.E.), is part of a four-gospel codex.

By the time of Irenaeus, writing around 180 C.E., we have clear reference to the four gospels as forming a distinct collection. Irenaeus refers to sets of four in the natural world (zones, winds, etc.) and hence concludes that it is appropriate that the gospel be transmitted in four aspects or faces, that is, the four gospels (Against Heresies 3.11.8–9).

The Muratorian Canon is probably a late second-century document (see below for discussion) that offers a list of New Testament documents. If this document does date to the second century, it attests to there being four gospels (the document is fragmentary and begins by mentioning the third gospel, Luke), as well as the book of Acts. The canon also mentions other canonical and extracanonical works, discussed below.

The papyrus manuscript P45 is the earliest extant document of the four gospels and Acts as a collection in apparently a single manuscript. This fragmentary manuscript, which probably dates to the early third century, contains portions of Matthew, Luke, John, Mark, and Acts, possibly arranged in that order, and apparently nothing more.

Other evidence could be cited, but this shows the progression of the emergence of an authoritative four-gospel collection, along with Acts. According to this developmental hypothesis, the four gospels, though written fairly soon after the events they report, existed alongside other traditions, both written and oral, into the second century, at which time the fourfold gospel of the New Testament took definitive shape. This fourfold gospel was then explicitly endorsed by the end of the second century and found in clear manuscript form in the early third century. Its recognition as a corpus, as shown below, continued into the fourth century and beyond.

Paul's Letters.

The forming of the Pauline letter collection is thought by most scholars to have taken less time than for the Gospels and Acts. There are five different views of the emergence of the Pauline letter collection worth considering here.

The gradual development theory, which is the predominant view (and similar to the process for the Gospels and Acts), argues either (according to Zahn) that the ten-letter Pauline collection (without the Pastoral Epistles, which were written later) was gathered together before the writing of 1 Clement (so before 85 C.E.) or (according to Harnack 1926) that the thirteen-letter Pauline collection was gathered around 100 C.E. For both views, Paul's letters were of high interest to the early church, and served an important liturgical purpose.

By contrast, Edgar J. Goodspeed (1927), followed by John Knox (1959), think that Paul's letters had fallen into neglect after his death until there was revived interest caused by the writing of the book of Acts, which prompted a collection of Paul's letters. Christians in Ephesus (possibly Onesimus, according to Knox) gathered the letters from surrounding cities and circulated the collection of ten letters with Ephesians at the head of the corpus.

Many scholars have recognized the role of a significant follower of Paul in the gathering of the Pauline letter collection. Two of the most plausible choices are Luke (Moule 1982) and Timothy (Guthrie 1970), both followers of Paul associated with him in the letters. Others have thought that a Pauline school of followers may have been responsible for gathering the Pauline letter collection.

The idea of personal involvement in the gathering of the Pauline letter collection has been extended by David Trobisch (1994). He argues that Paul, as a major letter writer, would have kept copies of his letters, and this formed the basis of his gathering his own letter collection. The organization of the canonical lists and various collections of Paul's letters according to decreasing length, with the exception that Galatians is shorter than Ephesians, indicates that Paul himself was responsible for collecting at least the four major letters of his collection—Romans, 1 and 2 Corinthians, and Galatians—before it was later expanded by inclusion of deuteropauline letters (e.g., Ephesians to the end of the collection) and then circulated in comprehensive and inclusive editions (e.g., the major codices).

The regular shape of the standard Pauline letter collections, especially as found in the papyrus manuscript P46, the earliest collection of Paul's letters, dated to around 200 C.E., with Ephesians preceding Galatians, has suggested to a few scholars that Paul may have been involved in the collection of his entire letter corpus. P46 is arranged in descending order according to both church and personal letters (hence Ephesians precedes Galatians), indicating continuity throughout the Pauline corpus.

The process for collection of the Pauline letter corpus involves several considerations. The first is the role that Marcion's canon may have played. Some think that the New Testament canon, and in particular the Pauline letter collection, was a response to Marcion's canon. However, most scholars think that Marcion was not the first to form a New Testament collection, and so his is probably a reaction to earlier collections. The question is whether Marcion knew of the Pastoral Epistles. Tertullian says that Marcion rejected them (Against Marcion 5.21), although some scholars think he simply did not know them (possibly because they were not yet written). A second consideration is the authorship of the Pastoral Epistles. As noted above, several reconstructions of the formation of the Pauline letter collection place the authorship of the (pseudonymous) Pastorals, as well as other letters, in the second century. A third consideration is the Muratorian Canon, which lists the thirteen letters of Paul in what is represented as chronological order for the ecclesial letters. If this document is early, it provides a clear terminus for the thirteen-letter Pauline collection. A fourth consideration is manuscript P46. Most scholars agree that P46 was written about 200 C.E. The manuscript includes all of the Pauline letters from Romans to 1 Thessalonians (as well as Hebrews after Romans and before 1 Corinthians). Most scholars think that the manuscript also included 2 Thessalonians and Philemon, but it is debated whether it included the Pastoral Epistles. In any case, a fragment of Titus (P32) has been dated to around 200 C.E., and so puts an end-date on the process.

The Pauline letter collection appears to have developed earlier than the fourfold gospel, perhaps as early as Paul himself, but at the latest probably by the time of Marcion, with a thirteen-letter Pauline collection (perhaps plus Hebrews; see below) by around 200 C.E.

Remaining Writings.

The remaining writings—the most diverse of the three divisions—were the last group to be gathered. The book of Hebrews has already been mentioned (see below), because of its appearance in the Pauline letter collection (P46) by around 200 C.E. The evidence is that 1 John and 1 Peter were widely recognized by the second century (according to Eusebius citing Papias, Hist. eccl. 3.39.17). There is, however, relatively little early church or transmissional evidence for the gathering of the remaining writings of the New Testament. Nevertheless, there are two manuscripts that merit consideration. A parchment manuscript (0232) that has been variously dated from the mid- third century to the fifth or sixth century (a mediating date of 300 C.E. has been proposed) has 2 John 1–2 on both sides of the page, with page numbers 164 and 165. This indicates that 2 John was part of a larger collection, probably with the Gospel of John, Revelation, and 1 John, possibly by the third century. The second manuscript is a third-century set of papyri manuscripts with 1 and 2 Peter and Jude (P72) that were found with other extracanonical manuscripts. Wasserman (2004) has proposed that these biblical documents were written in the third century for private use, indicating that at this date these books were collected together and had authoritative status.

Important Canonical Lists.

The above section offers some of the signposts along the way in the development of the various major divisions of the New Testament. As these scriptural groupings began to emerge, various authors and documents chronicled the status of these various collections. As a result, canonical lists are important for charting the progressive development of the New Testament canon.

Many scholars consider the Muratorian Canon the earliest list. The Muratorian Canon, though fragmentary, appears to list the four gospels (it begins with the third gospel, Luke, and then mentions the fourth, John), Acts, the thirteen letters of Paul, some noncanonical Pauline writings that are called forgeries, Jude and two Johannine epistles, the Wisdom of Solomon, the Apocalypse of John and the Apocalypse of Peter, Shepherd of Hermas (considered nonauthoritative), and the nonauthoritative works of several other authors. Excluded from the Muratorian Canon are (besides Matthew and Mark), Hebrews, James, and 1 and 2 Peter. The early date of the Muratorian Canon has been disputed on the grounds of its being such a list, its similarities to fourth-century lists, the nature of its Latin, and some of its internal references; for these reasons, some scholars argue for a mid-fourth-century date. Despite these objections, most scholars still find the second-century date most plausible, especially because of reference to the Shepherd of Hermas being written recently and in the author's own times. If the Muratorian Canon is early, it attests to a relatively stable group of New Testament writings, especially the Gospels and Acts and the Pauline writings, and most of the remaining books. The fact that Hebrews is not mentioned can be accounted for on the basis of the Muratorian Canon coming from the Latin church, where Hebrews was only slowly accepted as authoritative (see below). Other inclusions and exclusions are more difficult to account for, apart from acknowledging that the edges of what would become the New Testament canon were far less well established than its core.

Most of the lists of New Testament collections date to roughly the fourth century. Eusebius, writing around 320–330 C.E. from Palestine (representing the Eastern church), is a very important source of information on various collections of New Testament writings. Besides listing the New Testament books known to Irenaeus (Hist. eccl. 5.8.2–8), Clement of Alexandria (Hist. eccl. 6.14.1–7), and Origen (Hist. eccl. 6.25.3–14) (some scholars treat these as canonical lists), after discussing various books he gives a summary of the books of the New Testament (Hist. eccl. 3.25.1–7). Most scholars think that Eusebius gives a list divided into four types of books, although Kalin (2002) thinks that it is a list of three types. The first category includes the recognized books (Gospels, Acts, Paul's epistles probably including Hebrews, 1 John, 1 Peter, and Revelation), the second the doubtful books (James, Jude, 2 Peter, and 2 and 3 John), the third the rejected works (Acts of Paul, Shepherd of Hermas, Apocalypse of Peter, Barnabas, Didache, Revelation, and the Gospel of the Hebrews), and the fourth heretical books. The only anomaly is how Revelation can be both recognized and rejected; perhaps it was accepted by some churches and rejected by others in the early centuries, especially in the East. Eusebius, however, considers it authentic. It has been observed that Eusebius's recognized and doubtful books (the first two categories) comprise the twenty-seven books of the New Testament.

Codex Claromontanus (D), a Western bilingual manuscript (Latin and Greek) from the sixth century, has a list of New Testament books inserted in it. The date of the list is disputed, but it may be as early as around 300 C.E. The list includes the four gospels, ten Pauline letters (excluding Philippians and 1 and 2 Thessalonians), the Catholic Epistles (except Hebrews), Revelation, Barnabas, Shepherd of Hermas, Acts of Paul, and Apocalypse of Peter. However, the last four books have a short line in front of them, thought by some scholars to indicate that the writer of the list considered these disputed writings. The list is not a carefully prepared one, with 1 and 2 Peter said to be “to” Peter (1 Peter also has a short line in front of it), so it is unclear what to make of the inclusiveness of this list.

Cyril of Jerusalem in his Catechetical Lectures (4.36), written around 350 C.E., mentions the four gospels, Acts, seven Catholic Epistles, and fourteen Pauline epistles. He does not mention Revelation. He also distinguishes a Gospel of Thomas by the Manichaeans as pseudepigraphical (not the Coptic Gospel of Thomas discussed below).

Athanasius's Thirty-ninth Festal Letter, written around 367 C.E. from Egypt and representing the Eastern church—a letter written to announce the date of Easter for that year—mentions the four gospels and Acts, the Catholic Epistles, the Pauline letters including Hebrews listed after the ecclesial letters and before the personal letters, and Revelation. He then lists books that have edificatory value, but that are not canonical (e.g., Wisdom of Solomon, Sirach, Esther, Judith, Tobit, Didache, and Shepherd of Hermas). This list conforms to the canonical New Testament.

A list from northern Africa (also known as the Mommsen Canon, identified by Theodor Mommsen from a manuscript kept in Cheltenham, U.K.) dates to around 360–370 C.E. and represents the Western church. It mentions the four Gospels, the thirteen-letter Pauline corpus, Acts, Revelation and the Johannine and Petrine epistles, but not Hebrews or Jude.

Epiphanius writing from western Syria in around 374–377 C.E. in his refutation of heretics lists the four gospels, Paul's fourteen letters (presumably including Hebrews), Acts, the Catholic Epistles, Revelation, and two Wisdom books, one of Solomon and the other of Sirach (Panarion 76.5).

The Apostolic Constitutions compiled in the late fourth century (around 380 C.E.) include several canons, one of which (85) contains a list of New Testament books. These include four gospels, fourteen Pauline letters (presumably including Hebrews), two Petrine letters, three Johannine letters, James, Jude, two letters of Clement, the Apostolic Constitutions themselves, and Acts. Revelation is not included.

Gregory of Nazianzus writing in around 390 C.E. from Asia Minor (the Eastern church) offers a metrical list that includes the four gospels, Acts, fourteen Pauline letters (presumably including Hebrews), and seven Catholic Epistles. Revelation is not included.

Amphilochius of Iconium writing sometime after 394 C.E. from Asia Minor (the Eastern church), also writing metrically, includes four gospels, Acts, fourteen Pauline letters including Hebrews, seven Catholic Epistles, and Revelation—although he admits that some think Hebrews, some of the Catholic Epistles (one of the Petrines, two of the Johannines, and Jude), and Revelation are spurious. The list from the Council of Carthage (397 C.E.; which represents the earlier Council of Hippo [394 C.E.] as well) lists four gospels, Acts, thirteen Pauline letters, Hebrews, the Petrine and Johannine letters, James, Jude, and Revelation. This conforms to the canonical New Testament.

The Latin Western theologian Rufinus writing in the late fourth century (On the Creed 37–38) includes four gospels, Acts, fourteen letters of Paul, two of Peter, one of James, one of Jude, three of John, and Revelation, in other words the canonical New Testament. He also notes books that are useful for the church but not canonical (Shepherd of Hermas, the Two Ways, and the Judgment of Peter).

Jerome, writing in 394 C.E., lists the New Testament books as including the four gospels; seven letters to churches by Paul (with Hebrews as falling outside this group); letters to Timothy, Titus, and Philemon; Acts; the seven letters of James, Peter, John, and Jude; and Revelation (Epistle 53.9). Elsewhere Jerome endorses Hebrews (and Revelation), even if authorship is unknown (Epistle 129.3).

Augustine in his Christian Doctrine (2.13), writing in the late fourth century, lists the four gospels; fourteen letters of Paul; the seven letters of Peter; John, Jude, and James; Acts; and Revelation. This comprises the canonical New Testament.

Pope Innocent writing in 405 C.E. (in a letter reconstructed from later copies) lists four gospels, fourteen Pauline letters (although some of the earliest manuscripts only list thirteen), the seven Catholic Epistles (Jude is excluded in one manuscript), Acts, and Revelation.

A Syrian catalogue of books in the New Testament dated to around 400 C.E. lists four gospels, Acts, fourteen Pauline letters, confirming the list in the Doctrine of Addai from the last half of the fourth century. The Catholic Epistles and Revelation are not included, although the Peshitta from the early fifth century includes James, 1 Peter, and 1 John, and the rest of the Catholic Epistles and Revelation are included in the Philoxenian Syriac version of 508 C.E.

There are several relatively early complete New Testament manuscripts that provide lists of New Testament works by virtue of their contents or explicit lists. These include Codex Sinaiticus (fourth century), Codex Vaticanus (fourth century), and Codex Alexandrinus (fifth century). Vaticanus is incomplete, breaking off in Hebrews (which is placed after the Pauline ecclesial letters), but it is legitimate to conclude that it originally included the four personal Pauline letters and Revelation (the Catholic Epistles occur after the Gospels and Acts and before the Pauline letters). Sinaiticus is complete, with an order that is more like contemporary New Testaments, except for Hebrews being after 2 Thessalonians, and Acts after the Pauline letters. Sinaiticus also includes Barnabas and Shepherd of Hermas after Revelation. Alexandrinus follows Vaticanus, but includes 1 Clement and 2 Clement, and lists Psalms of Solomon following the Clementine letters, though after a notation that is not clear. There is dispute over the implications of the noncanonical books in Sinaiticus and Alexandrinus. Most scholars think that the edges of the canon were still not fixed, while others think that these extracanonical books may not have had the same authoritative status, even though included in these codices.

There are also dubious lists (e.g., Laodicea, Pope Damasus, John Chrysostom), as well as other later lists of New Testament collections.

On the basis of these lists, one can see that, by the end of the fourth century, the collection of authoritative writings that we would call the canonical New Testament is relatively firmly established. Although there are some small variations concerning Hebrews, especially among Western writers, and Revelation, especially among Eastern authors, as well as a few of the other Catholic Epistles, for the most part the twenty-seven-book canon is established. There are also a few extracanonical books that are still sometimes attached to this group of letters, including Barnabas, the Shepherd of Hermas, the Clementine letters, the Wisdom of Solomon, and occasionally others, although many of these are labeled as questionable or noncanonical by various writers as well.

Disputed Texts.

The second and third centuries were very prolific times for the production of Christian literature. During this period, there were numerous gospels, acts, letters, and apocalypses produced within the church, besides those that were canonized. Eusebius uses four (or three; see above) categories to describe the New Testament writings: recognized books, disputed books, inauthentic books, and heretical books (Hist. eccl. 3.25.1–7). The recognized books, as we have seen, were generally recognized fairly early, and include the Gospels and Acts, the Pauline letters, and 1 John and 1 Peter. The formation of the New Testament generally began with the formation of the Pauline letter collection, and was soon followed by the development of the fourfold gospel and Acts. The later lists of New Testament books generally confirm this.

The disputed books, according to the lists above, include the books that Eusebius mentions as disputed, as well as Revelation, which Eusebius acknowledges as problematic for some by placing it within the recognized books while also noting that some consider it inauthentic, and Hebrews, which he does not mention though it presumably is included within the Pauline letters. All of these books would be placed in the third grouping of books, the remaining writings, which were apparently the last division of the New Testament books to be collected. This process apparently took some time, as is indicated by how infrequently most of these books are cited by early church authors. Besides Philemon (one of the Pauline letters), the letters of 2 Peter, 2 and 3 John, and Jude were apparently not known to the Apostolic Fathers. The book of James was not known in the Western church until the second half of the fourth century, when it is cited by Hilary of Poitiers. Jerome and Augustine were instrumental in its acceptance. Second Peter is first cited by Ambrosiaster in the mid- to late fourth century, and then frequently by Jerome and Augustine, as well as others. Second John is cited from the third century on, but 3 John is not cited until Jerome and Augustine. Jude is known relatively early but then is disputed by many. Even though it is known from the mid-fourth century on, Jude still is not included in a sixth-century New Testament list.

There are several books that Eusebius categorizes as inauthentic but that were apparently recognized as having authoritative and even canonical status for some church groups and in particular locations. However, these books were not eventually included in the New Testament canon. The Acts of Paul is a second- or third-century work that also includes the story of the Acts of Paul and Thecla and 3 Corinthians. This derivative work was widely used, as attested by the numerous manuscripts of it, and 3 Corinthians was recognized in the Syrian and Armenian churches, and the object of commentary by Ephraem in the fourth century.

The Shepherd of Hermas is treated as scripture by several early church writers, especially in the late second century. It is, however, rejected in the Muratorian Canon because of its having been written recently (as opposed to being early and apostolic), but is still included at the end of the New Testament in Codex Sinaiticus.

The Apocalypse of Peter is paired with the Apocalypse of John in the Muratorian Canon, although some doubt is raised about it, and it is preceded by a short line, perhaps indicating its dubious status, in the Codex Claromontanus list. Despite this, the Apocalypse of Peter appears to have been used by the church at least for instructional purposes into the fifth century.

Barnabas seems to have had early acceptance especially in Egypt. It is cited by Clement of Alexandria and called by Origen a “catholic epistle” (Against Celsus 1.63), a term also used of 1 Peter and 1 John. Barnabas was valued by Jerome (even if he did not grant it canonical status) and is also found in Codex Sinaiticus.

The Didache may have been written as early as the mid-first century up to the early years of the second century, and therefore may have been written earlier than some of the books in the canonical New Testament. The Didache was known by Clement of Alexandria and Origen, and continued to be widely used in such places as Egypt and Syria (it was one of the sources of the Apostolic Constitutions), even though it was rejected by Eusebius and Athanasius.

First Clement was written early (around 96 C.E.) and had widespread circulation in the East. It was used by such Christian authors as Irenaeus, Clement of Alexandria, and Origen. First Clement is linked with 2 Clement in Codex Alexandrinus and in an eleventh-century manuscript, although for the most part it apparently circulated independently. First Clement was not recognized in the West.

In more recent times, there has been widespread discussion of the fragmentary Greek apocryphal gospel texts. Without necessarily endorsing their authoritative status, some scholars think that some of these documents once vied for canonical status. P.Egerton 2, a fragmentary gospel dated to the second half of the second century, with possible composition in the first half, is thought by a few scholars to represent a viable gospel tradition that preceded that of the canonical gospels. However, the heavy dependence of P.Egerton 2 upon the canonical gospels indicates its later origins. The Gospel of Thomas, found at Nag Hammadi in its Coptic version from the fourth century and at Oxyrhynchus in three fragmentary Greek manuscripts from the late second or early third century, has been argued to constitute a “fifth gospel” that, while not recognized as authoritative in earlier periods, should be recognized as providing authentic sayings of Jesus. However, the Gospel of Thomas probably dates to the late second century (whether or not it is the Gospel of Thomas mentioned by Hippolytus, Refutation 5.7.20).

Rationale for Inclusion and Exclusion of New Testament Books.

Discussions of the criteria for inclusion and exclusion of books in the canon of the New Testament differentiate between how these criteria functioned within the early church and how they might be viewed today. Four criteria are most commonly discussed.

The first is conformity to the “rule of faith.” This criterion indicates that books that were given scriptural status and eventually included in the canon of the New Testament were seen to conform to what had been taught by the apostles concerning the death and resurrection of Jesus Christ. In other words, books were judged regarding their doctrinal conformity and orthodoxy. Some contemporary scholarship has problematized the notion of orthodoxy by contending that the early church was theologically diverse and that what we would today call orthodoxy was simply one of several competing theological positions. However, this notion has also rightly been criticized because of evidence from the late first century forward (e.g., 1 Clement 72) that those within the early church understood that there was a determinative rule of faith. A good example of this process in practice is Eusebius's account of Bishop Serapion (Hist. eccl. 6.12.4). Serapion was at first willing to allow a Gospel of Peter, though knowingly pseudepigraphal (and this entered into his decision), to be read in the church at Rhossus until he learned that it had docetic tendencies, at which point he forbade its being used.

The second criterion is apostolicity. Some scholars would put the criterion of apostolicity—which is closely related to the first criterion—as the most important criterion in the early church. Tertullian distinguishes between apostles and “apostolic men” in his discussion of authorship of the Gospels, and questions Marcion's acceptance of Luke's Gospel because it was written by an “apostolic man” rather than an apostle, as were Matthew's and John's Gospels (Against Marcion 4.2.2). The book of Hebrews, we noted above, was often included within the Pauline corpus in early lists. The book of Hebrews, as anonymous, has no direct apostolic attribution, and had questionable status especially in the Western church. The book may have been placed within the Pauline letter collection as a way of establishing its apostolic authorship, at least until it was recognized as authoritative on other criteria. The Muratorian Canon makes clear that the Shepherd of Hermas should be rejected as authoritative because it is not apostolic but was written recently. Recent critical discussion of authorship of books of the New Testament—whether Gospels such as Matthew and John as apostolic or certain Pauline letters—has made this criterion problematic, and raises questions about the standards by which certain books may have been accepted as canonical.

The third criterion is widespread and continuous usage in the church, or what is sometimes referred to as “catholicity.” Although some scholars differentiate between widespread use and catholicity, this criterion asserts that those works that had a claim to authoritative status and hence to canonical inclusion were those that had not only local appeal or use within the early church, but widespread and longstanding appeal to the catholic church, that is, the church as it existed worldwide in both the East and the West. Hebrews and Revelation are cases in point. Hebrews was not universally accepted in the West but was accepted in the East. The West eventually recognized Hebrews, even though there was not clear apostolic authorship, in part because the Eastern church had long recognized it. Similarly, Revelation along with Hebrews was accepted by Jerome because of their previous widespread and longstanding use. We know, as the New Testament lists above illustrate, that other books were at one time considered authoritative but were not finally accepted into the New Testament canon. These books, though used by some churches and in some areas, apparently failed to have widespread and continuous usage within the catholic church.

The fourth criterion is inspiration. The criterion of inspiration has been understood in different ways through the centuries, and is arguably the most difficult of the criteria to define and understand. Difficulties revolve around ancient views of inspiration, and the question of the place where inspiration resides. Inspiration was a more widespread notion in the ancient world, with a variety of works said to be inspired, including writings that were not accepted into the canon (e.g., 1 Clement 63.2 regarding his own work; 2 Clement 11.2 regarding 1 Clement). There is also the question of whether inspiration is a feature of the author or of the work produced by the author. A traditional view of inspiration and canon is that canonization came about as a result of inspiration. In other words, the writings were the product of inspiration and this resulted in their canonization. More recent discussion has inverted this so that inspiration is the result of canonization. Thus, those works that came to be recognized as canonical on the basis of other criteria were then recognized as inspired. In this way, canonization and inspiration go hand in hand.

Later Discussions of Canon and “Canon within the Canon.”

The canon in the East remained somewhat flexible, due in part to contradictory decisions made at the Trullan Synod in the seventh century, so that differing lists of the books of the Bible were produced even into the tenth century. These differences can be seen in examining the canons of the various Eastern churches, such as the Syrian, Armenian, Georgian, Coptic, and Ethiopian, where as many as four more books are included than in the Western canon.

The canon in the West was relatively better defined, with the church using the New Testament of Jerome in Latin until the Middle Ages. However, there were anomalies, such as the non-Pauline Epistle to the Laodiceans appearing in numerous Vulgate and other manuscripts, including those in English and German. The Reformation—not surprisingly because of its emphasis upon the centrality of the Bible—brought a reexamination of the canon of the New Testament. As in earlier times, the books that were subject to the most canonical scrutiny were those in the third grouping, including Hebrews and Revelation. Andreas Karlstadt divided the New Testament into classes of letters much like Eusebius's categories. Martin Luther separated Hebrews, James, Jude, and Revelation from the rest of the New Testament writings (as did Tyndale), with a later Lutheran version of the Bible even labeling these works “Apocrypha.” The Council of Trent in 1546 confirmed the twenty-seven books of the New Testament as we currently have it.

There is still some debate about whether the canon should be open or closed. When such questions are not strictly academic, they face such difficulties as assessing the criteria by which inclusion and exclusion would be decided (the ancient criteria would preclude adding any new books because of failure to be in widespread and longstanding use, if not other criteria), and the consequences of such actions. A more pertinent discussion, however, revolves around the question of a “canon within the canon.” There is a sense in which a canon within the canon is almost inevitable, as one functionally must select certain books, ideas, or themes to represent the message of the entire canon. One cannot speak of the canon in terms of the entire canon at all times, but one is compelled to find a shorthand—a reduced canon within the larger canon—as surrogate for the entire canon. There is a more important formulation of the question, however, that focuses upon those who intentionally wish to identify and promote a canon within the canon. For example, there are those who desire to locate a central theme or idea that they think is the essential core of the New Testament, and that by definition excludes parts of the New Testament that do not reflect this core (this reflects earlier Lutheran thought in some ways). Or there are those who find the New Testament a far too diverse, or even self-contradictory, body of documents that requires that one find a palatable unifying idea or concept. Such canons within the canon may serve heuristic purposes, but they often appear to be one-sided characterizations that end up not commanding continued or widespread assent because they neglect as much as they accept. (Canonical criticism has helped interpreters to realize that all parts of the canon have a role to play in the canonical witness.) Attempts to find a canon within the canon, for whatever purpose, usually fail to appreciate the fullness of the canonical witness, and what all of the various parts of the canon contribute to the whole, by balancing and supplementing and completing other portions that can only be disproportionately emphasized at the expense of others.



  • Bruce, F. F. The Canon of Scripture. Glasgow: Chapter House, 1988. Detailed study of major issues in formation of both Old and New Testament canons. Tends toward earlier formation date.
  • Campenhausen, Hans von. The Formation of the Christian Bible. Translated by J. A. Baker. Philadelphia: Fortress, 1972. English translation of Die Entstehung der christlichen Bibel, first published in 1968. Historical and theological treatment of canon development.
  • Funk, Robert W., et al. The Five Gospels: The Search for the Authentic Words of Jesus. New York: Macmillan, 1993. Advocacy of the Gospel of Thomas as providing authentic Jesus sayings.
  • Gamble, Harry Y. The New Testament Canon: Its Making and Meaning. Philadelphia: Fortress, 1985. Concise treatment of major issues in New Testament canon formation.
  • Goodspeed, Edgar J. New Solutions of New Testament Problems. Chicago: University of Chicago Press, 1927.
  • Gregory, Caspar René. Canon and Text of the New Testament. International Theological Library. Edinburgh: T. & T. Clark, 1907. Historical treatment of canon formation.
  • Guthrie, Donald. New Testament Introduction. 3d ed. Downers Grove, Ill.: InterVarsity, 1970.
  • Hahnemann, Geoffrey Mark. The Muratorian Fragment and the Development of the Canon. Oxford Theological Monographs. Oxford: Clarendon, 1992. Most thorough discussion of a late date for the Muratorian Fragment, with implications for canon development.
  • Harnack, Adolf. Die Briefsammlung des Apostels Paulus und die anderen vorkonstantinischen christlichen Briefsammlungen. Leipzig: Hinrichs, 1926.
  • Harnack, Adolf. The Origin of the New Testament and the Most Important Consequences of the New Creation. Translated by J. R. Wilkinson. 2d ed. London: Williams & Norgate, 1925.
  • Kalin, Everett R. “The New Testament Canon of Eusebius.” In The Canon Debate, edited by Lee Martin McDonald and James A. Sanders, pp. 386–404. Peabody, Mass.: Hendrickson, 2002. Argues for tripartite division of books in Eusebius.
  • Käsemann, Ernst, ed. Das Neue Testament als Kanon. Berlin: Evangelische Verlagsanstalt, 1970. A collection of previously published essays by German scholars, followed by critical analysis.
  • Knox, John. Philemon among the Letters of Paul. Rev. ed. London: Collins, 1959.
  • Köstenberger, Andreas J., and Michael J. Kruger. The Heresy of Orthodoxy: How Contemporary Culture's Fascination with Diversity Has Reshaped Our Understanding of Early Christianity. Wheaton, Ill.: Crossway, 2010. Discussion of controversy about early orthodoxy and heresy.
  • McDonald, Lee Martin. The Biblical Canon: Its Origin, Transmission, and Authority. Peabody, Mass.: Hendrickson, 2007. Most recent major study of canon development. Tends toward later formation date.
  • McDonald, Lee Martin, and James A. Sanders, eds. The Canon Debate. Peabody, Mass.: Hendrickson, 2002. Major essays on most issues in contemporary canon discussion.
  • Metzger, Bruce M. The Canon of the New Testament: Its Origin, Development, and Significance. Oxford: Clarendon, 1987. Standard treatment of canon of New Testament. Tends toward earlier formation date.
  • Moule, C. F. D. The Birth of the New Testament. 3d ed. San Francisco: Harper & Row, 1982.
  • Parker, D. C. The Living Text of the Gospels. Cambridge, U.K.: Cambridge University Press, 1997. A study of the development of the Gospel as manuscripts.
  • Porter, Stanley E. How Did We Get Our New Testament? Text, Transmission, Translation. Acadia Studies in Bible and Theology. Grand Rapids, Mich.: Baker, forthcoming. A study of the major issues surrounding textual criticism, textual transmission, and translation of the New Testament.
  • Porter, Stanley E. “Paul and the Process of Canonization.” In Exploring the Origins of the Bible: Canon Formation in Historical, Literary and Theological Perspective, edited by Craig A. Evans and Emanuel Tov, pp. 173–202. Acadia Studies in Bible and Theology. Grand Rapids, Mich.: Baker, 2008. Treatment of the six major views of formation of the Pauline letter collection.
  • Skeat, T. C. “The Oldest Manuscript of the Four Gospels?” New Testament Studies 43, no. 3 (1997): 1–34. Defense of P4, P64, and P67 as from a single quire codex.
  • Skeat, T. C. “The Origins of the Christian Codex.” Zeitschrift für Papyrologie und Epigraphik 102 (1994): 263–268. Defense of P75 as part of a four-gospel codex.
  • Souter, Alexander. The Text and Canon of the New Testament. Revised by C. S. C. Williams. 2 ed. Studies in Theology 25. London: Duckworth, 1954. Older but still valuable discussion of canon.
  • Trobisch, David. Paul's Letter Collection: Tracing the Origins. Minneapolis: Fortress, 1994. Advocate of Pauline involvement in formation of the Pauline letter collection.
  • Wall, Robert W., and Eugene E. Lemcio. The New Testament as Canon: A Reader in Canonical Criticism. Journal for the Study of the New Testament: Supplement Series 76. Sheffield, U.K.: Sheffield Academic, 1992. Essays in canonical criticism.
  • Wasserman, Tommy. The Epistle of Jude: Its Text and Transmission. Coniectanea Biblica: New Testament Series 43. Stockholm: Almqvist & Wicksell, 2006. Discussion of P72.
  • Westcott, B. F. A General Survey of the History of the Canon of the New Testament. 6th ed. Cambridge, U.K.: Macmillan, 1889. Classic treatment of the canon formation of the New Testament.
  • Zahn, Theodor. Geschichte des Neutestamentlichen Kanons. 2 vols. Erlangen and Leipzig: Deichert, 1888–1892. The most comprehensive study of the New Testament canon.

Stanley E. Porter