This entry discusses how various writings and collections of writings were officially accepted by various religious authorities and communities as scripture; it consists of three articles:

The first article deals with the arrangement of books in the sacred text of Judaism, and the second article describes the processes by which those (and other) books were accepted as canonical by both jewish and Christian communities, in the latter case as the Old Testament. The last article deals with the canonization of the New Testament.

Order of Books in the Hebrew Bible

No traditions have survived concerning the authorities who fixed the canon of Hebrew scriptures, or about the internal order of the books, or about the underlying principles that determined their sequence. It is probable, but not certain, that the three distinct collections known as the Pentateuch, the Prophets, and the Writings (Hebr. tôrâ [Torah, nĕbîʾm, and kĕtûbîm, respectively, the initial letters of which form the acronym tanak, used in Jewish tradition for the Bible) represent the three successive stages of canonization. Apart from the Pentateuch, the order of whose five books is invariable, the arrangement of the contents of the other two corpora was not uniform in manuscripts and printed editions until fairly recently.

An anonymous tannaitic tradition (Bab. Bat. 14b), no later than ca. 200 CE, lists the order of the books of the Prophets and the Writings. This presents a problem because the codex form was not adopted by Jews before the fifth century CE and because the general and favored scribal practice—with one exception—was to restrict each scroll to a single biblical book. What then is the meaning of term “order” in the rabbinic text? The most likely explanation is that it refers to the manner of storage and the system of classification and cataloguing in vogue in the libraries and schools of Palestine. The library procedures of the Hellenistic world would have required each of the three collections of canonical works to be placed in a separate armarium, with the scrolls arranged in their appropriately assigned order.

The sequence of the Former Prophets following the Pentateuch is: Joshua, Judges, Samuel, Kings. This arrangement never varies and presents one long continuous history of Israel from the beginning of the conquest to the fall of the Judean kingdom, the Babylonian exile, and the release of King Jehoiachin from prison in 561 BCE. (See Deuteronomic History.)

The variations in the order of the books occur in the Latter Prophets and particularly in the Writings. A majority of manuscripts and most printed Bibles feature Isaiah, Jeremiah, and Ezekiel, which is the proper historical order. The above‐cited source, however, followed by some manuscripts, lists Isaiah in third place in juxtaposition with its contemporary Hosea. Another tradition has Jeremiah after Kings and before Isaiah and Ezekiel. This is because that prophet was active during the last years of the monarchy, and Jeremiah 39 and 52 largely duplicate 2 Kings 25.

The small prophetic books, generally known as the “Minor Prophets,” were habitually transcribed onto a single scroll and were collectively designated “The Twelve” (so already in Sir. 49.10, ca. 180 BCE). Their internal arrangement is: Hosea, Joel, Amos, Obadiah, Jonah, Micah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi. This is also the order of a scroll of the second century CE from Wadi Murabbaʿat (See Dead Sea Scrolls) containing the Hebrew Minor Prophets, and it apparently reflected traditional views about their historical sequence. The same order, but with Micah following Amos and succeeded by Joel, is given in 2 Esdras 1.39–40. This groups together three prophets of the eighth century BCE.

The order of the Writings in Hebrew printed Bibles is: Psalms, Proverbs, Job, Song of Solomon, Ruth, Lamentations, Ecclesiastes, Esther, Daniel, Ezra, Nehemiah, Chronicles. Passages like 2 Maccabees 2.13–14 and Luke 24.44 seem to attest to the great antiquity of the initial place of Psalms. The aforementioned tannaitic source has Ruth before Psalms due to the concluding genealogy of David, the reputed author of the Psalter. The Aleppo Codex (end of ninth century CE and the Leningrad Codex of 1008 CE both open the Writings with Chronicles, probably because that work duplicates the Pentateuchal genealogies and much of the Former Prophets.

The tannaitic practice, also found in manuscripts and ultimately standardized in the printed editions, was to conclude the Hebrew scriptures with Chronicles following Ezra‐Nehemiah. This must have been a very early tradition, for it is reflected in Matthew 23.35 and Luke 11.51. The inversion of the chronological order must have arisen out of a desire to close the canon on a note of consolation, and to make the statement that the fulfillment of biblical prophecy involves the return of the Jewish people to its ancestral land. Apart from this messianic exegesis, it also serves to encase the Hebrew scriptures within a framework of historical narrative, for Chronicles begins with Adam and its last sentence contains the same two key Hebrew verbs of redemption with which Genesis concludes (pqd, ʿlh, Gen. 50.24–25; 2 Chron. 36.23).

Christian editions reverse the order of Prophets‐Writings, so that the closing words of Malachi (4.5–6 [Heb. 3.23–24]) concerning Elijah become transitional to the New Testament, and connect with the role of John the Baptist (see Matt. 11.13–14; Mark 1.2; 9.11–13; Luke 1.16–17).

Least stable in respect of order are the small books in the corpus of the Writings. The tannaitic source follows Proverbs with Ecclesiastes and the Song of Solomon because all three are attributed to King Solomon. Most medieval manuscripts preserve this association in one way or another. Lamentations, Daniel, and Esther are grouped together since they all belong to the period of the exile. In medieval times, the Song of Solomon, Ruth, Lamentations, Ecclesiastes, and Esther were all clustered together in that order, based upon their use as lectionaries in the cycle of the Jewish religious calendar, commencing with Passover. This system became the rule in the printed editions. Greek Bibles differ considerably from the Hebrew scriptures in that the books are arranged according to genres of literature. Ignoring the additional Apocrypha that are interspersed among the canonical works, the following classification emerges. First comes a narrative‐historical collection that comprises the Pentateuch and Former Prophets, with Ruth attached to Judges, and Chronicles following Kings. Second is a prophetic collection consisting of: Isaiah; Jeremiah, to which is adjoined Lamentations for thematic reasons and traditions of authorship; Ezekiel; Daniel, because he is regarded as a prophet, a contemporary of Ezekiel, and is identified with the personality of that name mentioned in Ezekiel 14.14, 20; 28.3; and the Twelve in a slightly different internal order. The two complete Greek codices, the fourth‐century CE Vaticanus and the fifth‐century CE Alexandrinus, share these characteristics. However, the latter has Esther and Ezra‐Nehemiah immediately after the prophetical collection, while the former places Ezra‐Nehemiah after Chronicles. The third part is a poetic‐didactic collection. Codex Vaticanus has Psalms, Proverbs, Ecclesiastes, Song of Solomon, Job, and Esther. The order of Codex Alexandrinus is Psalms, Job, Proverbs, Ecclesiastes, and Song of Solomon. Also the sequence of the second and third collections interchanges in the two codices.

Other Greek codices feature different arrangements and there is no uniformity in the traditions of the churches. All extant Greek codices and lists are of Christian origin, and it is uncertain whether or not any represent alternative Jewish conventions about the order of the biblical books.

Nahum M. Sarna

Canon of the Hebrew Bible and the Old Testament

From the fourth century CE, the word “canon” (from a Greek word meaning “a rule”) has been used to denote the correct list of the biblical books, and in consequence the collection of books thus listed. It is important to distinguish between the composition of the biblical books and their recognition as scripture. Scholars sometimes envisage a five‐stage process of composition, circulation, revision, collection, and recognition as canonical, and such a long process may indeed have been involved for some books. Others of the books may have been much more rapidly acknowledged. But if we may judge from the history of the New Testament canon (see the next article in this entry), some books were probably recognized as having divine authority more slowly than others. It was at one time widely believed that the whole canon had been recognized by the time of Ezra, and this idea is already reflected in 2 Esdras 14.44–48, where Ezra is said to have “made public” twenty‐four books (the standard Jewish count of the canonical books). The idea that Ezra knew the canon is probably not entirely without foundation, for most of it had then been written, and an older tradition speaks of Ezra's contemporary Nehemiah, after the calamity of the exile, gathering together in a library “the books about the kings and prophets, and the writings of David, and letters of kings about votive offerings” (2 Macc. 2.13). In addition, Ezra would have known the books of the Pentateuch, which must also have been gathered together by this time, so most scholars agree, and which were the basis of his reforming work. A few of the latest books, however, had not yet been written, and they at least would obviously have had to be added to the sacred collection later. Moreover, there is reason to believe that Esther and Daniel were not finally accepted into the canon until the crisis of the second century BCE, related in the following verse of 2 Maccabees, where we are told that “in the same way Judas also collected all the books that had been lost on account of the war that had come upon us” (2 Macc. 2.14). This would have taken place about 164 BCE. The Judas in question is Judas Maccabeus, and the war is the persecuting campaign of the Hellenistic Syrian king Antiochus Epiphanes, who attempted to destroy the scriptures (1 Macc. 1.56).

Many factors contributed to the recognition of certain books as canonical; among them are the following: the tradition that many of the books came from Moses or one of the other acknowledged prophets; the spiritual authority of the books themselves, as it was experienced in public or private reading, and in exposition; the fact that the books had come to be laid up in the Temple as sacred; the opinions of religious leaders and the common convictions of the people about the books. And for Christians, there was the additional consideration that Jesus himself and his apostles, in the pages of the New Testament, often refer to the Jewish scriptures in general, and to many of the individual books, as having the authority of God.

The division of the Hebrew Bible into three sections (not four or five, as in Greek, Latin, and English translations), known as the Law, the Prophets, and the Writings, can be traced back to the second century BCE, when it is three times referred to in the prologue of Sirach, added by the Greek translator of the book in about 130 BCE. He refers to the three sections as “the Law and the Prophets and the others that have followed them,” “the Law and the Prophets and the other books of our ancestors,” “the Law and the Prophecies and the rest of the books.” It should be noted that the third section had not yet been given a definite name, though the use of the definite article and the expression “the rest of” suggests that it already had a fixed content. In other early references to the three sections (in Luke 24.44, the Dead Sea Scrolls, and Philo), the third section seems to be called “the Psalms,” taking its name from one of the chief books it included. The later Jewish name, the Writings (Hebr. kĕtûbîm, translated into Greek as hagiographa), is first found in rabbinic literature.

For the past hundred years it had been commonly believed that the three sections do not really have any distinct identity, but are accidents of history, reflecting the different stages at which books were accepted as canonical: the Law in the fifth century BCE, the Prophets in the third century BCE, and the Hagiographa at the “synod” of Jabneh or Jamnia, about 90 CE (an academic debate that discussed, among other things, the canonicity of Ecclesiastes and the Song of Solomon). The Law, however, clearly does have a distinct identity, consisting of the four books that comprise the life of Moses and the legislation bearing his name (Exodus to Deuteronomy), together with a historical introduction (Genesis) tracing the course of events between creation and his own day. This leads one to suspect that the other two sections may also have a logical rationale, even if it is less obvious; and, especially if one takes the books in traditional order, recorded in an early quotation in the Babylonian Talmud, it is not too difficult to see what this rationale is. Both sections, like the Law, include narrative books (covering the two subsequent periods of history) and books of another kind, not legislative, but in the case of the prophets oracular, and in the case of the Hagiographa lyrical and sapiential (wisdom literature). The four narrative books in the Prophets are Joshua, Judges, Samuel, and Kings, carrying on the history from the death of Moses to the end of the monarchy; these are followed by the four oracular books Jeremiah, Ezekiel, Isaiah, and the Twelve (the Minor Prophets). In the Writings the four narrative books Daniel, Esther, Ezra‐Nehemiah, and Chronicles continue Jewish history by covering the period of the exile and return, Chronicles being put last probably because it begins with Adam and ends with the return, thus recapitulating the whole of Israel's history. Preceding these are the six lyrical and sapiential books in the Hagiographa, Psalms, Job, Proverbs, Ecclesiastes, Song of Solomon, and Lamentations, with Ruth prefixed to Psalms (but counted separately) because it ends with the genealogy of the psalmist David. Thus, there is nothing irrational (as has usually been supposed) in the isolation of Chronicles from Samuel and Kings, and of Daniel from the Prophets, in a different section of the canon.

The belief that the Law was the first complete section to be recognized as canonical is very likely true, given its traditional association with Moses. It is quite possible, however, that some of the earlier books in the other two sections were recognized as canonical alongside the Law; and these may originally have been a single collection of non‐Mosaic books, like Nehemiah's “library,” which gradually increased and was only organized in two sections after it was complete. The view that the canon of the Samaritans, which consists of the Law alone, shows that when their schism with the Jews took place this was all the Jewish canon consisted of, cannot be sustained. Qumran evidence has now made it probable that the schism did not become final until the destruction of the Samaritan temple on Mount Gerizim by the Jews in about 120 BCE. So the Samaritans of that period must certainly have rejected books which were already canonical among the Jews, probably because many of those books explicitly recognized the Jerusalem Temple.

One of the main reasons for supposing that the Hagiographa were not received into the canon until a late date is that four of them (Proverbs, Ecclesiastes, the Song of Solomon, and Esther) raise serious problems that were debated among the rabbis of the first few centuries CE. We know, however, from rabbinic literature that similar problems were raised by most of the other canonical books, and that it was only because in these books they were particularly intractable that they were taken so seriously. Moreover, there was a fifth book that raised equally serious problems, Ezekiel, which is in the Prophets. If the canonicity of this book could be debated after the first century CE, clearly the question was not one of adding books to the canon but of removing books from it. Needless to say, nothing of the kind was actually done.

It has been widely supposed that certain schools of thought had divergent canons. Some of the church fathers say that the Sadducees accepted only the Law, like the Samaritans. The Sadducees may have joined up with the Samaritans during the century after the destruction of the Jerusalem Temple (the center of their influence) by the Romans in 70 CE, but prior to that time their canon seems to have been the same as that of the Pharisees. Again, the presence of books of the Apocrypha in Septuagint manuscripts has led to the suggestion that the Hellenistic Jews of Alexandria had a wider canon than that of the Jews of Palestine. However, the voluminous writings of the first century CE Alexandrian Jew Philo are against this theory. Moreover, the manuscripts in question were produced by Christian scribes, at a date when knowledge of the Jewish canon was becoming somewhat vague in Christian circles; and in any case they are evidence of what Christians regarded as edifying reading rather than as strictly canonical. Finally, the Essenes have been supposed to have included in their canon congenial pseudonymous apocalypses such as 1 Enoch and Jubilees. In fact, they seem rather to have treated these as a sort of interpretive appendix to the canon, on a lower level of inspiration. Jude 14–15, which quotes 1 Enoch, is certainly not saying more than this about it, and in Jude's Christian (not Essene) context is probably saying less, that is, he may be using the quotation as an ad hominem argument.

The New Testament shows Jesus and his apostles endorsing a canon wider than that of the Samaritans and indistinguishable from that of the Pharisees, which now seems to have been the standard (if not, indeed, the only) Jewish canon. It had probably closed, in the form found in the Hebrew Bible, not later than the second century BCE. The threefold division of the canon, the traditional order of the books, and their standard Jewish numeration as twenty‐four may well be due to Judas Maccabeus and his advisors (see 2 Macc. 2.14, and above); the slightly later numeration of the books as twenty‐two, found in Josephus, is based on the number of letters in the Hebrew alphabet, and appends Ruth to Judges and Lamentations to Jeremiah, thus reducing the number by two.

Today, the canon exists in two main forms: that found in the Hebrew Bible, followed by Jews, Protestants, and some Orthodox churches, and that found in the Septuagint, which includes the Apocrypha, followed by Roman Catholics and also some Orthodox churches.

See also Bible


Roger T. Beckwith

New Testament

The canon of the New Testament resulted from the interplay of various theological and historical factors. The decisive factor was the impact of the person and message of Jesus Christ, together with the Christian conviction that in him as the Lord, God had spoken his final and authoritative word in history. As the Christian movement was confronted with philosophical and religious trends current in the Mediterranean world of its time, the need for an authentic expression and preservation of the foundation of its belief became the basic motivation toward the realization of the New Testament canon. This grew the more acute after the demise of the first generation of eyewitnesses. Certainly the idea of an Old Testament canon functioned as an analogy and, to a certain extent, as a stimulus in this regard. Few factors, however, expedited the growth of the idea of the canon more than the attacks from heterodox quarters, like those of Marcion, the gnostics, and the Montanists (see below). Finally the need for one, universally accepted holy book for the whole church also played a role.

Historical Survey.

Historical processes do not lend themselves to neat chronological delineations. Nevertheless, we can roughly divide the far‐reaching process through which the twenty‐seven books of the New Testament were brought into a normative, carefully delineated, and ecclesiastically accepted unit, with an authority equivalent to that of the Hebrew Bible, into four periods:

The first phase (latter part of first century CE): Creation of various early Christian documents.

Initially the gospel message was transmitted orally. In this period the young church was guided, in addition to the Old Testament, first by the apostolic witness, which developed into the apostolic tradition, and second by early Christian prophecy. In particular, the first of these factors was destined to play a decisive role in the eventual decisions about the extent of the New Testament canon. The authoritative writings would be those emanating from the apostles or the circle of those standing in a not too indirect relation to them. The prophetic witness is represented in the New Testament canon by the book of Revelation, although its ultimate recognition was determined by its association with John the apostle. The authors of the early Christian documents did not visualize their writings as part of a future canon. Rather they intended merely to give pastoral guidance to young churches. But as foundational documents, standing so close to the origin of Christianity, they possessed the inherent possibility of later becoming part of a normative collection.

The second phase (roughly from the close of the first century to the middle of the second): Growing recognition of the normative character and collection into groups of a basic number of writings.

This period, demarcated by Clement of Rome (ca. 96 CE) on the one hand and Justin Martyr (ca. 150 CE) on the other, finds oral tradition increasingly replaced by the written Gospels. Initially oral tradition was used alongside and even preferred to the Gospels. But as the reliability of the former declined, it was gradually replaced by the four Gospels. In 1 Clement, the Didache, Ignatius, and Papias, the “living voice” (as Papias terms it) of the oral tradition still enjoys preference, while in Polycarp's letter to the Philippians (ca. 135 CE) the scale tips in favor of the written Gospels. In 2 Clement (ca. 140 CE we have at least twice as many quotations from the written Gospels as from all other sources, and from the middle of the second century the written Gospels are predominant. The gnostic Gospel of Truth, for example, knows all four Gospels, while only uncertain traces of the oral tradition occur. The same holds true for Justin Martyr who makes only sporadic use of the oral tradition. Parallel to this development is the increasing use of the written Gospels for liturgical readings. Justin (Apology I.67.3–4) reports that the memoirs of the apostles (i.e., the Gospels) are not only read but also commented upon in public worship, which would put them on a par with the Old Testament books. The growing recognition of a substantial number of writings is also evidenced by the way references to them are made. In 2 Clement 2.4 and Barnabas 4.14 the gospel of Matthew, for instance, is referred to as on equal footing with Old Testament writings. At this stage we also find the collecting of early Christian writings around two foci, the Pauline correspondence and the Gospels. We can reasonably accept that by the middle of the second century the Pauline corpus as well as the four gospels were available in collections. All these factors serve to illustrate that they were increasingly recognized as normative ecclesiastical documents. The idea of a New Testament canon was beginning to emerge.

The third phase (ca. mid‐second century to 190 CE): The New Testament canon becomes a reality.

Marcion was the first person, as far as we know, who actually visualized the idea of a New Testament canon. He deliberately excluded the Old Testament from his normative collection and included only Luke and ten Pauline letters (which he purged of Jewish traits). The official church reacted by emphasizing the normative character of all four Gospels as well as all thirteen letters ascribed to Paul. Irenaeus reflects this position. In addition he probably accepted all the other New Testament books (though he does not mention all the Catholic letters) and, perhaps, also Hermas. His evaluation of 1 Clement and the Wisdom of Solomon is positive, but it is doubtful whether he regarded them as “holy scripture.” It is clear, however, that by now the idea of the canon has materialized; its broad base is fixed, but uncertainty still exists over the books on its periphery.

Final stage (ca. 190–400 CE): The closing of the canon.

It was particularly the claims to having received new revelations made by the gnostics (against whom Irenaeus already had reacted) and the Montanists, members of an apocalyptic prophetic movement, that stressed the need for a clear demarcation of the canon. Whereas the canon of Clement of Alexandria was still open, Origen and Eusebius in the Eastern church, as well as the author of the Canon of Muratori, a late second‐century list, in the West, were convinced of the necessity of a clearly demarcated canon. For that reason both the former two writers, each in their own way, differentiated between three groups of writings: the generally accepted, the uncertain, and those that should be definitely excluded. In the wake of the vehement anti‐Montanist reaction, many Eastern churches now began to question the position of the one prophetic‐apocalyptic book in the canon, namely Revelation, a question sporadically recurring until the Middle Ages. In the West the anti‐Montanist reaction took another course: in reaction against their accent on passages like Hebrews 6.4–6, the authenticity of the letter to the Hebrews became a matter of dispute which lasted until the fourth century. Uncertainty still existed in various quarters over some of the Catholic letters and also over books like the Didache, the Shepherd of Hermas, and the Wisdom of Solomon.

In the East the uncertainty was cleared up by the thirty‐ninth Paschal Letter of Athanasius, metropolitan of Alexandria, written in 367 CE, which listed all the books of the present New Testament. The writings of the apostolic fathers are excluded, but allowance is made for Hermas, the Didache, and the Wisdom of Solomon (as well as certain Old Testament apocrypha) to be read privately. It is also noteworthy that in this letter the Greek equivalent for the verb “to canonize” (in the sense of: “officially recognize as normative”) is used three times of the biblical books. (Some fifteen years earlier Athanasius had used the word “canon” in a technical sense of the Bible—the first certain occurrence of this use of the word.) Although only intended for the churches under his supervision, Athanasius's influence was such that his canon was widely approved in the East, and it greatly expedited the movement towards uniformity in the whole church. In the Western church the canon of Athanasius was probably approved at the Synod of Rome in 382 CE and definitely confirmed by a papal declaration of the year 405. Under Augustine's influence the North African church followed suite at the Synods of Hippo Regius (393) and Carthage (397), and, owing to persisting uncertainties regarding Hebrews, James, and Jude, reiterated its decisions at Carthage (419). By now the New Testament canon, with its twenty‐seven books, was almost universally accepted as the second part of the Christian Bible. The one exception was the Syrian National Church where the popular Diatessaron of Tatian, a second‐century harmonization of the Gospels, had held sway for centuries at the cost of the four “separate gospels,” and where initially strong resistance existed against some general epistles and Revelation. Here it would still take some time for the Peshitta, the official Syriac translation, to oust the Diatessaron. The small East‐Syrian Nestorian church, however, persisted with a canon of twenty‐two books (excluding 2 Peter, 2 and 3 John, Jude, and Revelation). The other exception was the Ethiopian church, which included Hermas, the two Clementine epistles, and the Apostolic Constitutions, and whose New Testament canon consists, up to our own day, of thirty‐eight books (See Bible).

Criteria of Canonicity.

In determining the content and scope of the canon various criteria were applied, e.g., that of apostolicity, the rule of faith (regula fidei), and the consensus of the churches. Of these the first played the most important role. It would be a mistake, however, to deduce from this that apostolicity was, at least initially, treated as a merely formal criterion. As attested, for instance, by the Canon of Muratori, the real consideration was rather that of reliability. As primary sources the apostles and their followers were seen as the trustworthy exponents of the original revelation given in Jesus Christ. It would also be a mistake to regard the official recognition of our present twenty‐seven books by the church as the act which gave them their canonical status. The decisions of the church were in reality the acknowledgement of the intrinsic authority and power of these writings.

See also Interpretation, History of, article on Early Christian Interpretation


Andrie B. du Toit