A new Scripture

The New Testament canon in second-century Christianity

It has been clear for some time that the second century was a (indeed, perhaps the) crucial period in the development of the New Testament.

Larry Hurtado¹

As the prior chapters have now made evident, the second century was a dynamic period for the early Christian movement. It was during this century that Christianity came into its own – distinct from Judaism and (as a result) attracting greater attention from the Greco-Roman authorities. In the midst of opposition and persecution, and even martyrdom, Christians persevered in their commitment to gather together and worship Jesus as Lord and to spread the Christian message far and wide. In order to accomplish this task, they wrote books. A lot of them. Early Christianity was a robust intellectual and literary movement, leading some pagan writers to view it as more of a philosophy than a religion.

But one accomplishment of the second century might be Christianity’s most significant. Standing on the shoulders of their Jewish forefathers, and following the precedent set by the Old Testament writings, Christians in the second century began to recognize, use and assemble some of their own books together into a scriptural collection. It wasn’t just that Christians wrote books. They actually viewed some of them as the very word of God. And it was this collection of books, what we call a New Testament canon, that would not only guide the Church through the trials of the second century, but would shape its future for the next two millennia. Indeed, it is difficult to imagine writings that have had a deeper, more profound impact on the entire history of Western civilization than the books of the New Testament, particularly the Gospels and the letters of Paul.

To be sure, the story of the New Testament canon is not restricted to the second century. Already in the first century, there are hints of an emerging canonical consciousness (e.g., 2 Pet. 3.16; 1 Tim. 5.18). Moreover, the edges of the canon were not fixed when the second century was over. It would not be until the fourth (or even fifth) century that the boundaries of the canon hardened and the controversies over some of the smaller books began to dissipate. Thus, the story of the canon was a lengthy one. And things were not always neat and tidy. Nonetheless, the second century (as we shall see) rightly takes centre stage in the whole affair. It was then that a ‘core’ canon becomes visible and established within the early Christian movement. This core not only guided the Church in her battles with various second-century heresies, but it would become the foundation for the Church’s doctrinal discussions in years to come.

But not all scholars would agree. There is a great deal of debate about whether the New Testament canon can be dated to the second century. Some scholars have insisted that it is not the second century but the fourth century which is decisive.² It is only then, we are told, that the Church first had something we could call a ‘canon’. Before this time, the state of the canon was chaotic, unsettled and lacking meaningful boundaries. In support of a fourth-century date, scholars often point out that this was the time when Christians began to make lists of which books were ‘in’ and which books were ‘out’. Indeed, we do not have a list containing all 27 books, and only these 27 books, until Athanasius’ Festal Letter in 367.

Other scholars are willing to concede that a canon (of sorts) existed at the very end of the second century, but not any earlier. Before that, it is argued, Christians did not really view their books as Scripture.³ This abrupt emergence of the canon at the end of the second century has been attributed to a number of factors, but it is often linked to the activities of the heretic Marcion. Since Marcion created his own canon – an edited gospel of Luke and a collection of Paul’s letters – it has been argued that the Church responded by creating its own larger and more full-orbed canon.⁴ Thus, the canon (as we know it) was an anti-heretical move by Christians at the end of the second century. Other scholars, not willing to grant Marcion so much influence, have instead laid the creation of the canon at the feet of Irenaeus, the bishop of Lyons.⁵ It is Irenaeus, we are told, that basically invented the idea of the canon and pushed it onto the second-century Church.

The purpose of this chapter is to probe into the state of the New Testament canon in the second century, taking into account these different currents of scholarship. We shall argue that a canon (in at least some sense) emerges not in the fourth century, nor even at the beginning of the third century, but by the middle of the second.

Obstacles to a New Testament canon in second-century Christianity

Before exploring the primary sources from the second century, we shall begin with a brief exploration of some key issues that have affected scholarly analysis of the canon during this timeframe. The historical evidence is read in the light of what a scholar thinks is possible or impossible during this century, what is likely or unlikely, or what Christians may have expected or resisted. Thus, addressing these issues on the front end can hopefully provide a clearer perspective when we examine the primary sources below.

Definition of canon

Any discussion of the date of the New Testament canon must begin with a clarification of what we mean by the term ‘canon’. The issue is not the meaning or etymology of the word itself , borrowed from the Hebrew ) but our understanding of the concept. Unless we determine what counts as a canon in early Christianity – how do we know it when we see it? – then we will have little agreement on what we are looking for in the historical sources. Brevard Childs observed this very problem a number of years ago: ‘Much of the present confusion over the problem of canon turns on the failure to reach an agreement regarding the terminology.’⁶

One of the most influential definitions of canon in recent years, and one that has proved to be a significant obstacle to a second-century date for canon, is the one advocated by A. C. Sundberg. For Sundberg, a canon exists where there is a fixed, final and closed list of books – what we might call an exclusive definition of canon.⁷ Prior to this development, Christians may have had ‘Scripture’ – a fluid, open-ended set of texts – but they did not have a canon. Fundamental to the exclusive definition, then, is a sharp distinction between the terms ‘Scripture’ and ‘canon’.

The implications of this distinction for the date of canon are profound. If one adopts the exclusive definition, then that essentially rules out the possibility of a second-century canon. As Geoffrey Hahneman observes, ‘Once a distinction is made between scripture and canon, the idea of a New Testament canon does not appear applicable until the fourth century.’⁸

While there are many positives about the exclusive definition – certainly the edges of the canon were not solidified until the fourth century – other scholars have argued that it need not be the only way to understand the concept of canon. There is an established trend among biblical scholars (dating at least back to Harnack) where ‘canon’ is used for books that function as Scripture in the early Church. In other words, we have a canon when books are being used as an authoritative norm for Christian communities even if we do not yet have a final, closed list. This functional definition of canon has been advocated by a number of scholars including Brevard Childs who does not draw a sharp distinction between Scripture canon but suggests the two terms are ‘very closely related, indeed often identical’.⁹ On this definition, we would have a canon as soon as books are being used in a scriptural manner – something that was happening (as we shall see) by the early second century.

While the exclusive definition has certain advantages, the functional definition is more likely to capture what would have been in the minds of second-century Christians. It is difficult to imagine early Christians drawing a sharp distinction between the concepts of Scripture and canon, and it is even more doubtful whether they would have regarded Scripture as a loose, open-ended concept. On the contrary, if Christians in the second century were able to say that a certain book in their library was Scripture, then this implies that they were able to say that other books in their library were not. Thus, it seems that some degree of limitation is already implied in the term ‘Scripture’ – even prior to a universal, agreed-upon list.¹⁰

For these reasons, we will employ the functional definition in this chapter (while still acknowledging the validity of the exclusive definition). We will be looking into the primary sources for indications of when early Christians began to use these New Testament books as Scripture.

Interest in a canon

In addition to the issue of definition, scholars have also pushed back against a second-century canon because of their belief that Christians were simply not interested in such a thing. Indeed, some have implied that a new written scriptural collection was the furthest thing from their minds. C. F. Evans argues that the earliest Christians ‘could hardly have conceived . . . the creation of a further Bible to go along with that already in existence’.¹¹ Such an idea, says Harry Gamble, ‘was utterly remote from the minds of the first generation of Christian believers’.¹² And what is the reason for the lack of interest in canon? Because, argues James Barr, ‘The cultural presupposition of [early Christianity] suggested that committal to writing was an unworthy mode of transmission of the profoundest truth.’¹³ In other words, Christians were more concerned about oral tradition than written texts.

In many ways, we have already dealt with this objection in the prior chapter. It is certainly true that early Christians continued to appeal to oral tradition in the second century, but there are no reasons to think that Christians were averse to written documents in such a way that would hamper the development of the canon.¹⁴ On the contrary, we argued above that Christianity was a robustly textual movement with a great interest in books of all sorts. Oral tradition and written texts seem to have worked in tandem, side by side. And it must not be forgotten that Christians had already embraced a written canon even in the first century, namely the books of the Old Testament. While there is dispute among scholars about how settled the Old Testament canon was during this period, some have argued that the boundaries of the Old Testament were fairly well established even in the first century.¹⁵

The manner in which oral tradition and a New Testament canon are not mutually exclusive can be seen in the way the ‘rule of faith’ functioned in the early Church. As we explored in Chapter 5, the rule of faith was a convenient summary – often delivered orally – of what ‘orthodox’ Christians believed was the core teaching of the apostles. But this rule did not function as a replacement for a New Testament canon nor an obstacle to the development of the New Testament canon. On the contrary, the rule of faith was viewed as ‘a summary of the Scripture’s own story line’¹⁶ or, as Irenaeus put it, ‘the order and connection of the Scriptures’.¹⁷ Thus, the rule of faith and the books of the New Testament canon had a symbiotic and mutually affirming relationship. Sometimes the rule of faith was used to validate canonical books, and other times the canonical books were used to validate the rule of faith.¹⁸ This should come as no surprise, given that both the rule of faith and the New Testament books were deemed to contain a summary of apostolic teaching.

The existence of the rule of faith, then, did not mean that second-century Christians lacked an interest in canon or that they didn’t need or desire one. Rather, the opposite seems to be the case. The rule of faith was valued precisely because it was seen to be a summary of the canonical teachings.

Disagreements over canon

Another reason that some scholars have hesitated to affirm a second-century canon is because of all the disagreements over books during this period – what we might call canonical diversity. Not only did these disagreements involve disputes over books that would eventually become canonical (e.g., 2 Peter, 2 and 3 John), but they also involved disputes over books that would never become canonical (e.g., Gospel of Peter, Shepherd of Hermas). The level of diversity and disagreement was so significant, so the argument goes, that it is not reasonable yet to speak of the existence of a canon. We must wait for a later century when more unanimity is achieved.

The problem with this line of reasoning, however, is that the level of canonical diversity is not nearly as great as claimed. And this inflated perception of diversity is driven by a number of commonly used arguments that prove to be problematic upon closer examination. We will briefly examine some of these arguments here.

Use of apocryphal writings

A common way to show canonical diversity is to point out how early Christians used apocryphal books alongside the books of the New Testament. Clement of Alexandria is a key example of a patristic writer who regularly appeals to apocryphal gospels, including the Gospel of the Egyptians and the Gospel of the Hebrews. In addition, there is the well-known story of the Christians at Rhossus who were reading the Gospel of Peter. Their bishop, Serapion, offered no objection at first but when he later read the gospel for himself, and discovered its Docetic teaching, he condemned it as ‘heresy’.¹⁹ But the key question in both these examples is whether the mere use of apocryphal material means that a canon cannot exist. Are the two mutually exclusive? Apparently not in the minds of the early Christians. Clement uses these apocryphal gospels, but does not cite them as Scripture. For him, they are simply useful resources that fit his argument (in addition to many other non-canonical sources he cites). He is very clear that the Church receives four and only four gospels as Scripture.²⁰ In the case of the church at Rhossus, there is no indication that the Gospel of Peter was being read in worship as Scripture. Indeed, it is hard to believe a bishop would allow such a practice for a gospel he had never even read! These examples show that the mere use of apocryphal material by early Christians does not prove a high level of diversity and disagreement over the New Testament canon. To confuse the usage of a book with the acceptance of that book as Scripture can give the false impression that early Christianity was more diverse than it really was.

This is not to suggest, of course, that no second-century writers regarded apocryphal books as Scripture. In all likelihood, some of them did. No doubt some of the ‘heretical’ groups discussed in earlier chapters viewed some of their books as Scripture – though explicit statements to that effect are less common than one might think. Even so, we have argued that not only were these fringe groups in the minority, but the apocryphal texts they read, even if regarded as Scripture by some, never had the widespread popularity of the canonical books. As for whether ‘orthodox’ writers ever viewed apocryphal writings as Scripture in the second century, the historical evidence for this practice is (again) rather limited. There is a possibility that Irenaeus regarded the Shepherd of Hermas as Scripture,²¹ but it is difficult to be sure.²² The fact is that Irenaeus hardly ever used the book. Similarly, Clement of Alexandria seems to have a very high opinion of the Epistle of Barnabas, but his view of it is unclear since he never cites it as ‘Scripture’ and even seems willing to critique it at times.²³ But even if these authors did regard these books as fully scriptural, we are still faced with the reality that this phenomenon was fairly limited. As we shall see below, the vast majority of the books that second-century authors plainly regard as Scripture eventually came to be part of the New Testament canon.

Disagreement over canonical writings

Scholars intent on dating the canon later than the second century not only appeal to the use of apocryphal material (as just discussed), but they also point to disagreements among ‘orthodox’ Christian leaders over canonical books. A standard example is Origen’s statement on 2 and 3 John in the third century where he observes, ‘not all say that these are genuine’.²⁴ If such disagreements occur in the third century, it is argued, then surely there cannot be a canon in the second century. However, this line of reasoning runs into a number of problems. For one, it overlooks the implications of Origen’s comment in the opposite direction. Even if some doubt these books, apparently most Christians do accept them, Origen among them. It seems that Origen is simply observing exceptions to the general pattern during this period. Second, if the mere existence of disagreement over books means there cannot be a canon, then there could not even be a canon in the fourth century (there was disagreement then too), or even in the modern day (some groups still have a different canon). Such an argument seems to set up a nearly impossible standard for the level of agreement that must be achieved before we can say we have a canon. In fact, it seems to presuppose the exclusive definition of canon discussed above – we must first have a fixed, final and nearly unanimous list before a canon exists. But as we have seen, there is no reason to think this definition is the only way to think of canon. On the functional definition, a canon can exist when books are being used as Scripture, even if absolute unanimity has not yet been achieved. And third, the persistent appeal to disagreements among Christians over some of the smaller canonical books (e.g., 2 and 3 John, 2 Peter, Jude) actually misses the bigger picture, namely that there was a tremendous amount of unity around the core books from a very early time. What-ever disagreements did exist, it seems that they meaningfully affected only a handful of books.

Arguments from silence

Another argument used to show canonical diversity is the argument from silence. Some scholars will assume that if an author does not mention or use a certain New Testament book, then it must not be regarded as canonical. This can give the impression that there was more disagreement about books than there really was. So, for example, since Irenaeus does not mention Philemon, one might conclude that Irenaeus did not have Philemon in his canon. However, one has to be careful when using such an argument. The fact that Irenaeus is silent about a book cannot be used as evidence that he rejected it or did not know it. This is particularly true for small books like Philemon because they were used and cited so infrequently – probably due to their small size. The argument from silence is also commonly used when discussing the state of the New Testament canon in the Apostolic Fathers. Although these writings do refer to some canonical books (more on this below), a number of canonical books are not mentioned. Should we take this as evidence that the canon during this time did not include these omitted books? Not necessarily. Again, we cannot always know what books an author knew and used. And we should not forget that the extant writings we possess from these authors are often very limited. It is unreasonable to expect that all the books they received as Scripture would be mentioned in every single work they wrote, no matter the length.

Once the problems with the argument from silence are understood then we can reconsider the way we describe the canon of individual authors. For example, when it comes to Polycarp (whom we will discuss below) scholars often describe his Pauline letter collection as consisting of Romans, 1 Corinthians, Galatians, Ephesians, Philippians, and 1 and 2 Timothy. But instead, we should say that his collection consisted of at least these books – it may have been more extensive and could have (in principle) contained all 13 letters of Paul. In other words, it is not so much that Polycarp’s canon is limited to these books (though that is possible), but rather the available evidence about his canon is limited to these books.

The state of the New Testament canon in second-century Christianity

Now that we have addressed some of these preliminary considerations (and obstacles), we have a clearer path to explore the state of the New Testament canon in the second century. And, not surprisingly, we will begin with one of the most influential figures during this period: Irenaeus of Lyons.

Irenaeus and the ‘creation’ of the New Testament

The reason for Irenaeus’ place of prominence in most discussions of the canon is not difficult to discover. In the midst of his battles with the heretics of his day, he provides one of the clearest articulations of the state of the canon in the second century, affirming the scriptural status of at least the four Gospels, Acts, all the Pauline epistles (minus Philemon), Hebrews, James, 1 Peter, 1 and 2 John, and Revelation – 23 out of the 27 books in the New Testament. Thus, Irenaeus is an example of how there was a ‘core’ canon of books in the second century, despite the fact that there was continuing debate over some of the smaller books. But it is not just a matter of which books he cites but the frequency with which he cites them. He cites over 1,000 New Testament passages in total, demonstrating an impressive scope of knowledge (and use) of these books. Beyond this, he offers one of the most well-known defences of the fourfold Gospel canon when he declares: ‘It is not possible that the gospels can be either more or fewer than the number they are. For, since there are four zones of the world in which we live and four principal winds . . . [and] the cherubim, too, were four-faced.’²⁵

It is precisely these sorts of statement that have led some scholars to claim that Irenaeus was the creator of the New Testament canon. After all, it is argued, the tortured logic used to justify four gospels seems to suggest that Irenaeus is on the defensive and looking for a way to justify his new canon of Scripture.²⁶ He must be an innovator looking to bolster and promote his new position. The canon of Irenaeus, therefore, is viewed as ‘a radical departure from traditional Christian practice’.²⁷

But this sort of approach runs into a number of problems. Most significantly, Irenaeus does not use these scriptural books in a manner that suggests it was an innovation. He cites these books naturally, organically and unapologetically, with every confidence that his audience is familiar with them.²⁸ There is no indication that he is pushing a new scriptural collection of books onto an audience that did not previously have one. As for his justification of the fourfold Gospel, the appeal to the four creatures around the throne or the four zones of the world may sound unusual to our modern ears but would not have seemed so in the world of the early Christians. The number four was already seen as symbolically representing completeness and fullness. Thus, Irenaeus appears to be reflecting on the Church’s long-standing practice of using four gospels and is simply looking for a theological principle to explain it. And he found it in numeric symbolism.²⁹

But there is an additional (and more significant) reason to reject the Irenaeus-as-innovator model of canon. Simply put, Irenaeus was not the only one who had a canon at the end of the second century. We have a number of other sources from this same period that appear to have a remarkably similar collection of books to Irenaeus. And this suggests that the scriptural status of these books must have roots that go back further still. It is to these sources we now turn.

Further evidence of a ‘core’ canon in the second century

Muratorian Fragment

In approximately the same timeframe as Irenaeus, our earliest canonical list, the Muratorian Fragment, affirms nearly the same ‘core’ collection of books: the four Gospels, Acts, 13 letters of Paul, Jude, 1 and 2 John, and Revelation – 22 out of 27 books. Unmentioned are Hebrews, James, 1 and 2 Peter and 3 John, though this is not the same as rejection. The list also mentions the apocryphal work the Apocalypse of Peter, but the author seems reluctant as he admits ‘some of us are not willing that the latter [Apocalypse of Peter] should be read in church’. In addition, the list excludes a number of books including the Shepherd of Hermas and the writings of Valentinus, Basilides, and the Asian founder of the Cataphrygians (probably a reference to Montanus).

The existence of a list such as this in the latter half of the second century presents a serious challenge to the idea that Irenaeus was the sole architect of the canon. Indeed, this list does what some have regarded as unthinkable: it begins to draw hard lines around books by arguing some are ‘in’ and some are ‘out’. Such an approach seems difficult to square with the idea that the Christians in the second century only had a ‘boundless, living mass of heterogeneous’ texts.³⁰ On the contrary, the Muratorian Fragment seems to suggest that the canon was already at a relatively mature (though not complete) stage where restrictions and limitations were not uncommon.

Because the Muratorian Fragment does not fit with certain reconstructions of the canon during this time, some scholars have argued that it must belong to a later time, namely the fourth century. However, the attempt to push the date back has not found widespread support and the second-century date still seems to represent the consensus of modern scholars.³¹

Theophilus of Antioch

In his apologetic treatise to a pagan named Autolycus (c.177), Theophilus of Antioch seems to provide yet another echo of a ‘core’ canon at the end of the second century. Aware that the newness of the Christian Scriptures presented a philosophical problem for those in the Greco-Roman world, Theophilus presented a counter-argument that the Gospels shared the same inspiration of the Old Testament Scriptures and thus should be accepted. He writes: ‘Concerning the righteousness which the law enjoined, confirmatory utterances are found both with the prophets and in the Gospels, because they all spoke inspired by one Spirit of God.’³² As examples of what ‘Gospels’ he had in mind, Theophilus proceeds to cite Matthew³³ and Luke³⁴ alongside Old Testament texts and plainly affirms the scriptural status of John.³⁵ Although he is silent about Mark’s Gospel, the fact that he later produced a fourfold Gospel harmony (similar to Tatian’s Diatesseron) suggests that he received it as authoritative alongside the other three.³⁶ As for Paul’s letters, Robert Grant has made a case that Theophilus probably knew of a ‘collection of Pauline letters, probably including the Pastoral Epistles’,³⁷ and there is additional evidence that Theophilus regarded them as a ‘divine word’³⁸ that was on par with the authority of the Gospels.³⁹ He also seems to have regarded the book of Revelation as Scripture.⁴⁰ As for the Catholic Epistles, there are allusions here and there,⁴¹ but the evidence is less clear about which of these he accepted. Of course, it must be kept in mind that this treatise is the only extant work of Theophilus we possess, so it is difficult to know the full extent of his canon.

Regardless, and even with the limited information in our possession, it seems that the canon of Theophilus bears a notable resemblance to the canons of Irenaeus and the Muratorian Fragment. He holds to at least the four Gospels, the Pauline epistles, Revelation, and probably a few of the smaller books.

Clement of Alexandria

Writing at the end of the second century (c.198), Clement, head of the catechetical school of Alexandria, provides yet another glimpse into the ‘core’ canon of the second century. Sounding (again) a lot like Irenaeus, Clement affirms the canonical status of at least the 4 Gospels, 13 letters of Paul, Hebrews, Acts, 1 Peter, 1 and 2 John, Jude and Revelation. Given Eusebius’ testimony that Clement wrote commentaries on all the ‘Catholic Epistles’, it seems possible that his canon may have also included James, 2 Peter and 3 John, but it is difficult to be sure. As already observed, Clement often cited from a variety of sources outside the New Testament, even including apocryphal gospels, but there is no indication he regarded these books as sharing the same scriptural status as those books in the canon. Such a practice not only reflects Clement’s vast learning but also is fitting with the cultural milieu of Alexandria where intellectuals interacted with many other authors and currents of thought.⁴² The closest Clement came to affirming the scriptural status of a non-canonical book is probably the Epistle of Barnabas, but (as noted above) the evidence for this is less than clear. But even if Clement did regard Barnabas as Scripture, that does not affect the overall point we are making, namely that a core canon still existed even in the midst of continuing disputes about a handful of peripheral books.

Earlier roots of the canon in the second century

The above discussion demonstrates that Irenaeus was not alone when he affirmed approximately 23 out of 27 New Testament books at the end of the second century. On the contrary, a number of other sources confirm that this was a fairly widespread phenomenon by this period. But this raises the question of how far back we can see this New Testament canon developing. Unless we argue for a ‘big bang’ theory of canon where it popped into existence very rapidly, we might expect to see traces of this phenomenon even earlier. So we now explore evidence from the middle and early second century.

Justin Martyr (and Tatian)

When it comes to the four Gospels, Justin Martyr provides additional evidence that they were used and received as Scripture very early. Although Justin does not refer to the four Gospels by name, we have good reasons to think that he had Matthew, Mark, Luke and John in mind: (a) he refers to the Gospels as ‘drawn up by His apostles and those who followed them’,⁴³ indicating that he received (at least) two gospels from apostles and (at least) two gospels from their followers – a likely reference to the canonical four; (b) he cites from all three Synoptic Gospels,⁴⁴ even calling one of them ‘[Peter’s] memoirs’,⁴⁵ a likely reference to Mark; (c) he shows familiarity with the ‘logos’ theme of John’s Gospel⁴⁶ and even appears to cite John directly, ‘For Christ also said, “Except ye be born again, ye shall not enter into the Kingdom of heaven”’;⁴⁷ and (d) his disciple was Tatian who produced the famous four-Gospel harmony called the Diatesseron, making it highly unlikely Justin would have been unfamiliar with the canonical four.

But even if Justin was aware of (and used) these four Gospels, some scholars have argued he really didn’t think they were Scripture. This is why, it is argued, Justin doesn’t refer to them by name but instead calls them the ‘memoirs of the apostles’.⁴⁸ In response, it should be observed that Justin plainly affirms the authority these books when he tells us that they were read in worship right alongside the Old Testament Scriptures and even expounded upon by the pastor/preacher.⁴⁹ Moreover, Justin sometimes uses the phrase ‘it is written’ when citing these Gospels which typically is used to indicate scriptural books.⁵⁰ In fact, Martin Hengel argues that Justin’s testimony shows that the Gospels had, in some ways, even surpassed the Old Testament books by this time, and have ‘taken over the significance of the Jewish reading of the Torah’.⁵¹

In regard to how many other New Testament books Justin used, there is disagreement among scholars. His high respect for Revelation is clear since he regards it as apostolic and even calls it ‘one of our [Christianity’s] writings’.⁵² But it is unclear how many of Paul’s letters he knew or used. While we cannot be sure, it seems likely that Justin would have known Paul’s letters, given his interactions with Marcion who clearly did. As for why Justin didn’t utilize Paul more plainly in his Apology and his Dialogue, it may have been because of the apologetic nature of these works. Simply put, he used the sources that would be most effective in his interactions with unbelievers.⁵³

Marcion

As noted in Chapter 4, Marcion was in many ways the Church’s first ‘heretic’. He moved to Rome around ad 144 and began to promote his controversial teachings about the God of the Old Testament and the person of Christ. Most notably, however, Marcion was remembered for the way he edited and changed the New Testament writings. Irenaeus himself is plain about what Marcion had done:

Wherefore also Marcion and his followers have betaken themselves to mutilating the Scriptures, not acknowledging some books at all; and, curtailing the Gospel according to Luke and the Epistles of Paul, they assert that these are alone authentic, which they have themselves thus shortened.⁵⁴

According to Irenaeus, Marcion took an existing corpus of Scriptures – which would probably have included the four Gospels, the Pauline letters and perhaps some other books – and received only a portion of it. Marcion’s portion included Luke and just ten epistles of Paul (leaving out the Pastorals). This shrinking of the canon allowed Irenaeus to say that Marcion was guilty of ‘not acknowledging some books at all’. If Irenaeus is correct, then we have good reasons to think a larger collection of New Testament books existed before Marcion mutilated it in the 140s – a collection that apparently did not look all that different from Irenaeus’ own.

Of course, modern scholars have challenged Irenaeus’ own account, suggesting that he is not to be trusted. Instead, we are told, Irenaeus was simply rewriting history to suit his own purpose, namely to portray ‘orthodoxy’ as earlier and ‘heresy’ as later.⁵⁵ It was actually Marcion, so the argument goes, that invented the idea of a New Testament canon and it was the orthodox leaders (like Irenaeus) who were simply reacting to him. This idea that Marcion was the impetus behind the Church’s creation of the canon is a long-standing scholarly position dating back to the writings of Adolf von Harnack and still advocated by some today.

Even so, the scholarly tide has now turned against the idea that Marcion played such a prominent role in the development of the canon. Indeed there are now doubts about the level of his influence and whether his truncated canon was even intended by him to have a scriptural status at all.⁵⁶ Regardless, what does seem to be clear is that a larger corpus of books pre-dated Marcion’s activities. If so, then events surrounding Marcion provide yet another testimony to a ‘core’ canon in the earliest period. As John Barton observes: ‘The New Testament books, or at any rate the central “core” of the Gospels and the Pauline and the Catholic Epistles, were already used very widely in the time before Marcion, and continued to be so used after him.’⁵⁷ David Balás reaches a similar conclusion: ‘Thus Marcion did not initiate the notion of the Canon but attempted to reform an already existing body of normative writings.’⁵⁸

The fact that some sort of larger New Testament canon existed before Marcion reduced it will find further confirmation in a number of other historical sources we will explore below. But it is worth noting here that even Marcion’s own version of Luke’s Gospel seems to indicate that this was the case. Although reconstructing the text of Marcion’s Luke is notoriously complex – largely due to the fact that it is only available to us in the citations of patristic writers like Tertullian and Epiphanius⁵⁹ – scholars have observed for generations that it already exhibits harmonizing tendencies with the other Synoptic Gospels.⁶⁰ If so, then this may be an indication that, by the time of Marcion, the Gospel of Luke had already been circulating with Matthew and Mark or at least tightly associated with these other Gospels in the minds of the scribes that had been copying it. And if three Gospels were connected together in some fashion, it is hard to avoid the implication that we might have echoes here of a fourfold Gospel unit. This sort of scenario suggests that Marcion probably encountered a fourfold gospel that already existed and then, for his own theological reasons, cut it down to just a single gospel (Luke) – something that even Harnack himself affirmed.⁶¹ Of course, this sort of argument must still remain very tentative given that the harmonizations may have seeped into Marcion’s text in a number of other ways (including from the hands of Tertullian and Epiphanius!).

Aristides

As noted in an earlier chapter, Aristides of Athens penned his Apology to Emperor Hadrian in about ad 125.⁶² What is noteworthy about the Apology is that it invites the emperor to actually read what Aristides calls ‘the Gospel’.⁶³ He states: ‘This is taught in the Gospel . . . and you also if you will read therein, may perceive the power which belongs to it.’⁶⁴ While Aristides may have just a single Gospel in mind, the fact that he draws upon material from a number of our canonical Gospels suggests that he is using the singular ‘Gospel’ to refer to the plurality of written sources about Jesus that the Church had received.⁶⁵ This mix of singular and plural language for the Gospel(s) reflects the conviction of early Christians that there was only one Gospel even though it was presented through a multiplicity of witnesses. In fact, we see Justin and Irenaeus use the singular ‘Gospel’ in a similar manner, even though they clearly recognized four Gospels.⁶⁶ For example, at one point Irenaeus states: ‘Let us revert to the Scriptural proof furnished by those apostles who did also write the Gospel, in which they recorded the doctrine regarding God, pointing out that our Lord Jesus Christ is the truth.’⁶⁷ In this statement, Irenaeus even mentions plural apostles as the authors of the singular ‘Gospel’ – showing that he viewed it as a single entity despite its fourfold form.

As for which Gospels Aristides had in mind, he draws upon material that clearly comes from Luke and John, and possibly even Mark and Matthew.⁶⁸ It is unlikely that this material is drawn from oral tradition because, as noted above, he has told the emperor that he can ‘read’ about these things, clearly pointing to a written source. Indeed, this written source must be well known and well established because the language of Aristides implies that the emperor could obtain a copy if he so wished.⁶⁹ Aristides’ high opinion of these Gospels is also evident through his claim that whoever reads them would ‘perceive the power which belongs to it’ – a likely reference to his view that these books bear the power of the Spirit of God. At one point, Aristides also mentions ‘other writings’⁷⁰ beyond the Gospels, a likely reference to Paul’s letters and perhaps even Peter’s as indicated by echoes of these books in the Apology.⁷¹ Although we cannot be sure of the specific make-up of Aristides’ collection, it probably contained a number of our canonical Gospels (probably four of them) alongside of a collection of apostolic letters. Again this fits the pattern we have seen so far in the second century.

Apostolic Fathers

As observed in the prior chapter, one of the earliest collections of Christian writings we possess outside the New Testament is what scholars refer to as the Apostolic Fathers. These writings did not circulate together as a unit in the ancient world, but scholars link them together largely because they derive from the same general period, the first half of the second century (with a few exceptions). Space prevents us from undertaking a thorough investigation of these books, but a few select examples can still shed some light on the state of the New Testament canon during this period. Although the excerpts from Papias are considered part of this corpus, we will examine those in a separate selection below.

Before we look at these examples, it should first be acknowledged that some scholars are generally sceptical about whether the Apostolic Fathers can tell us much of anything about the development of the canon.⁷² The basis for this position is the belief that Christians during this period were relatively uninterested in written texts and relied almost exclusively on oral tradition. The use of oral tradition, it is argued, is also evident from the fact that the text of the citations in the Apostolic Fathers is different from what we find in the New Testament. The wording is just not the same. Thus, it is argued, whenever these authors seem to be citing New Testament books it is more likely they are actually drawing on oral material.

To some extent, this issue has already been addressed at prior points in this volume where we argued that the portrayal of early Christianity as an oral religion is a bit misleading. The Christian movement was a movement that possessed a robust textuality and commitment to written texts – fitting with its Jewish roots and the early Christian commitment to the Old Testament. Moreover, we should remember that while most Christians were illiterate, that is obviously not the case with the Apostolic Fathers themselves. They were clearly educated, composed their own texts, and interacted extensively with other written texts. Thus it is reasonable to expect that such authors, more than the average Christian, would be aware of (and utilize) early Christian writings and not just oral tradition. As for inexact citations, we must be careful not to assume this can only be explained by the use of oral tradition. On the contrary, it was not unusual for authors to paraphrase, summarize or conflate biblical passages when they are being cited.⁷³ Ancient authors, unlike modern ones, felt no need to offer an exact citation of the written text being utilized. In fact, we often see this practice when patristic writers cite the Old Testament, and even when the New Testament writers themselves cite the Old Testament (e.g., Mark 1.2–3).

With these considerations in mind, we now turn to a few examples:

The Epistle of Barnabas (c.130) seems to be citing Matthew 22.14 when it says: ‘As it is written, “Many are called, but few are chosen.”’⁷⁴ The wording of this citation matches only the text from Matthew and cannot be attributed to oral tradition due to the fact that the quote is introduced with the citation formula, ‘As it is written’. This citation formula also reveals that the author viewed the source of this citation as scriptural.

Ignatius, writing to the Ephesians on the way to his martyrdom (c.110), actually refers to a Pauline letter collection: ‘Paul, who was sanctified, who gained a good report, who was right blessed, in whose footsteps may I be found when I shall attain to God, who in every epistle makes mention of you in Christ Jesus.’⁷⁵ Although we cannot be sure of the extent of Ignatius’ collection, it would have at least included the letters in which Paul referred to the Ephesians (1 Corinthians, Ephesians, 1 and 2 Timothy),⁷⁶ and would have probably been larger since he shows evidence of also knowing Romans, Philippians and Galatians.⁷⁷ The fact that Ignatius would have regarded these letters as authoritative is shown by his very high view of the apostolic office and the person of Paul.⁷⁸ In addition, there is good evidence that Ignatius knew of writings from Peter⁷⁹ and also knew and used the Gospels of Matthew, Luke and John.⁸⁰

Polycarp (writing c.110) also seems to be aware of a Pauline letter collection, consisting of at least Romans, 1 Corinthians, Galatians, Ephesians, Philippians, and 1 and 2 Timothy. It is clear that he regarded Paul’s letters as Scripture because he not only acknowledges the unique authority of the apostolic office, but also cites from Ephesians with the introductory phrase, ‘As it is written in these Scriptures’.⁸¹ Polycarp’s commitment to the Pauline letters led Charles Nielsen to argue that by the early second century, ‘a sacred Christian Scripture was emerging with the Pauline corpus as its foundation’.⁸² In addition, Polycarp knows other apostolic letters such 1 Peter and 1 John,⁸³ and there are good reasons to think he would also have known at least the Gospels of Matthew and John.⁸⁴

The letter of 1 Clement also provides evidence of an early Pauline letter collection, consisting of at least 1 Corinthians, Romans, Hebrews (assuming it to be Pauline) and probably also Galatians, Ephesians, Philippians and Titus.⁸⁵ It is clear that the author attributes high authority to these writings because he affirms Paul’s unique apostolic office⁸⁶ and even exhorts the Corinthian church to ‘Take up the epistle of that blessed apostle Paul. . . . To be sure, he sent you a letter in the Spirit’⁸⁷ – a likely allusion to Paul’s inspired status. Donald Hagner also argues that the author of 1 Clement knew 1 Peter, James and Acts, but it is unclear how he viewed these books.⁸⁸ However, in the case of 1 Peter, it is likely he would have attributed the same level of authority to it as to the letters of Paul.

Although the evidence in the Apostolic Fathers (and this is just a sampling) is certainly less clear than what we find in the late second century – the further back we go the murkier our sources become – we still see the basic outlines and echoes of a New Testament canon. Generally speaking, the shape of the canon still seems to consist of Gospels, Pauline letters and a few smaller books, even though the specific number of each of these are not always knowable. And even without all the detail we would wish, a fundamental and overarching principle does emerge from the Apostolic Fathers, namely that Christians, even in the early second century, were quite comfortable regarding apostolic writings as Scripture alongside the Old Testament.

Papias

The final source we will examine in our exploration of the canon in the second century is Papias, bishop of Hierapolis (c.125). The testimony of Papias is particularly noteworthy for a couple of reasons. First, Papias was a bit of a detective, actively investigating, pursuing and collecting information about Jesus. We are told that he ‘inquired’⁸⁹ into the teachings of the Lord’s disciples and even penned his own (now lost) book, Exposition of the Sayings of the Lord, that no doubt contained the fruit of his inquiries. Second, Papias is noteworthy because of his unique historical position – he was a friend of Polycarp and a hearer of the apostle John⁹⁰ – and because he draws his information from an even earlier source, a person he refers to as ‘the Elder’ (or Presbyter). Most scholars agree that this is probably a reference to ‘John the Elder’ (cf. 2 John 1; 3 John 1) who, Papias tells us, was a ‘disciple of the Lord’ and thus an eyewitness and follower of Jesus himself.⁹¹ On a prima facie level, it seems likely that John the Elder is simply a reference to John the apostle, the son of Zebedee – especially since Papias knew this John.⁹² However, other scholars have argued that John the Elder is a different John from the apostle. Regardless of which John is in mind (and we cannot resolve the debate here), the key fact is that Papias is drawing upon a source that takes us back to the end of the first century, a remarkably early (and important) period in the development of the canon.

So, what does this ‘Elder’ tell us about the canon during this critical period? Here are the words of Papias:

The Elder used to say: Mark became Peter’s interpreter and wrote accurately all that he [Peter] remembered. . . . For he was intent on just one purpose: to leave nothing out that he had heard or to include any falsehood among them . . . Matthew collected the oracles in the Hebrew language and each interpreted them as best he could.⁹³

We can learn a number of important things from the testimony of Papias’ Elder. For one, it tells us that two of our canonical Gospels, Matthew and Mark, were known by name toward the end of the first century. This reality stands in contrast to the repeated claims of some scholars that Gospel material during this period was primarily oral with no concern for written accounts. Moreover, it demonstrates that these Gospels did not circulate anonymously only to have a name randomly and artificially attached to them in the middle of the second century, as some scholars maintain.

In addition, we learn that these Gospels had apostolic connections, Mark drawing from the teachings of Peter (a tradition repeated in a number of other historical sources) and Matthew being himself an apostle. In this way, Papias sounds like Justin Martyr as he acknowledges that some Gospels come from apostles themselves and some from companions of apostles. No doubt these apostolic connections were highlighted by Papias (or his Elder) to establish the unique authority of these particular accounts of the life of Jesus. This emphasis on the trustworthiness of these Gospels is also evident in the statement that Mark determined ‘to leave nothing out that he had heard or to include any falsehood among them’.⁹⁴ Clearly, these Gospels were viewed as fully authoritative and historically reliable even at the end of the first century.

Although the other Gospels aren’t mentioned in these excerpts from Papias, we have good reasons to think that he knew them.⁹⁵ Given that Papias also knew 1 John and Revelation, and sat under the apostle John’s preaching, it is highly likely he knew the Gospel of John – and he even appeals to information that only appears in John.⁹⁶ Papias also cites what appears to be Luke 10.18, suggesting he also knew that Gospel as well.⁹⁷ There are even indications he knew some of Paul’s letters, especially 1 and 2 Corinthians.⁹⁸ Thus, it seems that Papias had a canon composed of a possible fourfold Gospel, some letters of Paul, and a few other books like 1 John and Revelation – further confirmation of the trend we have seen all along. And the evidence he provides takes us further back than any other source, even into the first century.

Conclusion

Although the story of the New Testament canon certainly extends well beyond the second century, this period remains critical in our understanding of how and when the early Church recognized a new corpus of books as Scripture. It is in the second century, we have argued, that we can see the emergence of a ‘core’ canon generally consisting of the four Gospels, the letters of Paul, and a handful of other books. Despite the claims of some, there seems to be little reason to think that this core is due to the innovations of Irenaeus because it was present already in a number of his contemporaries, Clement of Alexandria, the Muratorian Fragment and Theophilus of Antioch. Moreover, we can see traces of this core canon even further back in the writings of Justin Martyr, the activities of Marcion, Aristides and the Apostolic Fathers (particularly Papias).

Certainly the boundaries of the canon had not yet hardened – that would not happen until the fourth century or later. But the essence of the canon seems to be in place even by the first half of the second century. Thus, we can agree with the assessment of John Barton: ‘Most of the present Christian Bible was already “Scripture” before the mid-second century, while on the other hand decisions about its exact boundaries still lay in the future.’⁹⁹

If we are correct, then the existence of a ‘core’ canon has important implications for other aspects of our study of the second century which we have already covered. For example, we can see more clearly why the persecution of early Christians – both intellectual and political – often involved their sacred writings. Celsus attacked not simply generic Christianity but the narratives of Jesus contained in the four canonical Gospels.¹⁰⁰ Also, in the Acts of the Scillitan Martyrs, the proconsul Saturninus interrogated one of the Christians, Speratus, by asking what was in his bag. Speratus responded: ‘Books and letters of Paul, a righteous man.’¹⁰¹ And even in Lucian’s mocking satire of Christians, he mentioned how Christians visited Peregrinus, and the ‘sacred books of theirs were read aloud’.¹⁰² And in response to such criticisms, the apologists would often cite these canonical books or even call upon their adversaries (or the emperor) to read them – as seen in Aristides of Athens, Theophilus of Antioch and Justin Martyr.

A second-century canon also illumines our understanding of early Christian worship which we explored in Chapter 3. Central to the worship gatherings of early believers was not only the reading of the Old Testament Scriptures – but books of the New Testament as well. Indeed, we have seen in Chapter 6 how the format of the New Testament manuscripts indicates they were designed for public reading. Moreover, Christian worship involved teaching, preaching and catechizing on the basis of these books. This present chapter reminds us that these New Testament writings really were known and used in the worship of second-century Christians.

But perhaps more than anything else, the existence of a second-century ‘core’ canon has significant implications for our discussions of ‘heresy’ and ‘orthodoxy’. While Bauer has argued that there was no clear definition of orthodoxy until the fourth century, and that early Christianity was wildly diverse with little agreement on much of anything, the existence of a core canon of New Testament writings suggests otherwise. The fact that the great bulk of the New Testament canon was in place so early suggests that the vast majority of Christians – what Celsus called ‘the Great Church’ – were quite agreed about which books to use and read. And this broad and widespread commitment to the four Gospels, the letters of Paul, and a few other books would have provided the doctrinal foundation for the Church to define orthodoxy and battle against what they perceived to be heresy. The subsequent debates about smaller books like 2 Peter or 3 John would do little to change this fact. Regardless of the outcome of the debates over these peripheral books, the larger and more foundational theological trajectory had already been determined. Thus it appears the existence of the New Testament canon is not the result of debates over orthodoxy and heresy, but the very thing that allowed the Church to survive in the midst of those debates.

↑ Back to top