When did the New Testament become Scripture?

Introduction

The earliest extant list (or ‘canon’) of books considered to be authoritative is ‘The Muratorian Canon’ (late second century). The first part of this text is missing but given it enumerates Luke and John as ‘third’ and ‘fourth respectively, it is likely to have begun with reference Matthew and Mark. It subsequently lists all the books of the NT as received by the Church, except Hebrews, James, 1 Peter, 2 Peter and one of the epistles of John.[1] Works that are considered useful but not received are also listed including the Apocalypse of Peter, the Wisdom of Solomon, and the Shepherd of Hermas.

Whilst this is earliest surviving list of NT books, this does not mark the point when the NT books were first recognised as Scripture. The fact that Marcion (c.140) rejected all but ten of Paul’s epistles and Luke’s gospel is a strong indication that the wider Christian community already recognised certain books as authoritative.

The following is a short historical study intended to show what historical evidence is available that witnesses to the authoritative status of NT books.

Early Christian Understanding of OT Scripture

In Greek, ‘Scripture’ is the noun grafh, (graphē) and literally means ‘a writing’; it was used in Greek texts to refer to any piece of writing. However, in the NT, this noun is used exclusively to refer to special sort of writing: Scripture. This exclusivity does not mean an early Christian could not have used γραφή to refer to an ordinary text. What primarily determines whether grafh, refers to Scripture or some other ordinary text is absence of any complement (i.e. ‘the writings’, rather than ‘the writings of x’).

The NT witnesses to a developed understanding of what Scripture is – an understanding inherited from Jewish precursors. Scripture, though given through humans, was spoken by the Holy Spirit (Acts 1:6; 2 Tim 3:16) and not the result of subjective speculations (2 Peter 1:20-21). It was not only held to be useful (Rom 15:4; 2 Tim 3:16) but also unbreakable (John 10:35). The NT writers not only believed Scripture to contain unshakeable predictions of future events (e.g. Matt 26:54; Luke 22:37; John 13:18, 19:24, 28, 36-37, 20:9; Rom 1:2; Gal 3:8; James 2:23; Acts 1:16) but to be an authoritative source of information (e.g. Matt 22:29; Mark 12:24; Rom 10:11; Gal 3:22; 1 Tim 5:18; 1 Pet 2:6).

The NT also refers to Scripture using the related verb gra,fw (graphō). This is not used exclusively of Scripture but, again, the way the word is used in context determines its meaning. When a quotation is introduced with the words ‘it is written’ without any further marker (e.g. ‘in the books of x’) then the quotation comes from Scripture.

Looking at the way NT writers use both noun and verb to introduce scriptural quotations we can sketch out which books they held to be Scripture:

• Genesis (Gal 3:8; James 2:23)
• Exodus (Matt 22:29; Rom 9:17)
• Leviticus (James 2:8)
• Numbers (John 19:36)
• Deuteronomy (Matt 4:4; Rom 11:8)
• 2 Samuel (Rom 15:9)
• 1 Kings (Rom 11:2)
• Nehemiah (John 6:31)
• Job (1 Cor 3:19)
• Psalms (Rom 15:4; John 13:18)
• Ecclesiastes (Rom 3:10)
• Isaiah (Rom 10:11; 1 Pet 2:6)
• Jeremiah (1 Cor 1:31)
• Ezekiel (Rom 2:24)
• Micah (Matt 2:5)
• Amos (Acts 7:42)
• Habakkuk (Rom 1:17)
• Zechariah (Matt 26:31; John 19:37)
• Malachi (Matt 11:10; Rom 9:13)

This list is a good sampling of the OT but is by no means exhaustive. Luke 24:27 refers to Scripture as including the books of Moses and all the Prophets; further down v. 44 refers to “the Law of Moses, the Prophets and the Psalms”, which would appear to be a reference to the three sections of the Hebrew Bible. We have good reason to suppose that the early Christian understanding of Scripture, like that of their Jewish contemporaries, included the 39 books of our OT.

Inclusion of NT as Scripture

Having demonstrated that the early Christians, as witnessed by the NT writers, had a specific understanding of Scripture, we can apply the same analysis of grafh, and gra,fw to early references to the NT. Other early Christians shared with the same understanding of Scripture with the NT writers. For example 1 Clem 45.2 says that the Scriptures were given “through the Holy Spirit” and contain nothing “unjust or counterfeit”. Similarly, the Preaching of Peter says that the OT prophecies are Scripture and were decreed by God (fr.2). Reasonably, when a NT text is described as being Scripture or a quotation from a NT text is introduced as scriptural then this is evidence that (some) early Christians included this text in their conception of Scripture, that is, as given by the Holy Spirit and authoritative.

(A) 1 Timothy

For the Scripture (h` grafh,) says, “You shall not muzzle an ox when it treads out the grain,” and, “The labourer deserves his wages”. 1 Tim 5:18

Paul introduces two quotations as Scripture; the first is from Deuteronomy 25:4; the second is from Luke 10:7. If 1 Timothy is written by the apostle Paul, then it must date before 66/7 AD, which is the likely date of Paul’s execution. J. A. T. Robinson, connecting 1 Tim 1:3 with Acts 20:1, assigns it to autumn of 55 AD,[2] but this would date it earlier than the composition of Luke.[3] Conservative scholars have sometimes speculated that Paul was released from prison and travelled to Spain prior to his final imprisonment and execution; on this basis, A. D. Norris dates 1 Timothy to the autumn 66 AD.[4] The current critical consensus is that 1 Timothy is pseudepigraphal and it is consequently dated late. Even if this is the case, it is likely to have been written before the end of the first century since it is known to Polycarp (c.110; Pol. Phil 4.1, 12.3). If we assign 1 Timothy to the range 60-100 (taking most options on board) and Luke to the range 60-80 (again, taking on most options) then Luke is acknowledged as Scripture (by one writer, at least) no more than 40 years after composition, probably much sooner.

(B) 2 Peter

And count the patience of our Lord as salvation, just as our beloved brother Paul also wrote to you according to the wisdom given him, as he does in all his letters when he speaks in them of these matters. There are some things in them that are hard to understand, which the ignorant and unstable twist to their own destruction, as they do the other Scriptures (grafa.j). 2 Pet 3:15-16

Peter refers to Paul’s letter as Scripture. The phrase ‘all his letters’ may not refer to the full number of Pauline letters in our NT but just those known to Peter. If 2 Peter is written by the apostle Peter then it must date before 66/7 AD, which is the likely date of his crucifixion. Robinson dates it to 61/2 AD.[5] Scholarly consensus considers 2 Peter to be inauthentic and late; it is acknowledged that the external testimony for 2 Peter is, arguably, the weakest of any NT book. However, the dependence of the Apocalypse of Peter (c.110) and 1 Clement on 2 Peter require that it be dated at least within the first century.[6]

Again, taking on board the various options for the sake of argument, if we assign 2 Peter to the range 60-100 and Paul’s letters to 50-66 then (some of) the Pauline letters were acknowledged as Scripture (by one writer, at least) no more than 50 years after composition, probably much sooner.

(C) Epistle of Barnabas

And all the more attend to this, my brethren, when ye reflect and behold, that after so great signs and wonders were wrought in Israel, they were thus [at length] abandoned. Let us beware lest we be found [fulfilling that saying], as it is written (ge,graptai), ‘Many are called, but few are chosen’. Barn. 4.14

This anonymous epistle, wrongly ascribed to Barnabas, introduces the words of Jesus (cf. Matt 20:16; 22:14) with the formula ‘it is written’. This indicates both that the author was dependent on a written gospel (not an oral tradition), almost certainly Matthew, and that he regarded that gospel as Scripture. Since the letter mentions the temple laying in ruins (Barn. 16.3-4), it must date between 70-135 AD. The expectation that the temple may be rebuilt could reflect the circumstances leading up to the Jewish revolt (132-135);[7] Robinson dates it much earlier (c.75).[8] If we assign the Epistle of Barnabus to the range 70-135 and Matthew to the range 60-80 (following majority opinion) then the gospel was acknowledged as Scripture within 75 years of its composition.

(D) Polycarp to Philippians

For I trust that ye are well versed in the Sacred Scriptures, and that nothing is hid from you; but to me this privilege is not yet granted. It is declared then in these Scriptures [scripturis], “Be ye angry, and sin not,” and, “Let not the sun go down upon your wrath. Pol. Phil 12.1

Polycarp quotes Eph 4:26 as Scripture. Polycarp wrote to the Philippians responding to their request for copies of the letters of Ignatius and so this is a covering letter for the collection he sends. This letter is therefore contemporary with the Ignatian letters (c.110). It has sometimes been proposed that the extant text is compilation of two letters, the first twelve chapters coming from second letter of Polycarp written in the mid-130s.[9] Ephesians, though sometimes considered inauthentic, may be reasonably dated c.60. It was acknowledged as scripture by Polycarp within 75 years of composition.

(E) 2 Clement

And another Scripture [gk]γραφή[/gk] saith, “I came not to call the righteous, but sinners”. 2 Clem 2.4; cf. Matt 9:13; Mark 2:17

The homily, known as Clement’s second letter to the Corinthians, is neither a letter nor can it be credibly assigned to Clement.[10] It cites the words of Jesus as Scripture, though it could derive from any of the synoptic gospels (cf. Matt 9:13; Mark 2:17; Luke 5:32). It can be reasonably dated to the 140s.[11]

Other NT Scriptural Quotations

Though used frequently in the NT, not all scriptural quotations were introduced with the formula “it is written” or “the Scripture says”. Sometimes quotations are introduced with a simple “for” (e.g. Rom 10:18) or “again” (Rom 15:11); others form part of a narrative (e.g. Luke 4:17; Acts 8:32). We may diagnose several other ways of introducing scriptural quotations. “God says” (e.g. 2 Cor 6:16) and “God through x says” (e.g. Rom 9:25) denote scriptural quotations because Scripture is regarded as the word of God. Similarly, the introduction “the Holy Spirit says” (e.g. Heb 3:7) and “the Holy Spirit through x says” (e.g. Matt 22:43) also denote scriptural quotations because Scripture is regarded as given through the Spirit. Lastly, quotations introduced as the words of a prophet (e.g. Acts 7:48), or a prophecy (e.g. Matt 13:14), would normally be taken to scriptural because prophets spoke by the Spirit (2 Pet 1:20), excepting those cases where a false prophet is quoted (cf. Tit 1:12).

We can contrast these indicators of scriptural quotations with cases of non-scriptural quotations. We have three examples in the NT of quotations from sources that were certainly not considered to be Scripture. These are the quotations from Epimenides and Aratus that Paul used in his speech on Mars Hill (Acts 17:28) and the quotation from Epimenides about Cretans in Paul’s letter to Titus (Tit 1:12). The former quotations are ascribed to “some of your own poets”, the latter to “a prophet of their own”; in both cases something about the source of the quotation is stated which indicates that it is not to be taken as Scripture.

Other Early Christian NT Quotations

Using the indicators of scriptural quotations discussed above, we can diagnose some more cases of early Christian writers citing the NT as Scripture. In some cases, the manner in which the quotation is introduced leaves it uncertain whether it is considered to be a scriptural quotation. For example, there are many instances of quotations introduced as the words of Jesus (1 Clem 13:2, 46:8; 2 Clem 3:2, 4:2, 5:2, 4, 6:1, 9:11; Ign. Smyrn. 3.2; Pol. Phil 2.3, 7.2; Did. 9.5). These quotations are clearly treated with reverence as sayings of the Lord but that, of itself, does not mean that the source of the quotation was treated as Scripture. For example, Acts 20:35 quotes words of Jesus which are not known from any gospel; Paul considers them authentic words of Jesus but if there was a written source then it presumably never received scriptural status. Arguably, on the basis of these quotations alone these early Christians might have regarded the words of Jesus as authoritative and the gospel records as merely reliable history. However there are some other examples which are stronger indicators of scriptural status.

(A) 1 Clement

Take up the epistle of the blessed Apostle Paul. What did he write to you at the time when the Gospel first began to be preached? Truly, under the inspiration of the Spirit, he wrote to you concerning himself, and Cephas, and Apollos, because even then parties had been formed among you. 1 Clem 47.1-3

Clement writing to the church at Corinth reminds them of a letter written to them by Paul “under the inspiration of the Spirit”. By definition, acknowledging 1 Corinthians as inspired by the Spirit is recognition of 1 Corinthians as Scripture. 1 Clement has been traditionally dated to the mid-90s,[12] though it has been dated earlier (c.70). [13] 1 Corinthians is usually dated to c.55.

(B) 2 Clement

For, on the one hand, they hear from us that God has said, “it is no great accomplishment for you to love those who love you; it is great if you love your enemies and those who hate you. 2 Clem 13.4

Here the writer quotes words of Jesus from Luke 6:32, 35 as though they were words of God. It is vastly improbable that the writer means to identify Jesus as “God”; neither Ignatius (c.110) nor Justin (c.150-160) call Jesus “God” in this unqualified way. It is more probable that the writer is referring to Scripture.

(C) Didache

In the Didache, we find a NT quotation introduced as “the Lord says in the gospel” (Did 8.2; cf. 2 Clem 8.5). If the word ‘gospel’ means ‘recognised account of Jesus’, then this might be an indicator of Scripture. However it is, perhaps, more probable that ‘gospel’ means a proclamation of the good news.

(D) Diognetus

In the Epistle to Diognetus, the author quotes 1 Cor 8:1 as the words of “the apostle” (Diogn. 12.5). It is possible that introducing a quotation as having apostolic authority was equivalent to introducing a quotation as having prophetic authority, and thus an indicator of scriptural status. However, this example is complicated by the fact that it is generally recognised that chapters 11 and 12 do not belong to the Epistle of Diognetus but were wrongly combined at some later date.[14]

Summary

The table below presents the evidence considered above along with the possible range for the period between composition and recognition as Scripture.

NT Text	Cited	Years to Recognition
Luke	1 Tim 5:18	0-40
Matthew	Barn. 4.14	0-75
Paul’s letters, (some of)	2 Pet 3:15-16	10-50
1 Corinthians	1 Clem 47.1-3	15-40
Ephesians	Pol. Phil 12.1	50-75
Luke	2 Clem 13.4	60-80
Matthew or Mark or Luke	2 Clem 2.4	60-90

This analysis is necessarily impeded by the paucity of the evidence, and so cannot be taken to represent the full extent of scriptural recognition. It does demonstrate that NT texts were recognised as Scripture very early; Luke, Matthew and some of Paul’s letters were acknowledged as Scripture within a generation.

But this analysis almost certainly under-represents what was regarded as Scripture. J. Barton notes that the apostolic fathers quote the NT far more than the OT.[15] He suggests that this demonstrates the inadequacy of judging scriptural status through quotations introduced as Scripture. Clearly the NT was considered authoritative by the early Christians even when not introduced as Scripture.

There are limitations to this analysis. Whilst it demonstrates what was considered Scripture, it gives us no indication what was not considered Scripture; there is no early canon to which we can refer. Barton describes three categories based upon the available evidence: a core of frequently cited texts (4 gospels and major Pauline letters); an intermediate class of infrequently cited texts (Acts, shorter epistles and Revelation); and a third class of texts occasionally cited (e.g. Shepherd of Hermas). This distribution suggests that a significant part of our NT was treated as authoritative by subsequent generations of Christians, but also suggests some fluidity around the edges.[16]

---