1

1he New 1estament in the Second Century:

1ext, Collections and Canon
1

L. \. lurtado, Uniersity o Ldinburgh


It has been clear or some time that the second century was a ,indeed, perhaps tbe, crucial period in
the deelopment o the New 1estament. 1he indiidual writings that comprise the New 1estament
were all, or nearly all, written across the latter hal o the irst century, but the second century is the
period in which most o them came to hold a special signiicance, at least or the great majority o
Christian circles. lacilitating and relecting this, the second century was the time when these writings
were copied and disseminated widely. It is all the more rustrating, howeer, that the extant eidence
rom the period is almost in inerse relation to its importance. Neertheless, this importance
demands and justiies our eorts to take stock o what we can say, and with what conidence, about
the New 1estament writings in the second century.
In what ollows, I make a modest eort toward this end by underscoring three crucial
processes in this period, which also constitute three major areas o scholarly inquiry and controersy:
,1, the textual transmission o the New 1estament writings, ,2, the phenomenon o early collections
o writings ,especially the Gospels and Pauline epistles,, and ,3, certain writings coming to enjoy a
special status, authority, and usage, which seems to be the crucial earlier stage o a process that led
eentually to a ixed, closed canon o the New 1estament. I shall surey these matters in the light o
current scholarly debates and recently aailable eidence ,e.g., the most recent Oxyrhynchus
olumes,. My aim here is not to proide some deinitie treatment o any o the data or the issues,

1
It is a personal pleasure to oer this essay to Carroll Osburn, with great appreciation or his own scholarly
contributions and his collegial manner. An earlier ersion o this essay was gien as an inited presentation in
the New 1estament 1extual Criticism program unit o the annual meeting o the Society o Biblical Literature,
1-20 Noember 2001, Dener, USA. I particularly acknowledge critical interaction with \illiam L. Petersen
then and subsequently in clariying the discussion o patristic citation o N1 writings, though I must take sole
responsibility or the iews I oer here. James Kelhoer also kindly read an earlier drat and gae helpul
suggestions. |1he published ersion appeared in 1rav.vi..iov ava Recetiov: ^er 1e.etavevt 1etCriticat ava
2
but rather to emphasize the importance o these three processes or dynamics, thereby to help to
ocus urther thinking about them.
1extual 1ransmission of the New 1estament Writings
lirst, what can we say about the transmission o the text o the N1 in the second century Some
scholars emphasize and allege great luidity and reedom in this period, in some cases so much that
the extant manuscripts are alleged to be seriously unreliable or reconstructing the original` o N1
writings, whereas others contend that the manuscript eidence shows suicient useulness to
encourage this text-critical eort.
2
Some see the undeniable textual luidity as indicatie that the
writings held something considerably less than scriptural signiicance, whereas others argue that it
shows the opposite.
As is reasonably well known, the two main types o eidence that hae been used in orming
our iews o the transmission o New 1estament writings in the second century are, irst, the extant
manuscripts rom that time and early centuries thereater, and, second, the citations,quotations o
New 1estament writings by second-century Christian authors.
3
Let us look briely at the latest
deelopments in these bodies o eidence.
As or the manuscripts o New 1estament writings, there is both bad news and good news.
1he well-known bad news is that the extant manuscripts that can plausibly be dated to the second

egeticat tvaie., eds. J. \. Childers & D. C. Parker ,1exts and Studies, 1hird Series,,` 4, Piscataway, NJ:
Gorgias Press, 2006,, 3-2.
2
lelmut Koester, 1he 1ext o the Synoptic Gospels in the Second Century,` in Co.et 1raaitiov. iv tbe ecova
Cevtvr,: Origiv., Recev.iov., 1et, ava 1rav.vi..iov, ed. \illiam L. Petersen ,Notre Dame,London: Uniersity o
Notre Dame Press, 1989,, 19-3, contends that the second century was completely a period o wild ariation
until sometime near 200 CL when he proposes that a textual recension was undertaken, rom which our extant
N1 manuscripts all derie. Note also the recent problematizing o the task o reconstructing an original` text
o the New 1estament writings: Lldon Jay Lpp, 1he Multialence o the 1erm Original 1ext` in New
1estament 1extual Criticism,` 1R 92 ,1999,: 245-81.
3
See now esp. Barbara Aland, Die Rezeption des neutestamentlichen 1extes in den ersten Jahrhunderten,` in
1be ^er 1e.tavevt iv art, Cbri.tiavit,, ed. Jean-Marie Serin ,BL1L 86, Leuen: Peeters, 1989,, 1-38. Stuart R.
Pickering has complained about inadequate attention gien to the potential importance o quotations and
allusions to passages in N1 writings in early papyri: 1he Signiicance o Non-Continuous New 1estament
Materials in Papyri,` tvaie. iv tbe art, 1et of tbe Co.et. ava .ct., 1be Paer. of tbe ir.t irvivgbav Cottoqvivv ov
3
century are lamentably ew in number, and none o them gies us a complete text o any New
1estament writing. In act, the extant second-century manuscript eidence consists largely in a
handul o incomplete single leaes, though they collectiely derie rom a number o manuscripts.
Len i we accept Skeat`s argument that P64, P6 and P4 all represent the same multi-gospel
manuscript rom the late second century, the amount o text presered in the total body o second-
century manuscript material is still rustratingly small.
4
1he earliest manuscripts that gie us
substantial portions o texts are dated palaeographically to the early third century or thereabouts.
P45 ,Gospels and Acts, and P46 ,Pauline epistles, date rom ca. 200-250 CL, Gospels codices P66
and P5 rom ca. 200 CL, P4 ,Reelation, ca. 250-300 CL, and P2 ,Jude and 1-2 Peter, third to
ourth century CL.
Newly Published Manuscripts
On the other hand, the good news is that the small und o second-century and,or early
third-century manuscript witnesses has been enriched with the publication o three recent olumes o
the Oxyrhynchus papyri. Volumes 64-66 gie us preiously unknown New 1estament papyrus
materials that comprise leaes rom seen manuscripts o Matthew, our o John, two o Reelation,
and one each o Luke, Acts, Romans, lebrews, and James, the dates ranging rom the second
century to the ith or sixth century CL.
5
1he earliest are leaes o three manuscripts o Matthew

tbe 1etvat Critici.v of tbe ^er 1e.tavevt, ed. D. G. K. 1aylor ,Birmingham: Uniersity o Birmingham Press,
Atlanta: Society o Biblical Literature, 1999,, 121-41.
4
1. C. Skeat, 1he Oldest Manuscript o the lour Gospels` ^1 43 ,199,: 1-34, reprinted in J. K. Llliott
,ed.,, 1be Cottectea ibticat !ritivg. of 1. C. /eat ,No1Sup 113, Leiden: Brill, 2004,, 158-92, also G. N. Stanton,
1he lourold Gospel,` ^1 43 ,199,: 31-46, reprinted ,with light reision, in G. N. Stanton, ]e.v. ava Co.et
,Cambridge: Cambridge Uniersity Press, 2004, 63-91.
5
1be O,rb,vcbv. Pa,ri ,London: British Academy or the Lgypt Lxploration Society,: 1otvve `1, ed. L.
\. landley et at., 199, has POxy 4401-6 ,pp. 1-13, ed. J. Daid 1homas,, 1otvve `1, ed. M. \. laslam et
at., 1998, has POxy 4445-48 ,pp. 10-20, ed. \. L. l. Cockle,, and POXy 4449 ,pp. 20-25, ed. R. lbner,,
1otvve `1, ed. N. Gonis et al., 1999, has POxy 4494-95 ,pp. 1-5, and POxy 449-98 ,pp. -10, ed. \. L. l.
Cockle,, POxy 4496 ,pp. -10, ed. 1im linney,, POxy 4499 ,pp. 10-35, ed. J. Chapa,, and POxy 5000 ,ed. \.
L. l. Cockle,. 1hese comprise a new portion o P ,P.Oxy 4405 part o the same codex as P.Oxy 2683,, and
seeral newly identiied manuscripts assigned N1 papyri numbers P100-115. Basic inormation and images on
the Oxyrhynchus web site: http:,,www.papyrology.ox.ac.uk,. See especially the aluable discussion by Peter
M. lead, Some Recently Published N1 Papyri rom Oxyrhynchus: An Oeriew and Preliminary
Assessment,` 1,vvt 51,2000, 1-16, which includes ample citation o other releant publications, and J. K.
4
dated to the second or early third century: P.Oxy. 4405 ,a new portion o P, containing Matt
23:30-34, 35-39, second,third century,, P.Oxy. 4403 ,P103, Matt 13:55-56, 14:3-5, second,third
century,, POxy 4404 ,P104, Matt 21:34-3, 43 &45, late second century,. Prior to the publication o
these ragments, per Nestle-Aland 2, the only second-century manuscripts aailable were the
amous P52 ,P.Ryl 45, John 18:31-33, 3-38,, P90 ,P.Oxy 3523, John 18:36-19:1, 19: 2-,, and P98
,P.IlAO 23b, Re 1:13-20,. Len i we add in the recently posited manuscript combination o P4-
P64-P6 mentioned already, and grant the proposal that the manuscript dates rom the late second
century, and i we also add New 1estament papyri usually dated ca. 200 CL, such as P66, P5, P46, it
is still clear that the ery recent Oxyrhynchus ragments add signiicantly to a ery limited body o
manuscript material or the second century.
Moreoer it is urther good news that, although comprising a small amount o the tet o
New 1estament writings, these ragments are actually rich with data. lrom my own ,thus ar limited,
consultation o the releant Oxyrhynchus olumes and rom Peter lead`s aluable surey o these
ragments, I mention a ew illustratie matters. lirst, though we hae in each case only a small
sample o the manuscript rom which they come, in general these ragments conirm the text o the
great uncials which orms the basis o the modern critical editions.`
6
In the main, they proide us
with earlier attestation o ariants that we already knew o rom later witnesses. In some cases,
howeer, these are ariants preiously attested only in the ersions, which warns us that in a good
many other cases as well readings presently supported only in the ersions may well relect ery early
readings that simply happen not to hae suried in the extant Greek witnesses.

Llliott, lie New Papyri o the New 1estament,` ^or1 41,1999, 209-13, reiews the New 1estament
ragments in Volume 65 o the Oxyrhynchus series, and ocuses almost entirely on what readings they contain.
6
lead, Some Recent Published N1 Papyri rom Oxyrhynchus,` 16.

Perhaps the most signiicant ariant is in P.Oxy 4445 ,P106, at John 1:34, o eklektoj where most witnesses,
including P66 and P5 read o uioj. P106 here gies early support or a reading ound also in Sinaiticus ,original
hand,, later minuscules ,, 218,, Old Latin manuscripts ,b, e,
2
,, and the Old Syriac ,syr
s, c
,. See lead,
Some Recently Published N1 Papyri,` 11.
5
But the larger point is that these ragments urther encourage us to think that the more
substantial witnesses rom the third century and later are ,covtra Koester, probably not the results o
some supposed major recension o New 1estament writings initiated toward the end o the second
century.
8
Instead, the Oxyrhynchus ragments urther justiy the iew that the more substantial early
third-century papyri are reliable witnesses o the text o the writings that they contain, as these
writings had been transmitted across the second century.
Second, these ragments also reinorce the impression gien by the New 1estament papyri
rom 200 CL and a bit later that there were rar,ivg scribal tendencies operatie in the textual
transmission o the New 1estament in the second century.
9
1hat is, the recently-published eidence
is consistent with the iew that the second century was a time o somewhat dierse textual dynamics.
1o quote rom lead`s surey, the ragments illustrate arious points along the spectrum rom more
controlled texts ,with corrections, literary eatures, etc., to comparatiely more ree or careless
copying.`
10
\e are thereby urther warned against oer-simpliications about the textual
transmission o New 1estament writings in the second century. Instead, with enhanced conidence
we may take up Lpp`s proposal that the early New 1estament papyri can be placed in seeral early
clusters` or textual groups,` and that these represent dierent textual complexions` already
operatie in the second century. Some o the newly published ragments relect a concern or a
high degree o accuracy,` and others indicate a reer readiness to adapt the text, exhibited especially

8
lelmut Koester, 1he 1ext o the Synoptic Gospels in the Second Century,` in Co.et 1raaitiov. iv tbe ecova
Cevtvr,: Origiv., Recev.iov., 1et, ava 1rav.vi..iov, ed. \illiam L. Petersen ,Notre Dame,London: Uniersity o
Notre Dame Press, 1989,, 19-3.

9
See esp. James R. Royse, Scribal labits in Larly Greek New 1estament Papyri,` ,1hD diss, Graduate
1heological Union, Berkeley, 1981,, ia., Scribal 1endencies in the 1ransmission o the 1ext o the New
1estament,` in 1be 1et of tbe ^er 1e.tavevt iv Covtevorar, Re.earcb, eds. Bart D. Lhrman, Michael \. lolmes
,SD46, Grand Rapids: Lerdmans, 1995,, 239-52.
10
lead, Some Recently Published N1 Papyri,` 10.
6
in stylistic changes, harmonizations, higher numbers o accidental changes, and een occasional
changes motiated by doctrinal concerns.
11

1o aoid misunderstanding in this controersial matter, I emphasize my point. I do not
deny at all that there was ,perhaps considerable, luidity in the transmission o the New 1estament
writings in the second century.
12
I simply stress that along with a readiness o some ,perhaps een
most, scribes to introduce ariants intended to harmonize the Gospels, remoe ambiguities, airm
doctrinal concerns, and een introduce new material intended as ediying, in at least some circles
there also appears to hae been a somewhat more conseratie copying attitude.
In addition to their readings, howeer, the small but ascinating body o early papyri gies us
other aluable eidence that should not be oerlooked. New 1estament scholars, including text
critics, hae tended to comb early manuscripts or readings, but we also must learn to harest the
uller and more dierse data that lie in these aluable arteacts.
13
lor example, the correctiov. in P.Oxy
4403 ,P103, and P.Oxy 4405 ,P, are noteworthy. 1he quality o the hands suggests that these
manuscripts were not produced by proessional calligraphers such as those who produced expensie
copies o literary texts. Neertheless, along with some other eatures, these corrections relect the
sort o vevtatit, ,though not the ully deeloped scribal skills, that we associate with a scriptorium. In
particular, the corrections show a concern or what those correcting the copies regarded as accurate
copying. O course, we must be careul to aoid anachronism in positing too conidently ormal

11
L. J. Lpp, 1he Signiicance o the Papyri or Determining the Nature o the New 1estament 1ext in the
Second Century: A Dynamic View o 1extual 1ransmission,` in L. J. Lpp, Gordon D. lee, tvaie. iv tbe 1beor,
ava Metboa of ^er 1e.tavevt 1etvat Critici.v ,SD 45, Grand Rapids: Lerdmans, 1993,, 24-9. See the
somewhat similar categories proposed by Aland, Die Rezeption des neutestamentliches 1extes,` 26-2.
12
The so-called Western text is perhaps the most striking expression of the fluidity in textual
transmission in this early period. But it appears that the Western text is probably more a body of
readings produced by somewhat similar scribal tendencies, rather than a cohesive recension. For a recent
study, see, W. A. Strange, The Problem of the Text of Acts (SNTSMS 71; Cambridge: Cambridge
University Press, 1992).
13
Gnther Zuntz, 1be 1et of tbe i.tte.: . Di.qvi.itiov vov tbe Corv. Pavtivvv ,London: Oxord Uniersity
Press or the British Academy, 1953,, is a classic study o P46 that demonstrates a uller use o the data
aailable in the early papyri. Among studies o the major codices, D. C. Parker, Coae eae: .v art, Cbri.tiav
Mavv.crit ava t. 1et ,Cambridge: Cambridge Uniersity Press, 1992,, is a model.
7
scriptoria early in the second century, at least in the sense o the sort o physical settings in which
multiple copies o Christian writings were prepared in later centuries.
14
Neertheless, there are
arious indications that the copying o early Christian texts in the second century inoled emergent
scribal conentions that quickly obtained impressie inluence, and, at least in some cases and
settings, that there was a concern or careul copying.
15

1o cite one particular matter, where these new ragments presere the words known to us as
voviva .acra, particularly the our words Oto,, Kupio,, Xpioo,, and Iqoou,, these words are
written in the sorts o abbreiated orms that we already know rom other ancient Christian
manuscripts.
16
1he minor ariations in the precise spelling o the abbreiations do not rightly count
against the conclusion that there was a widely attested conention among Christian scribes that
certain religiously loaded` words were to be written in a distinctie manner.
1
In so ar as earliest
Christian manuscripts were not copied by proessional` scribes ,or at least oten do not exhibit the
kind o calligraphy more characteristic o proessionally produced literary manuscripts o the period,,
such widespread and distinctie scribal conentions are all the more notable.
It is also signiicant that all o these ragments come rom codices. 1hus, they collectiely
reinorce the conclusions drawn rom preiously known eidence that, by sometime early in the
second century at the latest, Christians oerwhelmingly had come to preer the codex, especially, it

14
C., e.g, larry \. Gamble, oo/. ava Reaaer. iv tbe art, Cbvrcb ,New laen,London: \ale Uniersity Press,
1995,, 120-23, Kim laines-Litzen, Cvaraiav. of etter.: iterac,, Porer, ava tbe 1rav.vitter. of art, Cbri.tiav
iteratvre ,Oxord,New \ork: Oxord Uniersity Press, 2000,, 83-91. Part o the diiculty in the issue is that
scholars are not always clear as to what they mean by scriptorium.`
15
See, e.g., Zuntz, esp. 263-83.
16
Unortunately, the ragments dated to the second century ,P.Oxy 4405,P, P.Oxy 4403,P.103, and P.Oxy
4404,P104, do not presere portions o text where the words in question would hae occurred. But newly
published third-century ragments do: P.Oxy 4449 ,P100,, P.Oxy 4401 ,P101,, P.Oxy 4445 ,P 106,, P.Oxy
444 ,P108,, P.Oxy 4495 ,P111,, P.Oxy 449 ,P113,, P.Oxy 4498 ,P114,, P.Oxy 4499 ,P115,. See the table o
eatures in lead, Some Recently Published N1 Papyri,` 5. On the nomina sacra, see esp. Colin l. Roberts,
Mavv.crit, ociet, ava etief iv art, Cbri.tiav g,t, 1be cbreicb ectvre. 1 ,London: Oxord Uniersity Press
or the British Academy, 199,, 26-48, and now Larry \. lurtado, 1he Origin o the Nomina Sacra: A
Proposal,` ] 11 ,1998,: 655-3, or a ull discussion and bibliographic reerences.
1
1hus, I do not ind laines-Litzen`s eort to minimize the signiicance o the voviva .acra persuasie: c.
Cvaraiav. of etter., 91-96.
8
appears, or their scriptures ,Old 1estament, and the Christian writings that were coming to be
treated widely as scripture.
18
1he only extant examples o Christian texts written on unused rolls ,as
distinguished rom re-used rolls, opisthographs`, are theological tractates ,e.g., Irenaeus` .gaiv.t
ere.ie.,, and writings that may hae been regarded as ediying in some circles but did not gain
acceptance as part o the emergent New 1estament canon.
19

1hese new Oxyrhynchus ragments o New 1estament writings also exhibit arious aids to
reading, such as rough-breathing marks, punctuation, and, a matter o particular signiicance,
occasional spacing at the ends o sentences and perhaps paragraphs.
20
1hese readers-aids are ery
unusual or literary texts o the period, but there are some similarities to pre-Christian Jewish
manuscripts o the Old 1estament writings ,e.g., P.Ryl 458,. 1he most cogent inerence is that the
Christian manuscripts with these arious scribal deices were prepared or ease o vbtic reading in
churches. 1hat is, these small ragments probably gie us urther important arteactual eidence
conirming second-century reports ,e.g., Justin Martyr, o the liturgical practice o reading these New
1estament writings.
21
1hough inadequately noticed, such eidence was already proided in
preiously aailable ragments such as the amous ragment o the Gospel o John, P52 ,P.Ryl 45,,

18
C. P.Oxy 4443, a ragment o Greek Lsther dated to the late irst or early second century C.L., a roll ,a
luxurious copy`,, and with at least one occurrence o an uncontracted qeoj ,plus two other cases proposed or
lacunae,, which rightly led the editors to assign the manuscript to a Jewish proenance ,1be O,rb,vcbv. Pa,ri,
1otvve `1, ed. M. \. laslam, 4,. C. also, e.g., the Oxyrhynchus manuscripts o Co.et of 1bova., one o
which is a codex ,P.Oxy 1,, and the other two ,P.Oxy 654 and 655, rolls, one o these ,P.Oxy. 654, an
opisthograph written on the reerse side o a land register.
19
L.g., P.Oxy 3405 ,an laelst 61, is a late second or early third-century roll o Irenaeus, .gaiv.t ere.ie., as is
the ourth-century copy, an laelst 62. 1he layoum ragment ,P.Vindob. G. 2325, an laelst 589, is a roll.
P.Oxy 654 ,an laelst 593, is an opisthograph, but P.Oxy 655 ,an laelst 595, is another copy o Co.et of
1bova. written on a resh roll. P.Mich 130 ,an laelst 65, is a late second-century opisthograph o bebera of
erva., whereas P.Berlin in. 5513 ,an laelst 662, third-century, is another copy o erva. on a roll.
20
Again, I reer to the table in lead, Some Recently Published N1 Papyri,` 5. lurther details are gien in the
releant Oxyrhynchus olumes. See also Lldon Jay Lpp, 1he New 1estament Papyri at Oxyrhynchus in their
Social and Intellectual Context,` in a,ivg. of ]e.v.: cavovicat ava ^ovCavovicat, ..a,. iv ovovr of 1;ite aaraa,
ed. \illiam L. Petersen, Johan S. Vos, lenk J. de Jonge ,No1Sup 89, Leiden: Brill, 199,, 4-68 ,and Lpp`s
Appended Note` which appears in oprints o his essay and takes account o Oxyrhynchus olumes that
appeared ater the essay was written,.
21
See, e.g., Gamble, oo/. ava Reaaer., 205-8, 211-31.
9
which exhibits diaresis and curious spaces that seem to register clauses, places where the public`
readers were probably intended to make small pauses.
22

It is an unortunate weakness in Kim laines-Litzen`s recent ,and in a number o other
matters, ery helpul, study o early Christian scribal practice that she rather too simply assumes the
general literary practice o making priate copies or personal usage as the operatie setting and
model or the production o all early Christian manuscripts. I think that she gies inadequate
attention to strong indications that a good many Christian manuscripts were prepared or grov. and
or reading out as liturgical texts.
23
In the ollowing paragraphs, I cite briely the important matters.
Already by the date o our earliest extant eidence, Christians had come to preer a
distinguishable book-orm ,the codex, oer against the wider preerence o their culture or the
book-roll. 1he Christian preerence or the codex seems to hae been especially strong in copying
their most reered writings, those that were regarded as scripture and,or were coming to be so
regarded. As we hae noted, Christians also deeloped distinctie scribal practices, among which the
voviva .acra are the most striking, but including also the richer use o punctuation and spacing. 1hey
read texts, not simply priately or in the sort o reading circles o the cultured elite, but also, ery
importantly and characteristically, as a regular part o their liturgical practice and thus as a eature o
their gathered worship. In this they diered rom the literary and religious practices o the larger
culture. Reading texts does not typically eature in cultic practices,settings, and, in any case, literary
texts did not get this kind o usage. 1he only precedent and analogy or the early Christian religious
usage o texts was in the reading o scripture as part o Jewish synagogue practice. All this
cumulatiely signals what must be seen as the emergence o an identiiably and somewhat distinctie

22
Diaresis in recto lines 1 and 2, and erso line 2, spacing recto lines 2 and 3, and erso line 2. C. l. Roberts
commented on the spacing he ound in P.Ryl 458 ,Greek Deuteronomy, second century BCL, at the ends o
sentences or clauses and groups o words, noting how unusual such spacing was, and also that a similar system
might be identiied in P52 ,1wo Biblical Papyri in the John Rylands Library Manchester,` ]R 20|1936| 219-
36, esp. 226-2,.
23
laines-Litzen, Cvaraiav. of etter.. lor a broader and more ully nuanced treatment o the production and
use o early Christian texts, see esp. Gamble, oo/. ava Reaaer. iv tbe art, Cbvrcb.
10
Christian literary ethos. Indeed, in an essay in the recent Peter Richardson e.t.cbrift I hae proposed
that the early Christian manuscripts oer us our earliest indications o an emergent Christian
material culture.`
24

1he reason I underscore these matters is that there is increasing recognition that the
repeated liturgical reading o New 1estament writings is an important actor iv tbe tetvat trav.vi..iov
o these texts. It certainly helps to account or the obiously requent copying and wide
dissemination o these writings, which goes ar beyond anything else in antiquity. lurthermore,
liturgical usage is one o the actors that would hae helped to prompt the sort o small stylistic
improements` intended to make texts clearer and easier to understand that are so well known in
Christian manuscripts. 1he regular liturgical reading o the our canonical Gospels also helps to
account or the abundance o harmonizing ariants, especially requent in Mark.
But repeated public reading o New 1estament writings would also hae set real limits on
how much a writing could be changed, at least in a gien circle, without people noticing ,and
probably objecting,, as anyone amiliar with what happens when liturgical changes are introduced can
attest. It is, thus, likely not a coincidence that Mark, which appears to hae been the least widely and
requently used in liturgical reading, also exhibits the largest number and the most salient ariations
,especially, o course, the seeral endings,. By contrast, the most widely used Gospel, Matthew, has
probably the most stable and ixed text.
1hat is, the practice o reeatea liturgical reading o New 1estament writings is yet another
actor that ought to lead us to hesitate to characterize the second century as basically a period o
wild` textual tendencies. Along with the surprisingly well-attested preerence or the codex and the
ubiquitous scribal treatment o the voviva .acra, the practice o liturgical reading o writings proides

24
Larry \. lurtado, 1he Larliest Lidence o an Lmerging Christian Material and Visual Culture: 1he
Codex, the ^oviva acra and the Staurogram,` in 1et ava .rtifact iv tbe Retigiov. of Meaiterraveav .vtiqvit,: ..a,.
iv ovovr of Peter Ricbara.ov, eds. Stephen G. \ilson and Michel Desjardins ,\aterloo, Ontario: \ilrid Laurier
Uniersity Press, 2000,, 21-88.
11
us with indications o conentionalization o practice with regard to these writings at a chronological
stage o early Christianity to which we are otherwise accustomed to attributing great diersity.
It is certainly the case, howeer, that we hae not adequately mined all that is proided to us
in the early papyri, whether those that hae been known or some time or those newly published.
1here are some aluable research projects here. lor example, larry Sanders noticed long ago that
Codex \ ,the our Gospels, exhibited a system o spacing o sense-units that corresponded with
ersional eidence, and proposed that this might relect an ancient system o phrasing, used in
reading the Scriptures in church serice,` whose origin must hae been as early as the second
century.`
25
1he spacing ound in early papyri that hae appeared subsequently seems to support
Sanders` suggestion. Already in the second century there appears to hae been an embryonic system
o subdiision o the texts o the gospels that probably relected and supported the practice o gospel
readings as part o Christian worship gatherings.
26
But it remains or us to mine the releant material
on this and other intriguing matters. Prospectie doctoral students, take note!
2

Second-Century Citations
1he other major body o data that has oten been taken as giing eidence about the transmission o
the New 1estament writings in the second century are the citations,quotations in second-century
Christian writers.
28
In ery basic terms, examination o second-century Christian writers indicates

25
l. A. Sanders, 1be ^er 1e.tavevt Mavv.crit. iv tbe reer Cottectiov, Part , 1be !a.bivgtov Mavv.crit of tbe ovr
Co.et. ,New \ork: Macmillan, 1912,, 14.
26
See urther Victor Martin`s discussion o spacing signalling subdiisions in P66: Pa,rv. oaver , ravgite ae
]eav cba. 111 ,Cologny-Genee: Bibliotheque Bodmer, 1956,, 18-21.
2
Note the new series, Pericope: Scripture as \ritten and Read in Antiquity,` established in 2000, and
published by Van Gorcum ,Assen, Netherlands, under the editorship o Marjo Korpel ,Utrecht, et alia, which
ocuses on scribal unit delimitation` in biblical manuscripts ,lebrew, Greek, Syriac, Latin,.
28
Studies include Donald A. lagner, 1he Sayings o Jesus in the Apostolic lathers and Justin Martyr,` in
Co.et Per.ectire.: 1be ]e.v. 1raaitiov Ovt.iae tbe Co.et., ed. Daid \enham ,Sheield: JSO1 Press, 1984,, 233-
68, A Committee o the Oxord Society o listorical 1heology, 1be ^er 1e.tavevt iv tbe .o.totic atber.
,Oxord: Clarendon Press, 1905,, lelmut Koster, ,voti.cbe |bertiefervvg bei aev ao.toti.cbev 1aterv ,1U 65,
Berlin: Akademie-Verlag, 195,, Ldouard Massaux, 1be vftvevce of tbe Co.et of t. Mattber ov Cbri.tiav iteratvre
efore t. revaev.. Part 1. 1be ir.t ccte.ia.ticat !riter. ,Macon, GA: Mercer Uniersity Press, 1990,, Arthur J.
12
ew explicit citations o New 1estament writings, and, where it is clear or highly likely that a New
1estament writing is quoted, the quotation oten exhibits curious dierences rom the text o the
writing that is dominant in the extant manuscripts.
1o be a bit more precise, these things tend to be comparatiely more characteristic o
Christian writers o the irst hal o the second century. Noting this, Barbara Aland has proposed
that in the latter hal o the second century we see the emergence o a text-consciousness` that is
relected in a more requently explicit ,named, and exact citation o New 1estament writings. 1his
increased text-consciousness,` she proposes urther, may hae deeloped as a result o two major
processes: ,1, second-century controersies oer Christian aith that inoled questions about the
wording o texts ,e.g., Marcion,, and ,2, the prolonged eect o repeated liturgical reading o certain
texts.
29

As noted earlier, Koester, and \illiam Petersen also, in particular hae argued that the loose
and luid wording o the quotations o New 1estament writings in early second-century authors
means that the text o New 1estament writings was then considerably more luid than is relected in
the extant manuscripts.
30
Indeed, they hae contended that the eidence o the early citations should
be preerred oer the extant manuscripts o the New 1estament writings in characterizing their
textual transmission in the second century. But, as Barbara Aland complained in her 1989 essay, the
analysis o second-century Christian citations has tended too much to proceed with insuicient
attention to the wider literary practices o the time.
31
1hat is, the import o the citation practices

Bellinzoni, 1he Gospel o Matthew in the Second Century,` ecCevt 9 ,1992,: 19-258, ia., 1be a,ivg. of ]e.v. iv
tbe !ritivg. of ]v.tiv Mart,r ,No1Sup 1, Leiden: L. J. Brill, 196,.
29
Aland, Die Rezeption des neutestamentlichen 1extes,` 5-21.
30
Koester, 1he 1ext o the Gospels in the Second Century`, \illiam L. Petersen, \hat 1ext Can New
1estament 1extual Criticism Ultimately Reach` in ^er 1e.tavevt 1etvat Critici.v, ege.i. ava Cbvrcb i.tor,:
. Di.cv..iov of Metboa., ed. Barbara Aland, J. Delobel ,Kampen: Kok-Pharos, 1994,, 136-52, ia., 1he Genesis
o the Gospels,` in ^er 1e.tavevt 1etvat Critici.v ava ege.i.: e.t.cbrift ]. Detobet, ed. A. Denaux ,Louain:
Peeters, orcoming,.
31
Barbara Aland, Die Rezeption des neutestamentlichen 1extes,` 2-3.
13
relected in early second-century Christian authors may not be as obious or as decisie as has
sometimes been assumed.
\e still do not hae the thorough-going history o citation in antiquity` that Lduard
Norden urged long ago.
32
But we do hae Christopher Stanley`s recent, aluable study o the
citations o Old 1estament writings in Paul, in which Stanley includes a comparatie analysis o the
citation o sources in selected early Roman-era pagan writers and in Jewish writings o the period.
33

Stanley shows that in Jewish, pagan and Christian writers o the time, the citation o known written
sources is impressiely ree and adaptie. \riters omit words, phrases and whole lines that they
deem superluous or problematic or their own rhetorical aims, and they also add or substitute words
and phrases to sere as interpretie renditions` o the material cited, making the material it more
closely with the context o the text in which the citation is appropriated. Likewise, authors requently
combine and conlate material rom dierent contexts o a cited work and,or rom dierent works.
So, in general, the citation practices and techniques that we can obsere in the early second-century
Christian writers are not ery dierent rom the lexible treatment o written sources in the New
1estament and in the broader literary culture o the time.
1hat the wording o these citations is oten not attested in any o the extant copies o the
cited works suggests that authors exercised a certain reedom in amending what they cited. 1he
dierences between citations and the texts o the sources cited oten seem to be, not simply the
products o imprecise memory, but instead deliberate, sometimes artul adaptations. Moreoer, the
conidence with which authors made these adaptations o widely-known sources suggests that they
wrote or readers who accepted such reedom as a legitimate conention in the literary culture o the
time. 1hat is, readers amiliar with the sources being cited would likely hae recognized the

32
Lduard Norden, Die antike Kunstprosa ,reprint, Darmstadt: \issenschatliche Buchgesellschat, 1958,,
1:88-90.
33
Christopher D. Stanley, Pavt ava tbe avgvage of critvre: Citatiov 1ecbviqve iv tbe Pavtive i.tte. ava
Covtevorar, iteratvre ,SN1SMS 4, Cambridge: Cambridge Uniersity Press, 1992,. See esp. Citation
1echnique in Greco-Roman Literature` ,26-91,, and Citation 1echnique in Larly Judaism` ,292-33,.
14
adaptations. 1hey would hae objected to them only i they dissented rom the point being made by
the author doing the citation.
Moreoer, Stanley cogently obsered that it is oer-simpliied to imagine that ancient authors
only either cited a work open beore them or else worked rom memory or oral tradition.`
1here is impressie eidence, particularly the use o combined and conlate citations, to suggest that
authors oten worked rom written compilations o excerpts rom one or more sources, these
compilations probably arranged topically. lurthermore, Stanley proposes that the alterations that we
see in the cited material may oten hae been made at the point o making the excerpt. 1hat is,
authors likely combed releant sources looking or excerpts on some topic and,or in search o
support or some point that they wished to make. And, gien that the literary culture o the day ully
permitted adaptation o cited material, authors would likely hae adapted what they excerpted, at the
point o compiling excerpts, much in the way researchers today oten combine cited material and
their own obserations in their notes.
But was there a ully commensurate reedom in the co,ivg o these writings 1o be sure, the
extant eidence or the writings cited, whether Greek classics, Old 1estament, or New 1estament
writings, indicates sometimes impressie luidity, especially in early stages o the textual transmission
o these writings.
34
But, with a ew notable exceptions, the luidity eidenced in extant manuscripts
does not really match the extent o the ariations that we see in the citations o these works in the
authors o the early Roman period. 1his all means that we should probably think o the copying o
texts and the citation o them as somewhat distinguishable processes with distinguishable sets o

34
Stanley cites Stephanie \est, Ptotevaic Pa,ri of over ,Papyrologica Coloniensis 3, Cologne: \estdeutscher
Verlag, 196, on the highly luid` state o lomeric epics till they were standardized in mid-second century
BCL ,\est, 5-14,. Also Qumran studies, e.g., Shemaryahu 1almon, 1he 1extual Study o the Bible-- A New
Outlook,` Qvvrav ava tbe i.tor, of tbe ibticat 1et, ed. l. M. Cross, S. 1almon, 321-400 ,Cambridge, MA:
larard Uni Press, 195,, Lmanuel 1o, 1etvat Critici.v of tbe ebrer ibte ,Assen: Van Gorcum,
Minneapolis: lortress, 1992,, esp. 155-98, show an interesting diersity in the text o the Old 1estament
writings in the pre-Mishnaic period.
15
conentions. It ollows, thus, that it is dubious to take the orm o citations as direct eidence o the
state o the texts being cited.
1o come to the point releant here, I suggest that it is almost certainly dubious to play o
and priilege citations oer against our early manuscript eidence or the New 1estament writings.
1hough they are rustratingly ragmentary, our earliest manuscripts come rom within decades o the
dates o the early second-century patristic writers ,e.g., Justin,, and our more substantial manuscripts
are roughly contemporary with, or een earlier than, patristic writers o the third century and later.
\e must reckon with all the releant eidence in characterizing the transmission o texts in second-
century Christianity.
A similar cautionary note was sounded seeral decades ago by Bruce Metzger, who adised
that in dealing with patristic citations o the New 1estament that dier rom the textual readings in
extant manuscripts the textual critic must consider whether it was the lather or the scribe o an
early copy o the New 1estament who was more likely to alter the text.`
35
Subsequently, Gordon
lee also demonstrated problems in the use o Patristic citations or recoering the New 1estament
text o their times and locales. lee showed that Patristic authors can cite the New 1estament rather
reely, especially in sermons and related writings, whereas in commentaries they adhere more to the
wording o the manuscripts aailable to them.
36
I, Patristic writers so reely adapted the text o New
1estament writings in citations, well ater the New 1estament writings had acquired unquestionably
scriptural status and their text was airly stable, we are warned about taking citations in the writings o
second-century authors as direct eidence o the state o the text in their time.

35
Bruce M. Metzger, Patristic Lidence and the 1extual Criticism o the New 1estament,` in ^er 1e.tavevt
tvaie.: Pbitotogicat, 1er.iovat, ava Patri.tic ,Leiden: Brill, 1980,, 16-88 ,quotation rom p. 183,.
36
Gordon D. lee, 1he 1ext o John in 1be ]erv.atev ibte: A Critique o the Use o Patristic Citations in New
1estament 1extual Criticism,` in Lldon J. Lpp and Gordon D. lee, tvaie. iv tbe 1beor, ava Metboa of ^er
1e.tavevt 1etvat Critici.v ,SD 45, Grand Rapids: Lerdmans, 1993,, 335-43 |originally published in ]
90,191, 163-3|.
16
On the other hand, we should certainly not ignore patristic citations. In a number o cases
they appear to presere ariants otherwise attested only lightly in Greek manuscripts or ersions.
1hereore, in other cases also where the extant eidence does not allow us to eriy matters we may
suspect that this is so. Clearly, there was luidity, sometimes considerable, in the way that the text o
New 1estament writings was handled in the second century. All the same, we should not disregard
the other indications that among the scribal tendencies o the time there was also, in some cases, a
recognizable concern to copy relatiely careully and aithully.
3

But there is a good deal more to be said about what the citations o New 1estament writings
in second-century authors tell us. My comments here are not intended to pronounce with inality,
only to gie reason to recognize that preious analyses are not adequate, and to underscore some
important questions and issues or urther research.
38


Collections
It is a well-known eature o second-century Christianity that cottectiov. o writings that came to be
part o the New 1estament were ormed and circulated. \e know that at some point the our
canonical Gospels came to be thought o as complementary renditions o the gospel story o Jesus,
and came to orm a closed circle enjoying distinctie regard in many Christian circles. \e know also
that collections o Pauline epistles were circulating, probably rom the late irst century, and were
likewise treated as scripture in at least some circles.
39
1hese phenomena are regularly and rightly
noted in histories o the New 1estament canon. But I propose that these collections constitute a

3
Note, e.g., the judgment by J. Neille Birdsall in his study o the text o Luke in P5 and P66, Rational
Lclecticism and the Oldest Manuscripts: A Comparatie Study o the Bodmer and Chester Beattty Papyri o
the Gospel o Luke,` tvaie. iv ^er 1e.tavevt avgvage ava 1et. ..a,. iv ovovr of Ceorge D. Kitatric/ ov tbe
Occa.iov of bi. it,fiftb irtbaa, ,No1Sup 44, Leiden: Brill, 196,, 39-51.
38
See also James A. Kelhoer, Miracte ava Mi..iov: 1be .vtbevticatiov of Mi..iovarie. ava 1beir Me..age iv tbe ovger
vaivg of Mar/ ,\UN1 2,112, 1bingen: Mohr Siebeck, 2000,, esp. 123-30, in critique o Koester`s methods
in assessing use,inluence o written sources in second-century writers.
17
curious and possibly more signiicant phenomenon than is relected in the attention usually gien to
it in scholarly studies. 1o be sure, these collections contributed to the larger collection that we know
as the New 1estament canon. But, well beore there was a New 1estament` canon or the debates
about what it comprised, what was the motiation or establishing such cottectiov. o writings, and
what unction,s, were these collections intended to sere
It is remarkable how early these collections o writings appear. Seeral recent studies agree
in pushing back the likely origin o a our-old Gospels collection to the earliest years o the second
century. 1his scholarly agreement is all the more interesting in that these studies pursue dierent
approaches and questions. 1heo leckel argues that the crucial actor was the production o the
amiliar orm o the Gospel o John in Johannine circles, and leckel places a our-old Gospel
collection sometime around 120 CL.
40
In an astonishingly detailed study o the long ending o Mark,
James Kelhoer argues ,persuasiely to my mind, that these erses were composed sometime in the
irst hal o the second century, with conidence,` he judges, ca. 120-150 CL, and that they
presuppose a our-old Gospel collection that had been circulating and gien high respect` or some
time preiously.
41
In a recent study proposing identiication o a urther ragment o Papias`s
comments about the Gospels presered in Lusebius, Charles lill contends that Papias knew the our
canonical Gospels as a collection sometime ca. 125-135 CL.
42
In a recent book Martin lengel has
weighed in strongly in support o an early ourold Gospel collection as well.
43
But perhaps the most
programmatic sketch o the case or an early ourold Gospel collection has been oered by Graham

39
See, e.g., Gamble, oo/. ava Reaaer., 99-101.
40
1heo K. leckel, 1ov ravgetivv ae. Mar/v. vv rierge.tattigev ravgetivv ,\UN1 120, 1bingen: Mohr-
Siebeck, 1999,. See the reiew by Daid C. Parker, ]1 52,2001,: 29-301. C. the reiew by Kari Syreeni in
Rerier of ibticat iteratvre ,http:,,www.bookreiews.org,Reiews,31614199.html,.
41
Kelhoer, esp. 15, and 154-56, 158 ,n. 4,.
42
Charles L. lill, \hat Papias said about John ,and Luke,: A New` Papian lragment,` ]1 49,1998,: 582-
629, esp. 616-1.
43
Martin lengel, 1be ovr Co.et. ava tbe Ove Co.et of ]e.v. Cbri.t ,London: SCM, larrisburg, PA: 1rinity
Press International, 2000,.
18
Stanton in his 1996 SN1S Presidential Address.
44
\orking chronologically backwards rom
Irenaeus, Stanton concludes that a our-old Gospel collection was being promoted rom sometime
shortly ater 100 CL, though it took time to win its well-known supremacy.
\e can also note slightly earlier studies, in particular Bellinzoni`s careul analysis o Justin`s
use o Jesus` sayings.
45
le showed that Justin likely used compilations o Jesus` sayings drawn rom
the written Gospels, almost certainly the three Synoptic Gospels. But Justin`s reerence to the
liturgical reading o memoirs o apostles and those who ollowed them` ,Diat. 103.8, suggests
strongly that he knew o at least two Gospels attributed to apostles and at least two attributed to
others. 1he most likely conclusion is that Justin reers to our our canonical Gospels as regularly
read in worship.
As or a collection o Pauline epistles, the eidence points back at least as early. Indeed,
Daid 1robisch has proposed that Paul himsel may hae compiled the irst collection o his own
epistles.
46
1he reerence to all` Paul`s epistles in 2 Peter 3:16 probably takes us back to sometime
ca. 100 CL or earlier, although it is impossible to say what the all` comprised. lor, as Gamble
notes, it appears that the second-century Pauline collections were o arying dimensions, comprising
ten, thirteen, or ourteen letters.
4
Marcion`s exclusie claims or his ten-letter Pauline collection
sometime around 140 CL probably presupposes a widespread circulation o Pauline letter-collections
already by that point. By ca. 200 CL, howeer, there was an ao.toti/ov` category o Christian

44
Stanton, 1he lourold Gospel`.
45
Arthur J. Bellinzoni, 1be a,ivg. of ]e.v. iv tbe !ritivg. of ]v.tiv Mart,r ,No1Sup 1, Leiden: L. J. Brill, 196,.
See also lelmut Koester, .vcievt Cbri.tiav Co.et. ,London: SCM, Philadelphia: 1rinity Press International,
1990,, 360-402.
46
Daid 1robisch, Die vt.tebvvg aer Pavtv.brief.avvtvvg: tvaiev v aev .vfavgev cbri.tticber Pvbtii.ti/ ,N1OA 10,
Gottingen: Vandenhoeck & Ruprecht, 1989,.
4
Gamble, oo/. ava Reaaer., 59-63, 100. On the impact o Paul, see esp. Andreas Lindemann, Der Apostel
Paulus im 2. Jahrhundert,` in 1be ^er 1e.tavevt iv art, Cbri.tiavit,, ed. Jean-Marie Serin ,BL1L 86, Leuen:
Peeters, 1989,, 39-6, ia., Pavtv. iv atte.tev Cbri.tevtvv. Da. ita ae. .o.tet. vva aie Reetiov aer avtivi.cbev
1beotogie iv aer frvbcbri.tticbev iteratvr bi. Marciov ,1bingen: Mohr -Siebeck, 199,.
19
scriptures, comprising a Pauline collection plus letters attributed to other apostolic igures ,esp. 1-2
John, 1 Peter, James,.
In a sense, then, the New 1estament is a collection o prior collections. Daid 1robisch has
proposed igorously that the New 1estament as we know it was compiled as a single editorial project
sometime in the mid-second century.
48
lis argument is most intriguing in pointing to passages in
arious New 1estament writings that could be seen as intended to cross-reerence to, and accredit,
other New 1estament writings. I am not persuaded that a ull New 1estament collection such as
later came to be preerred was compiled and circulating as early as 1robisch contends. But it may
well be that the compilation o early collections o texts, such as a our-old Gospel and a Pauline
letter-collection, did stimulate the composition o other texts and helped to shape their contents,
including the embedding o the sorts o intriguing reerences that 1robisch highlights.
Collections also probably had an eect upon the transmission o the text o the component
writings. 1he most dramatic demonstration is, o course, 1atian`s Diate..arov ,ca. 12 CL,, a
thorough adaptation and expansion o earlier harmonizing texts ,such as may hae been used by
Justin,.
49
It appears that one o 1atian`s added eatures was a ull use o John. I am not persuaded
that the ew bits o material not paralleled in the extant texts o the canonical gospels is suicient or
the claim that Marcion used any ith gospel writing.
1he long ending o Mark,` as Kelhoer has powerully argued, is another striking textual
phenomenon relecting the our-old Gospel collection. 1his early addition to Mark appears to draw
upon the our canonical Gospels, and no other gospel writing. It shows, too, that the our canonical

48
Daid 1robisch, Die vareaa/tiov ae. ^evev 1e.tavevt.. ive |vter.vcbvvg vr vt.tebvvg aer cbri.tticbev ibet
,lreiburg,Gottingen: Uniersittserlag lreiburg,Vandenhoeck & Ruprecht, 1996,, Lnglish translation: 1be
ir.t aitiov of tbe ^er 1e.tavevt ,New \ork: Oxord Uniersity Press, 2000,.
49
See \illiam Petersen, 1atian`s Diatessaron,` in Koester, .vcievt Cbri.tiav Co.et., 403-30. But I am not so
persuaded as Petersen that the Diatessaron incorporated written sources other than the our canonical Gospels.
C., e.g., lengel, 1be ovr Co.et. ava tbe Ove Co.et of ]e.v. Cbri.t, 24-26. 1he ot-cited Dura ragment is not so
clearly a piece o the Diatessaron as scholars hae tended to think. C. now D. C. Parker, D. G. K. 1aylor and
M. S. Goodacre, 1he Dura-Luropos Gospel larmony,` in tvaie. iv tbe art, 1et of tbe Co.et. ava .ct., ed. D.
20
Gospels were not only collected but also compared with one another, which explains best why
someone thought that Mark`s ending was deicient and needed to be augmented along the lines o
the other Gospels.
50

In numerous smaller ariants as well, we can probably see the eects o a our-old Gospels
collection, especially the many harmonizations o one Gospel to another that I hae already
mentioned. Clearly, the special recognition gien to the our Gospels did not necessarily inole a
reluctance to make such improements` to their texts.
O course, it is oten thought that the Pauline letters exhibit signiicant eidence o the
eects o circulating in,as a collection. I, as it widely thought, 2 Corinthians is a composite writing,
might this composition hae taken place in some connection with an early Pauline collection But a
collection could dispose some toward shortening texts too. Gamble has argued that a iteen-chapter
ersion o Romans ,and perhaps a ourteen-chapter ersion as well, was prepared with a iew toward
wide ecclesiastical circulation, and a Pauline letter collection is the most likely ehicle or this.
51
le
has also noted eidence suggesting an early, certainly irst-century, eort to oercome the problem
|o the particularity o Paul`s letter-destinations| by deleting or generalizing the addresses o some o
the letters and sometimes by omitting other locally speciic matter as well . . .`
52

1he larger question, howeer, is why such collections emerged at all and became so
successully used. Other letter-collections are a known literary phenomenon in Roman-era antiquity,
both in Christian ,e.g., Ignatius o Antioch, and pagan circles. A Pauline letter-collection is, thus, not

G. K. 1aylor ,Birmingham: Uniersity o Birmingham Press, Atlanta: Society o Biblical Literature, 1999,, 192-
228.
50
Kelhoer, 154-55.
51
larry \. Gamble, 1be 1etvat i.tor, of tbe etter to tbe Rovav. ,SD 42, Grand Rapids: \illiam B. Lerdmans
Publishing Company, 19,.
52
Gamble, oo/. ava Reaaer., 60.
21
without precedent, though it is still a remarkable deelopment, and perhaps a precedent-setting
phenomenon or Christians o the early centuries.
53

But why did a our-Gospel collection emerge so early and manage to hae such success It
is obious that there were concerns about a plural and somewhat diergent testimony about Jesus.
54

Marcion is the most blatant illustration. But the harmonies, especially 1atian`s Diate..arov, also
indicate a certain discomort with our discrete accounts and in some circles a preerence or a more
cohesie rendition o Jesus. \hy, in particular, did Mark obtain a continuing place in the Gospels
collection, when in the eyes o many Christians in the second century Matthew seemed to hae
superseded it so adequately Collections o Pauline letters circulating by the late irst century might
be cited as a precedent and stimulus. But a Pauline letter collection represents one apostolic oice,
whereas a collection o Gospels, een the recognizably similar canonical our, embodies a diersity o
oices and contents.
1robisch is probably right to see in the our-old Gospel an early and deliberately
ecumenical` moe. Likewise, the inclusion o letters attributed to arious apostles in what became
the New 1estament ao.toti/ov represents a deliberate eort to express and airm a certain diersity
or breadth in what is treated as authoritatie.
55

Canon
1he inal phenomenon to consider briely is the emergence o a New 1estament canon, which
likewise appears to be well on its way by the end o the second century. \e can, thus, look to the

53
Daid L. Aune, 1be ^er 1e.tavevt iv it. iterar, vrirovvevt ,Philadelphia: \estminster Press, 198,, 10-2,
reers to collections o letters o Aristotle, Isocrates, Demosthenes, Apollonius o 1yana, Libanius, Cicero,
Pliny the \ounger, and the senior Pliny.
54
Oscar Cullmann, 1he Plurality o the Gospels As a 1heological Problem in Antiquity,` in Oscar Cullmann,
1be art, Cbvrcb: tvaie. iv art, Cbri.tiav i.tor, ava 1beotog,, ed. A. J. B. liggins ,London,Philadelphia:
SCM,\estminster, 1956,, 39-54, lelmut Merkel, Die Ptvratitat aer ravgetiev at. tbeotogi.cbe. vva eegeti.cbe.
Probtev iv aer attev Kircbe ,1raditio Christiana 3. Bern,lrankurt: Peter Lang, 198,, ia., Die !iaer.rvcbe ri.cbev
aev ravgetiev. bre otevi.cbe vva aotogeti.cbe ebavatvvg iv aer .ttev Kircbe bi. v .vgv.tiv ,\UN1 13,1bingen:
J.C.B. Mohr, 191,.
55
1robisch, Die vareaa/tiov ae. ^evev 1e.tavevt., 158-60.
22
second century as the period o the key impetus and proto-canonical` deelopments. In the
interests o limited space, I shall simply mention some key matters.
Aboe all, we again must reckon with the practice o liturgical reading. As we know, those
writings that contended or acceptance in the canonical decision-making attested in writings rom ca.
180 CL onward were those that had already enjoyed widespread inclusion among the texts read in
Christian worship gatherings. 1hose that made it` the most quickly were those that had the widest
usage rom the earliest years, and the remaining questions about the rest were settled largely on the
basis o whether they had been accepted or liturgical reading suiciently widely.
1he earliest precedent and impetus disposing Christian circles to include writings o their
own or such liturgical usage was almost certainly the letters o Paul. 1hey were composed as
liturgical texts, to be read out in the gathered assemblies to whom they were originally sent, and were
kitted out` with liturgical expressions to make them it this setting more readily, especially the well-
known letter opening and closing expressions. Moreoer, i Colossians is a deutero-Pauline
composition, it neertheless shows that the exchange o Paul`s letters among churches began within
the years ollowing his execution ,4:16,, and, i it is an authentic letter o Paul`s, then the practice was
earlier still.
As I hae indicated already, it appears also that the Gospels that became canonical circulated
impressiely widely and early. 1hat Mark was used so thoroughly as source and model by the authors
o Matthew and Luke shows that Mark circulated inluentially in arious Christian circles within a
short period ater its composition. 1hereater, at least to judge rom the comparatiely greater
number o early copies extant, it appears that Matthew outstripped all the others in breadth o usage
and requency o copying. But John too appears to hae enjoyed impressie success ery early.
leracleon`s commentary, written sometime ca. 150-15 CL suggests that John had or some time
enjoyed scriptural signiicance in at least some circles. ,It is noteworthy that we hae no such
commentary on any Christian writing that did not come to orm part o the New 1estament.,
23
1he arious early public-readers aids mentioned earlier as characterizing copies o New
1estament writings already in the early second century relect their usage as liturgical texts. Again,
the closest pre-Christian precedents and analogies or these scribal eatures are ound in Jewish
copies o Old 1estament scriptural writings that came to be included in the closed canon o
Judaism.
56

All this early interest in the public reading o certain writings as part o the liturgical lie o
Christian groups suggests that we might need to re-think the iew that it was only in the later decades
o the second century that a text consciousness` came to be inluential. \e hae, perhaps,
somewhat romantically regarded the earliest Christian circles as so gien to oral tradition that their
writings took a distant second place in their alues. I submit that rom the earliest obserable years
Christianity was a prooundly tetvat moement.
1o cite an early indication, although Paul was an intrepid itinerant preacher, and
characterized himsel primarily as such ,Rom 15:1-21,, een in his own lietime his critics reerred to
the eects o his tetter. ,2 Cor 10:9-11,. 1he production o the deutero-Pauline letters and, indeed,
the larger production o pseudonymous letters as well, attest that writings were early an inluential
mode o Christian discourse, persuasion, and promotion o religious ideas. 1he reerence to the
books and aboe all the parchments` in 2 1imothy 4:13 shows how much Paul was associated with
texts in the subsequent circles that reered him. John o Patmos coneyed his colorul isions and
words in a tet, which he clothed in prophetic authority and or which he demanded respectul
reading and copying ,Re 21:18-19,.
1he production o multiple written renditions o Jesus in the irst century and onward shows
also that texts were an early and aored mode or transmitting traditions about him. Len earlier
than the canonical Gospels, the Q sayings-source illustrates this as well. 1he continuing prolieration
o gospels` beyond the our that became canonical was apparently already well under way in the

56
See, e.g., 1o, 1etvat Critici.v of tbe ebrer ibte, 208-1, on scribal practices that can be traced back to pre-
24
second century, and urther shows how gien to texts early Christian were or circulating their
traditions about Jesus.
\e hae, perhaps, read too much into the ot-quoted words o Papias about preerring the
reports o liing and suriing` oices oer books.
5
Papias` proession simply echoes the sort o
claims that ancient historians regularly made or their works, claims that they either were
eyewitnesses themseles or had learned o the eents they narrate rom witnesses.
58
1hat is, Papias`
words do not really represent a preerence or oral tradition, but instead relect the titerar,
conentions o his time, in which one sought authority or one`s rrittev reports through claiming that
they rested on authentic witnesses. And need I remind us that this Papias who supposedly disdained
books is himsel reported to hae written a ie-olume rrittev exposition o the sayings o Jesus
It is true that Christian writers o the decades prior to ca. 150 CL do not characteristically
cite texts explicitly in the way that it is done much more requently in subsequent times. But is the
practice o the post-150 CL period indicatie o an emergent text consciousness,` or is it more
correct to see an emergent avtborcov.ciov.ve.. 1hat is, I suggest that what changes in the post-150
CL period is a greater tendency to see texts as the ror/. of avtbor., and so to cite them as such, rather
than simply appropriating the contents o texts. And I urther suggest that a major reason or a
greater emphasis on texts as products o particular authors is the swirling controersies o the second
century oer heresies. 1his led Christians to place greater emphasis on authorship o writings as a
way o certiying and,or promoting them. So, or example, whereas the canonical Gospels were
composed without the authors identiying themseles, across the second century we see an increasing
tendency to attribute and emphasize authorship o writings, including a greater tendency to attribute

Christian manuscirpts.
5
As quoted in Lusebius, 3.39.4. But see L. C. A. Alexander, 1he Liing Voice: Skepticism 1owards the
\ritten \ord in Larly Christian and in Graeco-Roman 1exts,` 1be ibte iv 1bree Divev.iov., ed. D. J. A. Clines
,Sheield: Sheield Academic Press, 1990,, 221-4.
58
See, e.g., the discussion o lellenistic historiography by Aune, 1be ^er 1e.tavevt iv it. iterar, vrirovvevt, 80-
83.
25
authorship to writings or which authorship was not an explicit eature o the text ,e.g., the canonical
Gospels, lebrews,.

Conclusion
Gien the breadth o phenomena and issues inoled in the three processes that I hae addressed
here, it has been necessary to limit the extent o my discussion o any o them. \here I hae taken a
position on controersial matters, the unaoidable breity means that I cannot hope to hae
persuaded anyone irmly holding another iewpoint. But I hae aspired here, not only to reiew the
releant phenomena and issues, but also to underscore the importance o the second century or the
writings that came to comprise our New 1estament. I hope also to hae helped to dispose scholars
o the New 1estament, and scholars o the text o the New 1estament in particular, to make a ull
harest o the materials aailable or researching how New 1estament writings were treated in the
second century. Recent studies, and recently aailable manuscripts and their data as well, proide us
with some potentially exciting prospects or urther knowledge and insight about this crucial period.
More than eer, it is in the interests o any particular question or line o inquiry into the second
century that we try to take as much account as we can o the spectrum o questions, issues, and
aailable eidence.