LUKE–ACTS: THE PROBLEM OF EDITING A TEXT

WITH A MULTIPLE TEXTUAL TRADITION

Jenny Read-Heimerdinger

Introduction

The aim of my paper is to present the recent edition of Luke’s Gospel

and Acts according to Codex Bezae (D), published in collaboration
with Josep Rius-Camps from the Facultat de Teologia de Catalunya: Lluc.
Demostració a Teòfil: Evangeli i Fets dels Apòstols segons el Còdex Beza
(Barcelona: Fragmenta Editorial, ). I would like to give an account
of the reasons for creating this edition and to demonstrate the purpose
that we believe it serves.

. Presentation of the
Edition of Luke–Acts in Codex Bezae

The edition sets out the text of Luke and Acts on two facing pages, with
the Greek text of Codex Bezae on the left hand side and the Catalan
translation on the right hand page in corresponding lines. We have kept
the lines short in order to make it easier to maintain the parallels between
the two pages, following any indications given by the copyist of Codex
Bezae as to the stichoi of the exemplar (or even earlier stages of the
text), which appear to have been frequently combined to make longer
lines.1 The translation is designed to reflect as closely as possible the
nuances of meaning in the Greek, being based on the results of our earlier
exegetical study of the Bezan text.2 The edition is designed in such a

1 The question of the lines in the exemplar of Codex Bezae is complex. In general, we
have followed the conclusion reached by David C. Parker, Codex Bezae. An Early Christian
Manuscript (Cambridge, ): “I suggest that the argument for medial points and small
spaces both representing a line division in the exemplar is well established” (p. ). In
addition, as he also suggests, the double point, the high point (in Acts) larger spaces and
projecting lines may well serve as indicators of earlier line length, though apparently not
always at the same time in the history of the text (ibid., pp. –; –).
2 With regard to Luke, see, for example, J. Read-Heimerdinger, ‘Enslavement and
 jenny read-heimerdinger

way that the Catalan can be easily replaced with other languages, such
as English or French, for example. In the samples set out in the appendix,
the text is printed in parallel columns in the interest of the economy of
space. The letters in square brackets in the translation are provided to
indicate the narrative structure, starting from the smallest division of
the text (a sentence with a finite verb and dependent clauses indicated
with small letters) and ending with sections made up of groups of related
episodes.3 The text is annotated (not shown in the sample) in order to
comment on specific features of the Bezan text or to justify the choice of
translation.
In presenting the manuscript of Codex Bezae as a continuous text, the
edition treats it as a text in its own right, rather than as a series of variant
readings. The text was initially established from Scrivener’s edition of
Codex Bezae, with reference made to the actual manuscript of Codex
Bezae, kept in Cambridge University Library, whenever there was doubt
over the original reading.4 A difficulty arises for Acts because of the large
number of lacunose passages in Codex Bezae. In those places, we have
cited the text of witnesses that give occasional and sporadic support to
Codex Bezae elsewhere. In consequence, those passages do not have the
same status as that of the secure Bezan text. I will say more about this
issue later.

Redemption: The Census of Augustus and the Birth of Jesus in Luke .– Codex Bezae’,
in Facultat de Teologia de Catalunya (ed.), A la recerca del sentit de la paraula, Revista
Catalana de Teologia  (), pp. –; ‘Where is Emmaus?’, in D.J. Taylor (ed.),
The Early Text of the Gospels and Acts (Birmingham, ), pp. –; J. Rius-Camps—
J. Read-Heimerdinger, ‘After the Death of Judas: A Reconsideration of the Status of the
Twelve Apostles’, Revista Catalana de Teologia  (), pp. –. With regard to
Acts, see especially J. Rius-Camps—J. Read-Heimerdinger, The Message of Acts in Codex
Bezae: A Comparison with the Alexandrian Tradition. I. Acts .–.: Jerusalem; II. Acts
.–.: From Judaea and Samaria to the Church in Antioch; III. Acts –.: The Ends
of the Earth; IV. Acts .–.: Rome (JSNTSup. ; LNTS , , ; London,
, , , ), where details of studies on specific passages are given at the
appropriate points.
3 The thinking behind the structural analysis is explained in Rius-Camps—Read-

Heimerdinger, The Message of Acts, vol. III, pp. –.

4 A facsimile edition of Codex Bezae was published in  by F.H. Scrivener, Bezae

Codex Cantabrigiensis (Pittsburgh, Pennsylvania, repr. ), which is largely accurate; its
Greek text was reproduced in an edition by A. Ammassari, Bezae Codex Cantabrigiensis
(Città del Vaticano, ), where the original arrangement of the Latin text was unfortu-
nately not respected. A digitalized online copy of the manuscript has recently been made
available by Cambridge University library at http://www.lib.cam.ac.uk/exhibitions/KJV/
codex.php?id=.
 editing a text with a multiple textual tradition 

In the edition of the text, we do not give a critical apparatus. For

the book of Acts, this is to be found in our -volume commentary
comparing the text of Codex Bezae with that of the Alexandrian tradition
(represented by Codex Vaticanus B, and Codex Sinaiticus !);5 a
similar comparative commentary on Luke is anticipated. Our aim in the
commentary was to compare not only the wording of the manuscripts,
but especially to bring out the differences in the message each tradition
transmits. Thus, the first step for each section of the narrative was to note
every variant reading between the selected manuscripts and to list its
support among the available witnesses (Greek, versional, patristic and
lectionary).6 The readings were evaluated one by one, examining them
from a linguistic point of view as well as analysing their contribution to
the distinctive message of each of the two texts. A separate section of the
commentary discusses in greater detail this latter aspect, bringing out the
differences in their theology and historical perspective.

. The Textual Tradition of Luke’s Work

The primary purpose of this edition of Luke’s work is to make accessible

his text in its entirety as it has been preserved in Codex Bezae, in the same
way that the text of the Alexandrian manuscripts has long been available,
whether in Greek in the current editions or in modern translations
since the beginning of the th century. I would like to explain in more
detail the rationale behind the edition. Why was it important to combine
Luke’s Gospel and Acts in one volume? And why in this particular
manuscript?

5 The Message of Acts, see note .

6 In addition to a range of specialist studies of individual verses, the following critical
editions were consulted: for the Greek text of Luke and Acts, Eb. Nestle, Novi Testamenti
Graeci: Supplementum editionibus de Gebhardt Tischendorfianis. Codex Cantabrigiensis
Collatio (Leipzig, ); R. Swanson, New Testament Greek Manuscripts: Variant Readings
Arranged in Horizontal Lines against Codex Vaticanus. Luke; The Acts of the Apostles
(Sheffield, , ); for Luke in general, The American and British Committees of
the International Greek New Testament Project (eds), The Gospel According to St. Luke.
Part I, Chapters –; Part II, Chapters – (Oxford: Clarendon Press, , ); for
Acts in general, M.-E. Boismard—A. Lamouille, Le texte occidental des Actes des Apôtres.
Reconstitution et ré-habilitation. Vol. I, Introduction et Texte; vol. II, Apparat critique
(Paris: Editions Recherche sur les Civilisations, ); J.H. Ropes, The Text of Acts, in
F.J. Foakes Jackson—K. Lake (eds), The Beginnings of Christianity, Part I. The Acts of the
Apostles, vol. III, (London, ).
 jenny read-heimerdinger

The need for a continuous edition of Luke’s work in Codex Bezae arose
from our study of the text of Acts first of all. The textual situation of the
book of Acts is well-known, that it has come down to us in two forms,
the familiar Alexandrian form and the so-called ‘Western’ one which
has generally been classed as secondary. The impression has often been
given that the texts represent two distinct traditions of Acts, two separate
developments with which all other witnesses are to be assimilated and
classified as having either a “Western (or D) text” or an “Alexandrian
(or B) text”.7 Our analysis of the text of Acts leads us rather to the
conclusion that the text of D and that of B stand at the two extremes
of a period of development, a period in which changes were introduced
progressively, and to some extent freely; and furthermore, that the text of
D represents the earlier of the extremes.
The text of Acts known to the Church in the West, that of the Alexan-
drian manuscripts, has few differences between its main representatives,
! and B, and has a extensive support from other witnesses. The alter-
native form of text is represented by a diverse range of witnesses, of which
D is the principal extant Greek witness and the manuscript that differs
the most from ! and B—indeed, in over one quarter of the length
for Acts. Approximately   of this variation arises through material
present in D that is absent from the Alexandrian manuscripts, and
some   through the reverse—material absent in D but present in
the Alexandrian manuscripts; some   arises through material that is
present in both texts but in a different lexical or grammatical form; and
finally   of variation is accounted for by identical material with a dif-
ferent order of words. The result is a considerable amount of variation,
rather more complex than the usual picture that is painted, of one text
that is simply much longer than the other and to which the label ‘the long
text’ is given. Despite the figure of   that is often repeated to refer the
extent by which D is longer than ! and B, more precise calcula-
tions show it to be . .8
The text of the witnesses that attest a text that diverges from the
Alexandrian tradition tends to be treated globally as a type of text.
The fallacy of regarding the variation as a ‘text-type’ is illustrated by

7 The portrayal of a two-fold textual tradition of Acts is challenged by D.C. Parker—

S.R. Pickering, : Acta Apostolorum –, –, in The Oxyrhynchus Papyri 
(GRM, ; London, ), pp. –, who describe the early text of Acts as a ‘free text’.
8 The figures are taken from the analysis presented in J. Read-Heimerdinger, The

Bezan Text of Acts: A Contribution of Discourse Analysis to Textual Criticism (JSNTSup.

; Sheffield, ), pp. –.
 editing a text with a multiple textual tradition 

the lack of homogeneity among the witnesses. Codex Bezae stands out
not only by the amount and the consistency of its difference from the
Alexandrian text, but also because it presents a coherent and cohesive
text that has a perspective and a message quite distinct from those of
the Alexandrian text. Some of its readings are singular in so far as they
have no known support. However, this situation is a relative one, for
some Bezan readings were considered to be singular until the discovery
in the s of such versions as the Middle Egyptian manuscript of
Acts –, known as mae or G67, or at around the same time, the
discovery of the fragment in Syro-Palestinian found at Khirbet-Mird
(sypal). Versional support is, indeed, present for many of the Bezan
readings, especially in the earliest translations in Latin and Syriac,9 and
over a wider geographical area that is far from being confined to the
west. The recent collation of a th century Gospel manuscript in Arabic
has revealed further support for some Bezan readings of Luke’s Gospel.10
The citations in the writings of the Church Fathers in Greek, Latin and
Syriac lend additional support. From among the Greek manuscripts, the
papyrus P38 from around  is the earliest to display Bezan readings;11
there is further sporadic support, in content if not in form, in the text
of the Latin-Greek bilingual manuscript, Codex Laudianus (E). It is
not uncommon, either, for the Byzantine manuscripts, such as H,
L P Ψ or the mass of minuscules (M), to support a Bezan

9 Caution must nevertheless be exercised in appealing to the versions for support of

Codex Bezae, as underlined by the work of Peter Williams on the Syriac tradition (Early
Syriac Translation Technique and the Textual Criticism of the Greek Gospels, [Texts and
Studies, ; Piscataway, NJ, ]).
10 See H. Kachouh, The Arabic Versions of the Gospels: The Manuscripts and their

Families (Arbeiten zur neutestamentlichen Textforschung, ), Berlin, forthcoming July

 (available online as his PhD thesis, Birmingham, UK,  [vol. I, pp. –] at
http://ethos.bl.uk/). For a study of the manuscript, see Idem., ‘Sinai Ar. N.F. Parchment
 and : Its Contribution to Textual Criticism of the Gospel of Luke’, NovT  (),
pp. –.
11 For a detailed analysis of the Greek papyrus support for Acts in Codex Bezae,

see J. Read-Heimerdinger—J. Rius-Camps, ‘Tracing the Readings of Codex Bezae in

the Papyri of Acts’, in C. Clivaz—J. Zumstein (eds), Reading New Testament Papyri in
Context/Lire les Papyrus du Nouveau Testament dans leur contexte (Leuven—Paris—
Walpole, MA, ), pp. –. The study develops and refines the presentation of
the Acts papyri in B. Aland, ‘Entstehung, Charakter und Herkunft des sog. westlichen
Textes untersucht an der Apostelgeschichte’, EThL  (), pp. –; and J.K. Elliott,
‘Codex Bezae and the Earliest Greek Papyri’, in C.-B. Amphoux—J.K. Elliott (eds.), The
New Testament Text in Early Christianity: Proceedings of the Lille Colloquium, July /Le
texte du Nouveau Testament au début du christianisme: Actes du colloque de Lille, juillet
 (Lausanne, ), pp. –, esp. pp. –.
 jenny read-heimerdinger

reading against the Alexandrian text, but this is far from consistent for
at other times these manuscripts support the Alexandrian readings or
have a conflation of both readings.
The fact remains that Codex Bezae is the only manuscript to differ
consistently from the Alexandrian text of Acts. A problem thus arises for
the extensive passages of Acts missing from Codex Bezae when it comes
to presenting a text for the parallel edition, as noted earlier. What we have
done is to give the readings of the manuscripts that commonly support
Codex Bezae in its extant text, the Middle Egyptian, or the Old Latin
Fleury palimpsest (h) or Codex Laudianus, for example. It turns out
that at any given place of variation, there is usually only one reading that
varies from ! and B, but the attestation for the variant differs from
one place to another. Much the same picture emerges in the support for
the extant text of Codex Bezae, with the variants supported here by one
witness and there by another. The solution for the lacunose passages is,
nevertheless, far from satisfactory, not least because the high number of
singular readings evident in the extant Bezan text means that there may
well be readings of the missing Greek text of Codex Bezae that have not
been preserved in any other witnesses.
Our conclusion is quite different from that arrived at by Boismard
and Lamouille in their edition of Acts. They, too, present two texts in
parallel columns, one Alexandrian and one ‘Western’. The latter, ‘le texte
occidental’, was reconstituted from a wide range of available witnesses
including late versions, Church Fathers and lectionaries, on the basis of
a statistical analysis of recurring stylistic features, which they deemed
to be characteristically Lukan. According to their criteria, the text of
Codex Bezae is ‘un témoin très abâtardi [du texte occidental]’12 but, as
I have argued elsewhere,13 the identification by Boismard and Lamouille
of linguistic features that are presumed to be typically Lukan and that
serve as a standard against which variant readings are judged, depends
on circular reasoning and a priori assumptions.
The preceding description in this section relates to the book of Acts.
The textual picture of Luke’s Gospel is somewhat more complex. Apart
from the shorter readings in the final chapters of the Gospel, which are a
well-known feature of non-Alexandrian manuscripts, detailed compari-

12 Boismard and Lamouille, Le texte occidental, I, p. .

13 See my review of J. Taylor, Les Actes des Deux Apôtres, Journal of Theological Studies
 (), pp. –.
 editing a text with a multiple textual tradition 

son of its witnesses and the development of its text remains to be carried
out. Our exploratory analysis tends to lead to the same conclusion as for
Acts, namely that the Bezan text predates the Alexandrian one. This view
was arrived at on the basis of the cohesion and coherence of the Bezan
text, aspects of its language and of its theological message and historical
perspective already alluded to above. The unique readings of its text in
Luke’s Gospel display the same linguistic features as the firm text (the text
without variants) and, most importantly, serve as anchors to the essential
differences in the message as will be illustrated in Section  below.
With regard to the content of Codex Bezae in general, the view of Kurt
Aland was that its text is the end product, ‘ein Höhepunkt’, of a series
of texts which sought to provide a paraphrase of a previous version.14
David Parker likewise deemed it to be made up of successive layers of
emendation and error.15 Our conviction is that the inner coherence of
the language of Codex Bezae and of its message, both with regard to the
text of Acts as also the Gospel of Luke, speaks against its representing
the culmination of a process of revision, or its being the work of different
hands from different times with different intentions.

. Luke and Acts

I have mentioned the book of Acts and the Gospel of Luke together,
and indeed, the parallel edition is intended to present both books as a
unity, as two volumes of the same work. As far as we are aware, this has
not been done before. The reason for this decision is quite simple: as we
examined the text of Acts, it became increasingly clear with every chapter
of Codex Bezae that Acts was intended as the second volume of a single
work by one author, whose first volume prepared for the second, and
whose second volume depends closely on the first for its full meaning
and sense. The fact that they have a common author, whom tradition has
designated Luke, and a common addressee whom the author refers to by

14 K. Aland, Text und Textwert der griechischen Handschriften des Neuen Testaments.
III. Apostelgeschichte ( vols., ANTF, –; Berlin, ), vol. I, pp. –. See also
B. Aland (‘Entstehung, Charakter und Herkunft des sog. westlichen Textes’) who likewise
assigns to D a position at the end of a period of development.
15 D.C. Parker, Codex Bezae, pp. –; and see also his ‘Professor Amphoux’s

History of the New Testament Text: A Response’, New Testament Update  (), pp. –
.
 jenny read-heimerdinger

name as ‘most excellent Theophilus’, is a minor consideration compared

with the cohesion of the work from the point of view of its purpose and
message. The common author and addressee did not prevent the two
volumes becoming separated at an early date, as evidenced by the lack
of any manuscript that presents them as successive books. It is probable
that the Gospel was separated from Acts in order to group it with the
other Gospels; certainly, it seems to have taken longer for Acts itself to be
viewed as having the authority of a canonical book.16
The unity of Luke and Acts is contested by some17 because of a per-
ceived theological disparity between the two books. Most scholars, never-
theless, argue for its unity on the basis of a similarity of language, themes,
and narrative style. Codex Bezae, more than any other manuscript, dis-
plays a high degree of homogeneity in both language and message. In
particular, the Gospel anticipates the narrative of Acts, and in Acts there
is a constant echoing and development of themes and terms used in the
Gospel; further evidence is to be found in the way Luke holds over ele-
ments of Mark’s Gospel to Acts instead of using them in his own Gospel.
All of these factors are more evident in the text of Codex Bezae compared
with the Alexandrian text. As for the theological discrepancies between
the two volumes, these can be accounted for by recognizing the true
nature of the speeches in Acts, which constitute the main passages that
have a theological (as opposed to narrative) content. In Codex Bezae, it is
clear that the speeches are devices Luke uses to express the thinking of his
protagonists. Their thinking could, and in many cases did, change over
time, and it could well differ also from that of another character. In any
case, Luke by no means necessarily endorses their theology, and his prac-
tice of placing words in their mouths to convey the development of their
thought explains why it is so difficult to construct a consistent and uni-
fied theology from Luke’s writings. As far as the Bezan text is concerned,
it is possible to identify Luke’s own theology, but only from the actions
and the speeches of the characters for whom he indicates his approval by
mentioning their being in harmony with the Holy Spirit (Barnabas gen-
erally; Stephen also generally; Philip after Acts .; Peter only in Acts
; Paul only after Acts .).

16 The evidence for knowledge of the Gospel of Luke and Acts among the earliest
Church Fathers is examined by A. Gregory, The Reception of Luke-Acts in the Period before
Irenæus: Looking for Luke in the Second Century (WUNT ,; Tübingen, ).
17 M.C. Parsons and R.I. Pervo, Rethinking the Unity of Luke and Acts (Minneapolis,

).
 editing a text with a multiple textual tradition 

The unity, and singularity, of Luke and Acts in Codex Bezae is the
justification for publishing the text of the manuscript in its entirety, and
as a consecutive text.

. Difference in Message

I have mentioned several times a distinct narrator’s perspective in the text

of Codex Bezae and also a distinct message. In summary, in Codex Bezae,
the narrator tells his account from a Jewish perspective. His purpose
in the Gospel is to show how the person of Jesus and the events of
his life fit into the expectations of Jewish tradition, both written (the
Scripture) and oral (their interpretation in teachings and legends). He
is not telling a chronological story but presenting a theological account
from an insider Jewish point of view, using sophisticated techniques of
Jewish exegesis to make his points and alluding to complex scriptural
interpretations transmitted in oral tradition. The same is true of the book
of Acts, which covers the years of the early Church. There, the characters
are not only portrayed as re-enacting early Jewish history, but they are
also compared with Jesus as their model. The comparison is by no means
always favourable, for the disciples are seen to struggle to come to terms
with the radical nature of Jesus’ teachings and to accept the changes these
require to their traditional Jewish mentality and expectations.
The purpose of the narrator in the Bezan text impacts on the classifica-
tion of the genre of his work and results in a description that differs some-
what from those proposed in studies of the Alexandrian text. Though we
retain the conventional title ‘Gospel’, the first volume is not a gospel in
the sense of proclaiming good news and to that extent is not comparable
with the other gospels. The second volume, likewise, is neither biogra-
phy nor history. Both volumes share the common purpose of providing,
for Theophilus as a particular enquirer, trustworthy information about
known persons and events, and evaluating their worth and significance.
As such, it fulfils the function of epideictic rhetoric, or a ‘demonstration’,
the term we have chosen as a title for Luke’s work.
In the Introduction to the published edition, the considerations out-
lined here are presented in some detail in order to facilitate the reading
of the text and the appreciation of the features of Luke’s work in Codex
Bezae. Two examples, one from the Gospel and one from Acts, will serve
here to illustrate these features and bring out their distinctive character
compared with the perspective and message of the Alexandrian narrative.
 jenny read-heimerdinger

a. Luke .–
In the story of the journey away from Jerusalem made by two disciples
after the crucifixion, during which the resurrected Jesus appears to them,
the name of the village in Codex Bezae is Oulammaous, with no support
in any other extant document. All other witnesses read the familiar
‘Emmaus’. In the Bezan version of the story, the name ‘Oulammaous’ is
the key to the narrator’s purpose in telling the story; it serves as a hook
to connect the incident with a paradigmatic event in the history of Israel,
God’s self-disclosure to Jacob and his designation as the father of Israel
(Gen. .–). The name occurs at the point when Jacob awakes from
his dream in which God spoke to him; he calls the place Bethel, the ‘house
of God’, because of the presence of God there, and the Hebrew text then
continues:
. MT: "#$!%& %'(")*$ +& *&,!,
‘however, Luz was the name of the place beforehand’:
This half of sentence is translated in the lxx as:
, lxx: κα$ Ο&λαµλο)ς +ν -νοµα τ/0 π2λει τ5 πρ2τερον
The Greek translation renders the first two Hebrew words—oulam luz—
as a single word, thus giving the former name of the place as ‘Oulamlous’
which, by phonetic transformation, becomes ‘Oulammaous’ in several
minuscules of the lxx (as well as the references of Justin and Eusebius
to Gen. .), the form that is found in the Bezan text of Lk. .. The
use of the name by Luke confers on the scene a spiritual reality, whereby
the two disciples are presented as re-enacting the flight of Jacob from his
brother whom he has just tricked out of his birthright.18 In the Bezan text
of Luke , the betrayal of Jesus by Judas was already modelled on this
act of deception,19 and now the two are presented as taking flight just as
Jacob did.

18 A different justification for the Bezan reading than the one given here is indicated
by C.-B. Amphoux, ‘Le Chapitre  de Luc et l’ origine de la tradition textuelle du Codex
de Bèze (D. du NT’, Fil Neo  () pp. –. He proposes an underlying play
on words between Bethel (house of God, temple worship) and Bethlehem (house of
bread, represented by Jesus who replaces the ancient cultic practices with the breaking
of bread). Though his proposal is intriguing, this interpretation tends to see the coded
language as concealing the deeper meaning whereas Luke’s purpose in using Scriptural
allusions is to unveil it. Furthermore, there would have been no need for Luke to use the
singular designation of Bethel as ‘Oulammaous’ if it were the name ‘Bethel’ itself that were
important.
19 Compare Jacob who approached his father Isaac and kissed him as a (false) proof
 editing a text with a multiple textual tradition 

The parallels between the two scenes—the divine revelation in the

evening, the characters being unaware of what was taking place because
they were asleep or had their eyes closed, the subsequent realization—are
obvious and do not need to be spelt out in detail here.20 What is striking
in the Bezan text, however, is that by means of the paradigm—aligning
the disciples of Jesus with the Father of Israel—Luke also draws atten-
tion to a negative contrast. Whereas Jacob aligned himself totally with
the plan of God for Israel, the disciples understand so little of it that
when Jesus disappears from the scene in the Lukan story, the disciples
are left in a state of profound grief (λυπο8µενοι, . D, a verb found
only once more in Luke’s writings but only according to Codex Bezae, in
his account of the reaction of Jesus’ parents when they thought they had
lost him in Jerusalem [Lk. . D]). The reason for their grief is that
they have not properly understood what he had been trying to explain to
them about the Messiah in the Scriptures—they comment that their heart
was ‘veiled’ (κεκαλυµµ9νη, . D), and not ‘burning’ (καιοµ9νη) as
in the other manuscripts. Indeed, Jesus had only begun to interpret the
Scriptures according to this text (+ν ;ρξ=µενος . . . >ρµηνε8ειν α&το?ς @ν
τα?ς γραφα?ς, . D, cf. Cνοιγεν Dµ?ν τEς γραφ=ς, v. ), rather than
bringing the task to completion by interpreting all the Scriptures thor-
oughly (cf. B: διερµGνευσεν α&το?ς @ν π=σαις τα?ς γραφα?ς [v. ] . . .
διGνοιγεν Dµ?ν τEς γραφ=ς [v. ]); again, their eyes were opened, but
not completely (IνοJχLησαν οM NφLαλµοJ, . D; cf. διηνοJχLησαν
οM NφLαλµοJ, B). They have understood something of Jesus’ explana-
tion and they have realized that he has been resurrected—but now he has
disappeared for a second time, and they have no idea what this means.
Consequently, they immediately go back again to Jerusalem, OΙηρουσα-
λGµ, the religious centre of Judaism (.). It will take  chapters of
the book of Acts in Codex Bezae for them to detach themselves from the

of his identity (Gn . lxx @γγJσας @φJλησεν α&τ2ν) with Judas who approached Jesus
and kissed him (Lk. . D: @γγJσας @φJλησεν τ5ν OΙησο)ν) to confirm the identity
of Jesus to the chief priests and the guards. All other manuscripts apart from D read:
Cγγισεν τQR OΙησο) φιλ0σαι α&τ2ν, so losing the linguistic clue to the parallel.
20 Discussion of the parallels is developed in detail in J. Read-Heimerdinger, “Where

is Emmaus? Clues in the Text of Luke  in Codex Bezae”, in D.C. Parker—D.G.K. Taylor
(eds), Essays in New Testament Textual Criticism (Texts and Studies n.s., /; Birming-
ham, ), pp. –; see also J. Read-Heimerdinger—J. Rius-Camps, ‘Emmaous or
Oulammaous? Luke’s Use of the Jewish Scriptures in the Text of Luke  in Codex Bezae’,
Revista Catalana de Teologia  (), pp. –.
 jenny read-heimerdinger

ancient Jewish traditions and ways of thinking, which Luke indicates

by associating them for the first time at . with the neutral city of
SΙεροσ2λυµα.21
Even from this brief consideration of the readings in this passage
of Codex Bezae it can be seen how the whole text hangs together to
form a coherent story, with a consistent message. It is quite a different
one from that derived from the story in the other manuscripts. How
often this passage is taken in Church sermons precisely to exemplify the
joy of the resurrection! In the usual account, the disciples understand
perfectly what Jesus teaches them in his final hours before he ascends to
heaven, and they are ready to begin spreading the good news exactly in
accordance with the divine plan. This complete understanding sets the
scene for the book of Acts in which the apostles, to whom Paul is added,
are presented as infallible figures of authority. The narrative of Codex
Bezae, in contrast, continues with the same picture that had been painted
in the Gospel, of fallible human beings who only gradually come to grasp
the full extent of the radical nature of Jesus’ message.

b. Acts .
The portrait of Paul in the Bezan Acts is carefully and subtly painted,
showing how he gradually progressed in his acceptance of the mission to
the Gentiles. A major difficulty that Luke presents is his determination to
maintain the privileged status of Israel as the chosen people of God, into
which the nations are to be grafted. This view was, of course, at the very
centre of his Jewish heritage and is taken up in his letters, but Bezan Acts
consistently presents it as erroneous. Paul’s insistent attempts to persuade
the Jews about the messiahship of Jesus cause him to disobey the divine
will on repeated occasions.
His disobedience is seen particularly in his plan to take a collec-
tion of money to Jerusalem from the churches where there are Gentile
believers—this is the purpose of his final journey to Jerusalem where he
ends up being taken prisoner. From the outset, the Bezan text presents
this fatal journey as being Paul’s own plan which the Holy Spirit opposes,
as seen in the text at . D: Θ9λοντος δU το) Πα8λου κατE τWν XδJαν
βουλWν πορε8εσLαι εXς SΙεροσ2λυµα, εZπεν α&τQR τ5 πνε)µα [ποστρ9-
φειν εXς τWν OΑσJαν. Following this initial conflict, Luke in Codex Bezae

21On the distinction between Ierousalem and Hierosoluma in Codex Bezae, see Read-
Heimerdinger, The Bezan Text, pp. –.
 editing a text with a multiple textual tradition 

will describe five more attempts to dissuade Paul from carrying out his
plan because it was contrary to the will of God. The group represented
by the intermittent first person plural narrative (the ‘we’ group) plays an
important part in these attempts, for they have the specific function of
reminding Paul of the divine will and persuading him to accept it, all
of which Paul doggedly resists until he finally capitulates in Rome when
Luke brings his story to a close.
The Spirit’s guidance at . D is missing from the Alexandrian
text, which simply reports that Paul went to Ephesus while Apollos was
in Corinth. Indeed, in this text only four warnings of the danger that
Paul would meet in Jerusalem are recorded, as against six in Codex
Bezae, and the wording of these four is so much weaker than that of
the corresponding Bezan passages that they are generally not interpreted
as expressing opposition so much as prophecies of suffering which Paul
will, nobly and heroically, bear. In the Bezan text, Paul does not emerge
as a hero but rather as a flawed, very human character whose arrest
in Jerusalem is viewed as arising through his own stubbornness and
as constituting a hindrance to the mission with which he had been
entrusted.

. An Historical Explanation

On the surface, the opposition of Luke in Codex Bezae to Paul’s teaching

about the privileged status of Israel looks like the work of a Gentile
believer who is critical of the ancient claims of Judaism.22 That it is
the work rather of an insider is seen by repeated references in that text
(more frequent and more intricate than in the Alexandrian text) to Jewish
scriptural paradigms that are used to justify the change in the status of
Israel. The use is quite different from the later typology of the Church
Fathers—in Codex Bezae as in Jewish exegesis, the events and characters
of ancient Israel are viewed as foundational and the contemporary events
and characters as re-enactments of the ancient ones. There is no sense of
triumphalism but instead a sense of tragedy and regret. The narrator’s
opposition to traditional Jewish beliefs and expectations is made with

22 This was the contention of E.J. Epp, The Theological Tendency of Codex Bezae
Cantabrigiensis in Acts (SNTS Monograph, ; Cambridge, ). See also Idem., ‘Anti-
Judaic Tendencies in the D-Text of Acts: Forty Years of Conversation’, in T. Nicklas—
M. Tilly (eds), Apostelgeschichte als Kirchengeschichte. Text, Traditionen und antike Ausle-
gungen (BZNW, ; Berlin-New York, ), pp. –.
 jenny read-heimerdinger

insider knowledge, in Jewish terms, in much the same way as it is in the

books of the Prophets. This is seen in chapter after chapter of the Bezan
text of Luke and Acts.
When the evidence for the Jewish perspective of Luke in the Bezan
readings is compared with the Alexandrian text, it can be seen that not
only is the criticism of the apostles toned down in that text but, more-
over, the narrative is presented more as a straightforward chronological
account than a theological demonstration. A great number of the Jewish
allusions is absent.
The combination in Codex Bezae of the Jewish perspective and the
critical evaluation of the disciples suggests an explanation for the devel-
opment of two such different texts of Luke-Acts: the text written from
a Jewish perspective dates from a time when the Church was still very
much a part of Judaism. It was composed for a distinguished and highly
educated member of Jewish society who had a sophisticated knowledge
of Judaism and who was familiar to some extent with the people and
events in the narrative. He was aware of the conflicting ideas and deeds of
the disciples, and of the contrast between many of the followers of Jesus
and the model laid down by the master, and he wanted to understand
what it all meant. To this person Luke, a similarly educated Jew who was
a believer in Jesus, set about explaining from the point of view of their
shared heritage just what happened.
The more realistic picture of fallible leaders of the Church would also
be more acceptable the closer it was to the date of the actual events.
However, as the Church moved away from its Jewish roots and the apos-
tles took on the status of infallible heroes, the more subtle Jewish allu-
sions, especially to oral teachings or legends, would have been incom-
prehensible, or even offensive. Consequently, the narrative was toned
down to soften them or remove them altogether. At the same time, where
the narrator indicated his disapproval of the apostles, showing them
to have made mistakes or taught things that were contrary to the plan
of God, the text was modified in order to remove traces of such criti-
cism.
Both books were known to the communities that received them
(indeed, that continued to receive them, in the case of the Alexandrian
text), and served as their two volumes of Luke’s work. It is because the
text of Codex Bezae is so different from the familiar text, as I have aimed
to show from some few examples here, that it is important to let it stand
as a continuous text instead of relegating its readings to a critical appara-
tus. This is the justification for the edition of Luke-Acts in Codex Bezae,
 editing a text with a multiple textual tradition 

whose first publication has been in Catalonia, Spain, and which we hope
will contribute to reinstating the text of Codex Bezae as a text of unique
importance for the historical and theological study of the Church in the
early centuries.
 LUKE, DEMONSTRATION TO THEOPHILUS

Volume : The Gospel According to Luke

SECTION XX. THE RESURRECTION

[B] Appearances to the Disciples

[B–A] The Disciples of Oulammaous

, ]Ησαν δU δ8ο [a] . There were two

πορευ2µενοι @ξ α&τRν who travelled away from them
@ν α&τ/0 τ/0 Dµ9ρ_α that same day,
εXς κ`µην to a village
;π9χουσαν σταδJους >ξGκοντα  stadia away
;π5 OΙερουσαλGµ, from Ierusalem,
Nν2µατι Ο&λαµµαο)ς, called Oulammaous.
 aµJλουν δU πρ5ς >αυτοbς [b]  They were talking between
περ$ π=ντων ⟨τRν⟩ themselves about all these things
συµβεβηκ2των το8των. that had happened.

[B–B] Jesus Teaches Them about the

Messiah

. Κα$ @γ9νετο [a] . It happened

@ν τQR Nµιλε?ν α&τοbς while they were talking
κα$ συζητε?ν and trying to work things out
κα$ f OΙησο)ς @γγJσας that Jesus himself came up to them
συνεπορε8ετο α&το?ς. and started travelling with them.
 (οM δU NφLαλµο$ α&τRν [b]  (Their eyes, however,
@κρατο)ντο were prevented
το) µW @πιγνRναι α&τ2ν.) from recognizing him.)
 f δU εZπενg [c]  He said,
ΤJνες οM λ2γοι οiτοι “What are these words that
οjς ;ντιβ=λλετε πρ5ς >αυτοbς you are arguing over between yourselves
περιπατο)ντες σκυLρωποJ; as you walk, looking so sad?”
 ;ποκριLε$ς δU εkς [d]  One of them,
l
Q -νοµα Κλεοπmς by the name of Cleopas,
εZπεν πρ5ς α&τ2νg answered him,
Σb µ2νος παροικε?ς “Are you the only person
OΙερουσαλGµ, who has been staying in Jerusalem,
ο&κ oγνως who doesn’t know
τE γεν2µενα @ν α&τ/0 the things that happened there
@ν τα?ς Dµ9ραις τα8ταις; these past days?”
 luke, demonstration to theophilus 

 f δU εZπεν α&τQRg Πο?α; [e] a He said to him, “What things?”

ΤE περ$ OΙησο) το) ΝαζωραJου, [f] b “About Jesus the Nazorene,
qς @γ9νετο ;νWρ προφGτης a man who was a prophet,
δυνατ5ς @ν λ2γQω κα$ oργQω powerful in word and in deed
@ν`πιον το) Lεο) before God
κα$ παντ5ς το) λαο), and all the people,
 aς το)τον παρ9δωκαν  how the high priests
οM ;ρχιερε?ς and our leaders
κα$ οM rρχοντες DµRν handed him over
εXς κρJµα Lαν=του to be sentenced to death
κα$ @στα8ρωσαν α&τ2ν. and they crucified him.
 Dµε?ς δU IλπJζοµεν  But we had been hoping
sτι α&τ5ς +ν N µ9λλων that he would be the one
λυτρο)σLαι τ5ν OΙσραGλg who was going to redeem Israel;
;λλ= γε κα$ not only that, though,
σbν πmσιν το8τοις but on top of it all,
τρJτην Dµ9ραν σGµεραν rγει today is the third day
;φO οi τα)τα γ9γονεν. since this happened.
 ;λλE κα$ γυνα?κ9ς τινες  And yet some women
@ξ9στησαν Dµmςg startled us:
γεν2µεναι NρLρινα$ they went at dawn
@π$ τ5 µνηµε?ον to the tomb
 κα$ µW ε&ρο)σαι  and when they couldn’t find
τ5 σRµα α&το) his body,
+λLον λ9γουσαι they came back saying
NπτασJαν ;γγ9λων >ωρακ9ναι, they had seen a vision of angels.
οt λ9γουσιν α&τ5ν ζ0ν. who said he was alive.
 κα$ ;π0λL2ν τινες  And some of our company
@κ τRν σbν Dµ?ν went off
@π$ τ5 µνηµε?ον, to the tomb,
κα$ εiρον οuτως and they found it
aς εZπον αM γυνα?κες, just as the women had said,
α&τ5ν δU ο&κ εvδοµεν. but we didn’t see him.”
 f δU εZπεν πρ5ς α&το8ςg [f ’]  He said to them,
]Ω ;ν2ητοι “O you foolish men,
κα$ βραδε?ς τ/0 καρδJ_α how dim-witted you are
@π$ πmσιν regarding everything
οkς @λ=λησαν οM προφ0ταιg the prophets spoke,
 sτι τα)τα oδει παLε?ν  when they said that the Messiah
τ5ν Χριστ5ν had to suffer these things
κα$ εXσελLε?ν εXς τWν δ2ξαν α&το). and enter into his glory!”
 κα$ +ν ;ρξ=µενος [e’]  And he made a start
;π5 Μωϋσ9ως with Moses
κα$ π=ντων τRν προφητRν amd all the prophets
>ρµηνε8ειν α&το?ς to explain to them
@ν τα?ς γραφα?ς from the Scriptures
τE περ$ α&το). the things relating to himself.
 jenny read-heimerdinger

 κα$ Cγγισαν εXς τWν κ`µην [d’] a They had come close to the village
οi @πορε8οντο, where they were going,
κα$ α&τ5ς προσεποιGσατο [c’] b and he made as if
πορρωτ9ρω πορε8εσLαι. he were travelling on further.
 κα$ παρεβι=σαντο α&τ5ν [b’] a They pressed him,
λ9γοντεςg saying,
Με?νον µεLO DµRν, “Stay with us
sτι πρ5ς >σπ9ραν for the day has turned
κ9κλικεν D Dµ9ρα. to evening”,
κα$ εXσ0λLεν µε?ναι µετO α&τRν. [a’] b and he went in to stay with them.

[B–C] Jesus Makes Himself Known

. Κα$ @γ9νετο [a] .a It happened that

@ν τQR κατακλιL0ναι α&τ5ν while he was reclining at table,
λαβ{ν rρτον η&λ2γησεν taking bread he said a blessing
κα$ προσεδJδου α&το?ςg [b] b and shared it with them.
 λαβ2ντων δU α&τRν [c] a As they took
τ5ν rρτον ;πO α&το) the bread from him
IνοJ⟨χL⟩ησαν οM NφLαλµο$ α&τRν their eyes were opened
κα$ @π9γνωσαν α&τ2νg [d] b and they recognized him;
κα$ α&τ5ς rφαντος [e] c and he disappeared
@γ9νετο ;πO α&τRν. from their sight.
 οM δU εZπον πρ5ς >αυτο8ςg [d’]  They said to each other,
Ο&χ$ D καρδJα +ν DµRν “Wasn’t our heart
κεκαλυµµ9νη veiled
aς @λ=λει Dµ?ν @ν τ/0 NδQR, while he was talking to us on the road,
aς Cνοιγεν Dµ?ν τEς while he was opening the Scriptures up to
γραφ=ς; us?”
 κα$ ;ναστ=ντες λυπο8µενοι [c’] a They got up, profoundly sad,
α&τ/0 τ/0 |ρ_α and at that very hour
[π9στρεψαν εXς OΙερουσαλGµ, went back to Ierusalem.
κα$ εiρον [b’] b They found
ILροισµ9νους τοbς ~νδεκα the Eleven gathered together
κα$ τοbς σbν α&το?ς, and those who were with them
 λ9γοντες  and they told them
sτι Οντως Iγ9ρLη “The Lord really
N κ8ριος has been raised
κα$ €φLη ΣJµωνι. and he appeared to Simon!”
 κα$ α&το$ [a’]  And they went on
@ξηγο)ντο τE @ν τ/0 fδQR to relate what had happened on the road
κα$ sτι @γν`σLη and that he had made himself known to
α&το?ς them
@ν τ/0 κλ=σει το) rρτου. when he broke the bread.
 luke, demonstration to theophilus 

Volume : Acts of the Apostles

PART IV. ROME: VIA EPHESUS

AND JERUSALEM
SECTION I. THIRD PHASE OF THE
MISSION: EPHESUS

[A] The Baptism of John

[A–A] . . .
[A–A’] The Twelve Disciples and the
Baptism of John

. Θ9λοντος δU το) Πα8λου [a] .a Although Paul wanted

κατE τWν XδJαν βουλWν according to his own plan
πορε8εσLαι εXς SΙεροσ2λυµα, to go to Hierosoluma,
εZπεν α&τQR τ5 πνε)µα the Spirit told him
[ποστρ9φειν εXς τWν OΑσJαν. to return to Asia.
διελL{ν δU τE ;νωτερικE [b] b Having gone through the upper
µ9ρη parts
oρχεται εXς Εφεσον . . . he arrives in Ephesus . . .