Textual Evidence

Billings (2006)
on Luke 22:19b-20

Review: Bradley S. Billings, Do This In Rememberance of Me ,
The Disputed Words in the Lukan Institution Narrative (Luke 22:19b-20): An Historico-Exegetical, Theological and Sociological Analysis

Page Index

Last Updated: Feb 19, 2009

Prologue: - Introduction to Luke 22:19a-20

Review: - Billings on Luke 22:19a-20: - the Textual Problem
    Nazaroo's Notes on the Syriac Versions

Return to Index



Bradley S. Billings' book "Examines the difficulty represented by the textual tradition in Codex Bezae at the point of the Last Supper narrative in St Luke's Gospel . With a survey of explanations of the difficulty, this title examines the disputed words of Luke 22:19b-20 in regards to their style, grammar and theology, to ascertain their source and non-Lukan features."

Billings' book is offered for sampling on Google Books, a new way of marketing and raising the profile of otherwise hopelessly obscure scholarly works.
Billings' Book <-- Click Here.

In doing this, he necessarily reviews the available textual evidence; not just the Greek manuscript variants, but also the early versions, essentially the Early Syriac translations.

That this is not an easy or trivial textual problem, is obvious from both a history of scholarly work in this area, and Billings' own description of the task he has set before himself:

Many a scholar has met his Waterloo in an attempt to account for, or explain away, the existance of the Bezan text.

- E.J. Epp, quoting Streeter,
'The "Ignorance Motif" in Acts and
Anti-Judaic Tendencies in Codex Bezae',

HTR 55 (1962), pp.51-62.

"I hope not to become a further casualty, but cannot deny the risk. The text of Codex Bezae is an often perplexing and frequently difficult area of study. This is nowhere more so the case than in the Lukan corpus of writings it preserves, and perhaps nowhere more importantly than in the Bezan text of Luke's Gospel at the point of the Last Supper narrative (Lk. 22:15-20).

Here scholars and commentators have, since ancient times, contemplated the existence of two streams of textual tradition; the so-called 'longer reading' which retains all of the familiar words of institution attested to by the vast majority of the external manuscript evidence, and a so-called 'shorter reading' preserved only by Codex Bezae and a few other Western and Syriac witnesses which omit the words of Jesus after 'this is my body' producing a decidedly truncated recension of the Last Supper tradition that lacks altogether the words over the cup and the dominical command to rembrance and perpetual observance.

Theories and explanations for the textual problem represented by Codex Bezae at the point of the words of institution in Luke's Gospel have accrued over the centuries. Since the rise of Biblical criticism the passage has been the focal point for textual critics, perhaps most prominently in the epoch making labours of Westcott and Hort who numbered the passage among their 'Western Non Interpolations'.

Virtually no commentator on Luke can, or indeed does, ignore the conundrum but offers some rationale or theory for the authenticity of either longer or shorter ending."

- B. Billings, 'Do This...', Introduction

Billings' book presents eleven whole chapters on this one textual problem, covering the gamut from the textual evidence and theories (Part I), to a thorough discussion of (non)Lukan authorship of the disputed words (Part II), and proceeding to three chapters of sociological analysis, climaxing in a chapter which attempts to answer the question,

Why Codex Bezae was Altered: A Sociological Explanation

Obviously Billings' interest is centered upon the question of the text of Codex Bezae, especially at this point (Luke 22:19a-20).

The Significance of the Textual Evidence

However, our interest is actually in the portion of the problem which he discards, namely, the explanation for and categorization of the Syrian texts in this passage.

For it is in the behaviour of these Syriac texts that we can acquire an understanding of their nature, and from this correctly assess their relevance for the question of the Pericope de Adultera (John 8:1-11).

It is well known that the Syriac texts omit John 8:1-11. However, the standard explanation, namely that the passage was simply unknown and not yet 'inserted' into the Greek version of John is wholly inadequate to explain the bulk of the textual and internal evidence.

But in the proper characterization of the goals and basis of the Syriac versions, we will find the true reason and explanation for omissions like John 8:1-11.

Return to Index

Billings on
Luke 22:19a-20


Billings presents a useful summary of the textual evidence, even though he is wearing the usual blinders imposed by giving undue credit to Westcott and Hort, Metzger, and even Bart Ehrman. The spin can be intelligently ignored.

The Textual Problem

"As G.B. Caird remarked some time ago,

'the Lukan account of the Last Supper is the scholar's paradise and a beginner's nightmare; for it raises problems in almost every department of NT study and has provided a basis for a welter of conflicting theories'.

Before proceeding to the complex matters of form, source, and redaction criticism, or even attempting an exegesis, it is of course necessary by way of prolegomena to first establish the text. Here it is quickly realised that at the heart of all of the difficulties, problems, and conflicting theories surrounding the Lukan institution narrative as a whole lies the perplexing textual problem of Lk. 22:19b-20, a passage that raises according to Joseph Fitzmyer 'the most notorious' textual difficulty in the entire Gospel.

The crux of the difficulty is, succinctly, that although almost all of the manuscript evidence, including that of the extant Greek texts, the ancient versions, patristic citations, and most other witnesses include vv. 19b-20 (the so-called 'longer reading'), a small number of manuscripts follow Codex Bezae-Cantabrigiensis (hereafter D) in breaking off the narrative at 19a and resuming it at v. 21, thus omitting vv. 19b-20 and producing the enigmatic 'shorter reading'.



Chapter 1

The Nature of the Problem

If only the external evidence were taken into view the longer text could not possibly be disputed. All but one of the extant Greek manuscripts of Luke's Gospel and almost all other textual authorities read it, including such weighty witnesses as the 3rd century Bodmer Papyrus (P75) together with codices Sinaiticus (א), Alexandrinus (A), Vaticanus (B), and Ephraemi (C). Among the extant Greek manuscripts of the NT only Codex Bezae-Cantabrigiensis (D), a 5th century bilingual Uncial belonging to the Western 'family' of textual witnesses, reads the shorter text both in its Greek and Latin columns . The reading preserved by D is not followed by the Western family of textual witnesses as a whole but is found only in a few other Old Latin manuscripts ( ita,d,ff2,i,l ).

In addition to the witness of D and its Italic allies the manuscript evidence presents four further , less problematic, variations:

(i) Two more Old Latin manuscripts (itb,e ) modify the shorter reading by placing v. 19a before v. 17 so as to preserve the usual sequence of bread-cup.
(ii) The Curetonian Syriac (Syrc ) also places v. 19a before v. 17 and enlarges the wording with an interpolation from 1st Cor. 11:24.
(iii) The Sinaitic Syriac (Syrs ) is further expansionist, placing both v.19 and v.20a before v.17.
(iv) The Peshitta Syriac (Syrp ) followed also by two Sahidic and one Bohairic manuscript together with l32 lack both v. 17 and v. 18 but preserve the longer reading immediately after v. 19a. 1

The presence of the same textual difficulty represented by D in a second group of independent and early (2nd century) witnesses in the form of the Syriac manuscripts cited above is, in itself, highly significant for this additional set of evidence corrroborates the Western Text, at least to some extent and provides external testimony that the manuscript tradition presented at this point a difficulty that 2nd century scribes have attempted to remedy. 2

This is still the case even if, as Bruce Metzger claims, the four variant witnesses (above) can be accounted for as scribal compromises producing only minor modifications of either the shorter of longer reading. 3

The unanswerable question is whether the Syriac manuscript evidence represents various attempts in antiquity to deal with the problem presented by two versions of the Lukan text (a longer and a shorter text) or of the problem presented by the cup-bread-cup sequence in the longer text only. 4

Despite its great value in alerting us to the unmistakable presence of a textual difficulty in the 2nd century, the Syriac evidence does not confirm the nature or form of this difficulty and cannot, therefore, in its own right establish the authenticity of either the longer or shorter reading. 5

This leaves two principal forms in which the textual tradition of the Lukan Last Supper narrative has come down to us - that of the Majority Text 6 preserving the longer reading and that of D and its few Italic allies preserving the shorter reading.

Original Footnotes:

1 . Bruce Metzger suggests the omission is probably due to homoeoteleuton. , A Textual Commentary... (1994) p. 148.

2 . Johannes Knudsen, 'The Problem of the Two Cups', LQ 2 (1950)pp. 74-85.

3 . B. Metzger, Textual Commentary p. 174. See also the discussion by Felix L. Cirlot, The Early Eucharist (Lon.SPCK, 1939), pp. 242-4; and R. D. Richardson, 'A Further Inquiry into Eucharistic Origins with Special Reference to the NT Problems', in Hans Lietzmann, Mass and Lord's Supper: A Study in the History of the Liturgy (trans. Dor. H.G. Reeve; repr., Leiden: E.J.Brill, 1979),pp.219-701 (229).

4 . John Nolland, Luke (WBC, 35a-c; 3vols; Dallas: Word Books, 1993), vol.3, p. 1041.

5 . Bart Ehrman says of the four variant forms of textual evidence represented above, they 'can readily be dismissed as altogether lacking adequate documentary support and internal claims to authenticity' ('The Cup, the Bread', p. 577).

6 . By 'Majority Text' is meant the consensus of P75 א A B C K L T W supported by almost all miniscules and a large number of ancient versions including the Vulgate, as represented by the text of UBS4.

Return to Index

Removing the Usual Spin

Its hard to believe that in spite of completely inadequate footnotes for the textual section, Billings has managed nonetheless to squeeze in Bruce Metzger (twice, referring to the same tired old commentary, hardly changed from its original 1960's point of view), Bart Ehrman (notorious atheist/propagandist against Christian dogma who has taken over the late Metzger's mantle), and one parting shot against the Majority Text by redefining it as though it were based upon a concensus of Uncials and a Papyrus.

Still, Billings gives a concise overview of the situation regarding the Syriac evidence. And this is really all we are after.

The Textual Omissions: A Visual Look

It will help if we visually diagram the basic situation of the text:

Luke 22:14-23
22:14 And when the hour was come, he sat down, and the twelve apostles with him.
15 And he said to them, "With desire I have desired to eat this passover with you, before I suffer:
16 For I say unto you, I will not any more eat thereof, until it be fulfilled in the kingdom of God.
17-18 And he took the cup, and gave thanks, and said, Take this, and divide it among yourselves: For I say unto you, I will not drink of the fruit of the vine, until the kingdom of God shall come. 17-18 is simply omitted by Peshitta, along with 2 Sahidic & 1 Bohairic MSS, & l32
19a And he took bread and gave thanks, and brake it, and gave unto them, saying, "This is my body" placed before v.17
by Old Latin & Syr c, Syr c also adds words from 1st Cor 11:24. Syr s also moves v. 19b & v. 20a before 17.
19b-20 "...which is given for you: this do in rememberance of me."
(20a) Likewise also the cup after supper, saying,

"This cup is the new testament in my blood, which is shed for you.
19b-20 is simply omitted by
Codex Bezae
21 "But behold, the hand of him that betrayeth me is with me at the table.
22 "And truly the Son of Man goeth, as it was determined: but woe unto that man by whom he is betrayed!"
23 And they began to inquire among themselves, which of them it was that should do this thing.

Billings' explanation for the omission of Bezae (and its two Latin companions) is probably more than adequate. He first fairly presents the majority view of most scholars:

"The paschal context of the disputed words provides a ready solution to the longstanding problem posed by the two cups of Luke 22:17 and 22:20 (a phenomenon unique in the NT to Luke) and to the difficult cup-bread-cup sequence of the narrative, in that the form of the Passover Seder as it was almost certainly celebrated at the time of Jesus involves four cups, at least one of which was rituallistically partaken of prior to the dinner and at least one after it. If this is accepted then both the two cups and the sequence of cup-bread-cup present in the longer reading can be explained.

Pierson Parker, for instance, argues that the first cup in Luke's narrative (Lk. 22:17) is 'not eucharistic in any sense' but belongs to the description of the actual Passover meal that precedes the institution narrative in vv. 14-16. (Parker, 'Three Variant Readings', p.166.).

The presence of the two cups and their sequence might not then be problematic if the original hearers and readers of Luke's Gospel were themselves Jews, proselytes, or Jewish converts to Christianity, or if the Lukan community was sufficiently close in time and cultural milieu to the Jewish origins of the Christian movement to know of the Passover and its structure.

For Gentile converts to the faith, however, the problem of abuses and misunderstandings surrounding that largely Gentile dominated community's celebration of the Lord's Supper has arisen (1st Cor. 10-11) and must be addressed at length by Paul.

In later decades and centuries, as the Christian community was increasingly further removed in time, space, and ethnicity from its Jewish origins, Luke's two cups might have become more and more problematic and perplexing.

This unsatisfactory situation was then overcome by the Gentile scribe of D or one if its ancestors who removed the difficulty altogether by expunging the words of Lk 22: 19b-20 complete with their problematic second cup, at the same time bringing the Lukan account into harmony with the Markan and Matthaean forms, both of which know only the one cup.

Hence, J.H. Petzer concludes:

'a scribe must have omitted these disputed words in order to avoid the difficult cup-bread-cup sequence so that Luke's account of the institution of the Lord's Supper could be harmonized with the other institution narratives.'

- J.H. Petzer, 'Luke 22.19b-20', p. 252.

Billings, Do This... p. 83

Yet Billings presents two significant objections to the status quo solution:

a. Objection One: The Nature of the Lukan Community

... J. Nolland rightly notes, however, that Luke (possibly alone among the NT writers) is a 1st century Gentile who writes for the 1st Century Gentile Church at at time when Christian outreach to Jews was substantially past, even if some Jewish Christians continued to play a leading role in the life of the Christian community. Even if Gentiles attracted to Judaism are the intended audience,...the assumption that they would be familiar with the Passover liturgy to the extent of knowing the number of cups and their sequence is a large and uncertain one.

The view that Jews ...are the primary intended audience..is a minority one, although much has been done in recent decades to recover the very Jewish matrix and milieu from which all of the Gospels sprang. Although the Jews and certainly Jewish Christians must be among the intened audience...'the Jews' in Luke/Acts typically appear in polemic contexts. What appears most probable...is that the Gospel is directed to ...a 'mixed' audience of Jews, Samaritans, and Gentiles. ...the 'mixed communities' conclusion proposed by Eugene La Verdiere seems best:

The Lukan community consisted not in one, but in many communities, whiose members were mainly Gentile origin from the very beginning. Luke wrote...for communities ...established during the great Pauline missions, and for those they themselves founded later.

b. Objection Two: The Liturgy Customary in the Early Church

For the scribe of D, or an earlier exemplar, to be so perplexed by the cup-bread-cup sequence in the Passion Narrative that he excised the troublesome second cup altogether assumes that the liturgy customary surrounding the community meal was both sufficiently fixed and widely known so as to warrant such a substantial emendation. Here the evidence is, however, both diverse and contradictory. Within the NT itself the meal was given different terms - 'breaking of bread' (Acts), 'love feast' (Jude), and 'Lord's Supper' (1st Cor). ...there is considerable variation of practice ..during the early...decades of the Christian community, the very ...time the NT texts were being circulated ...and emended by scribes such as that of Codex Bezae...Ignatius himself refers to the community meal as both 'eucharist' and 'love feast'...the two earliest 2nd century descriptions are the Didache and Justin Martyr's 1st Apology. Here there is considerable fluidity.


The textual difficulty witnessed by D...attracts the thesis that the truncated words ...are an intentional emendation introduced by either the scribe of D or ...an earlier exemplar, to protect what was the most important rite of the Christian community from profanation and/or misunderstanding by outsiders who might have had access to his finished work.

Inadequate attention has, however, been given to the ...possibility that the shorter reading ...serves to safeguard the community cultus against misinterpretation and misunderstanding by outsiders, not for esoteric reasons, but to avoid socially and politically destructive allegations.

Only from the 4th century with the rise of monasticism...does a professional class of scribes ...emerge to assume responsibility for the transmission of the text. Throughout...the 2nd and 3rd centuries manuscripts were produced by and for the use of, religious communities.

...that the dogmatic and theological controversies of the times are reflected in the textual transmission...is already well established,...Manuscripts are however...more than the history of dogmatics. They function as textual mirrors...into all aspects of ...the community for whom they write. ...a window into the social world of early Christianity.

Return to Index

The Syriac Texts
and John 8:1-11

Much of what Billings has said regarding Codex Bezae and its deliberate emendations can also be applied to the Syriac manuscripts and versions.

"Throughout...the 2nd and 3rd centuries manuscripts were produced by and for the use of, religious communities." More specifically, these early manuscripts were prepared for liturgical and religious use.

The omission for instance of Luke 22:17-18 by the Peshitta should be grasped on the basis of the content of the omitted material, since it can hardly be the original text (the Peshitta was the final Syriac text for this period), and the rest of the passage was untouched by the editor of the Peshitta version. What was his purpose?

This editor Eliminated the 'first' cup from the Last Supper along with the pronuncement by Jesus that He would not drink of the 'fruit of the vine' ('oinos'/wine/grapejuice) until the Kingdom of God was (in fullness) come.

This effectively solved two problems simultaneously:

(1) The potential or excuse for 'overdrinking' (or the accusation of same) was eliminated by limiting the drink at the Eucharistic meal to a single cup of wine (whether per meal or per person).

(2) The problematic Nazarite vow is eliminated from the Lukan text. (This is the strongest formulation of a Nazarite vow appearing anywhere in the Bible!) This is desirable for two reasons.

(a) If this text was kept, the expectation would be to follow Jesus' example with similar vows 'until the kingdom come'. This would in effect obseletize and eliminate the 'wine' portion of the eucharist! This would be undesirable to those who wanted to perpetuate a eucharistic ritual with wine.

(b) The command to 'take and divide' is gone, and the second cup can then be interpreted as a 'one time' occurance, not a perpetual eucharistic ritual. This might be very attractive to the Eastern Christians in cultures which abhorred the Greek and Roman traditions of drinking and winemaking, along with its cultic debauchery.

The Syriac Editing Process and John 8:1-11

How does this relate to the omission of the Pericope de Adultera (John 8:1-11) by the Syriac textual traditions?

We have now accumulated more independant evidence that the Syriac texts were edited along the lines of the religious and moral standards of the communities they served.

Its very easy to see now how a passage like John 8:1-11 could be eliminated from the texts, not just to discourage the idea of committing adultery with the expectation of forgiveness, but also with the direct intent of removing the text from liturgical texts used for public worship, where such teachings might be misinterpreted or abused either by members of the community or outsiders looking for excuses to persecute the Christians.

Return to Index