² Cf. e.g. Metzger, B. M., A Textual Commentary on the Greek New Testament (United Bible Societies, 1971) xxviiGoogle Scholar; Elliott, J. K., The Greek Text of the Epistles to Timothy and Titus (Salt Lake City: University of Utah Press, 1968, Studies and Documents 34) 7–8Google Scholar; Aland, K. & Aland, B., Der Text des Neuen Testaments (Stuttgart: Deutsche Bibelgesellschaft, 1982) 282Google Scholar, to mention only three of the more influential methodologies currently in use.

³ Cf. e.g. the following criticism by Elliott, J. K., ‘The United Bible Societies' Greek New Testament: An Evaluation’, NT 15 (1973) 295–7Google Scholar; ‘The United Bible Societies’ Textual Commentary Evaluated’, NT 17 (1975) 142–4Google Scholar; ‘An Eclectic Commentary on the Greek Text of Mark's Gospel’ in G. D. Fee & E. J. Epp (eds.), New Testament Textual Criticism: Its Significance for Exegesis. Essays in Honour of Bruce M. Metzger (Oxford: Clarendon, 1981) 47–60Google Scholar; ‘The Text of Acts in the Light of Two Recent Studies’, NTS 34 (1988) 252–7Google Scholar; Kilpatrick, G. D., ‘The Greek New Testament Edited by Aland, K., Black, M., Martini, C. M., Metzger, B. M. and Wikgren, A. (1976).Google ScholarMetzger, B. M.: A Textual Commentary on the Greek New Testament (1971)’, TLZ 104 (1979) 267Google Scholar; De Jonge, H. J., ‘De Nieuwe Nestle: N²⁶’, Nederlands Theologisch Tijdschrift 34 (1980) 312.Google Scholar

⁴ Cf. e.g. the discussions at Jn 9. 38–39a; Lk 22.19b–20; Rm 14. 23; Mk 5. 22; and the note in brackets at Mk 10. 2 in Metzger, Textual Commentary.

⁵ Cf. e.g. Metzger, , Textual Commentary, 176Google Scholar; Greijdanus, S., Het Evangelie naar de Beschrijving van Lucas (v. 2, Amsterdam: Bottenburg, 1941; Kommentaar op het Nieuwe Testament) 149Google Scholar; Burkitt, F. C., ‘On Luke 22. 17–20’, JTS 28 (1927) 180Google Scholar; Kilpatrick, G. D., ‘Luke 22. 19b–20’, JTS 47 (1946) 51Google Scholar; The Eucharist in Bible and Liturgy (Cambridge: Cambridge University Press, 1983) 31–2Google Scholar; Chadwick, H., ‘The Shorter Version of Luke 22. 15–20’, HTR 50 (1957) 252CrossRefGoogle Scholar; Marshall, I. H., The Gospel of Luke: A Commentary on the Greek Text (Exeter: Paternoster, 1978) 800Google Scholar; Schürmann, H., Traditionsgeschichtliche Untersuchungen zu den synoptischen Evangelien (Düsseldorf: Patmos, 1968, Kommentare und Beiträge zum Alten und Neuen Testament) 172, 177Google Scholar; Taylor, V., The Passion Narrative of St. Luke. A Critical and Historical Investigation (Cambridge: Cambridge University Press, 1972, SNTS Monograph Series 19) 55Google Scholar; Jeremias, J., Die Abendmahlsworte Jesu (2nd ed., Göt-tingen: Vandenhoeck & Ruprecht, 1949) 76–7Google Scholar; Rese, M., ‘Zur Problematik von Kurz- und Langtext in Lk 22. 17 ff.’, NTS 22 (1976) 30–1Google Scholar; Zahn, T., Das Evangelium des Lucas (1st & 2nd ed., Leipzig: Deichert, 1913, Kommentar zum Neuen Testament) 676–7Google Scholar; Stagg, F., ‘The Lord's Supper in the New Testament’, Review and Expositor 66 (1969) 12.CrossRefGoogle Scholar

⁶ In addition to the presence of non-Lucan features, it has also been said that there are no typical Lucan features (‘lucanisms’) present in the disputed passage, cf. Taylor, , Passion Narrative, 55Google Scholar; Rese, , Problematik, 30.Google Scholar

⁷ Cf. Burkitt, , On Luke 22, 180Google Scholar; Rese, , Problematik, 15–31Google Scholar; Chadwick, , Shorter Text, 249–56Google Scholar; and in particular Kilpatrick, Luke 22. 19b-20, 51, who, after listing five non-Lucan features concludes that ‘… we have a considerable number of features either foreign to or unusual in Luke's style …’, which he takes as ‘… weighty against the authenticity of the passage’, cf. also by the same author, The Eucharist, 31–2.

⁸ Cf. e.g. Jeremias, , Abendmahlsworte, 77Google Scholar; Stagg, , Lord's Supper, 12Google Scholar; Schürmann, , Untersuchungen, 177Google Scholar; Marshall, , Gospel of Luke, 800Google Scholar; Taylor, , Passion Narrative, 57.Google Scholar

⁹ The analysis of style is approached in the traditional way in this article, i.e. by analysing the occurrences of words or grammatical constructions in the disputed passage elsewhere in the book under discussion. It follows in broad terms analyses of style such as that of Elliott, J. K., ‘The Text and Language of the Endings to Mark's Gospel’, TZ 27 (1971) 255–62Google Scholar; ‘The language and style to the concluding doxology to the Epistle to the Romans’, ZNW 72 (1981) 124–30Google Scholar; Fee, G. D., ‘On the Inauthenticity of John 5. 3b-4’, EQ 54 (1982) 207–18Google Scholar; Metzger, , Textual Commentary. Though this definition of and approach to style is very different from recent definitions of and approaches to style in other disciplines – disciplines upon which the text-critical use of style is based – it is nevertheless followed in this article with the intent of illustrating one of the most basic problems of this traditional text-critical use of author's style.Google Scholar

¹⁰ In the discussion of the features no attention is given to the question whether the feature under observation is or is not a non-Lucan feature. Departing from a number of features identified by other scholars, it is rather attempted to create some kind of standard, by means of which the context of the Lucan narrative and other narratives in the Greek New Testament could be analysed, in an attempt to create a basis of comparison. The purpose of the article is therefore not to attempt to prove that the style of the disputed passage is Lucan (or rather not non-Lucan), but rather to put the importance of the occurrence of those non-Lucan features into a relative perspective.

¹¹ It occurs as a variant for περί at Lk 6. 28 and Acts 12. 5 and 26.1.

¹² It is difficult to decide whether one should consider the evidence from Acts as well when analysing Lucan style in the Gospel. In this case, as well as in some other cases below, it seems that there might be some discrepancy between the style of the Gospel and the style of Acts. Without going into this problem, it needs to be pointed out that this problem has in the past led scholars to believe that these two books were not written by the same author. Cf. e.g. Clark, A. C., The Acts of the Apostles (Oxford, 1933) 393 ff.Google Scholar; and more recently the article by Argyle, A. W., ‘The Greek of Luke and Acts’, NTS 20 (1974) 441–5CrossRefGoogle Scholar, wherein he lists a number of differences as to vocabulary between these two books. For the sake of the discussion of the style of Lk 22. 19b-20 in this article, all the features which seem non-Lucan as regards the Gospel have been discussed below, with a note as to their occurrence in Acts.

¹³ Jeremias, , Abendmahlsworte, 76; Kilpatrick, Luke 22. 19b-20;The Eucharist, 31; Taylor, Passion Narrative, 55.Google Scholar

¹⁴ That is if λογούς is regarded as the original reading here. If λογούς is not original the situation changes and it leaves έμός in the disputed passage as the only instance where the word is used as attribute. Cf. Jeremias, , Die Abendmahlsworte, 76–7Google Scholar; Kilpatrick, , Luke 22. 19b-20, 51Google Scholar; The Eucharist, 31Google Scholar; ‘The Possessive Pronouns in the New Testament’, JTS 42 (1941) 185–6Google Scholar; Taylor, , Passion Narrative, 55.Google Scholar

¹⁵ The same is in broad terms true of σός and σο, the first being used only seven times, four times in the Gospel, always substantively (5. 33; 6. 30; 15. 31; 22. 42), and three times in Acts (5. 4; 24. 2, 4).

¹⁶ It should be noted that the question whether this is a non-Lucan construction or not, hinges on the status of λογούς, which is text-critically problematic in both Mt 8. 38 and Lk 9. 26. If the short reading is preferred in Lk 9. 26, it makes the occurrence in Lk 22. 19 unparalleled and thus non-Lucan, since the only similar construction is then removed.

¹⁷ Jeremias, , Die Abendmahlsworte, 76Google Scholar; Kilpatrick, , Luke 22. 19b-20, 51Google Scholar; The Eucharist, 31Google Scholar; and Taylor, , Passion Narrative, 55 refer to Acts 10Google Scholar. 4, where μνημοσύνη is used in a similar context.

¹⁸ Some critics regard the occurrence of this word with the article in the disputed passage as non-Lucan, because of the fact that both ποτήριον and ἄρτος in vs. 17 and 19a are anarthrous and it is therefore to be expected that the same construction will appear here in vs. 19, if Luke were to have written these words. Cf. Taylor, , Passion Narrative, 55Google Scholar; Jeremias, , Abendmahlsworte, 77Google Scholar; Zahn, , Lucas, 676–7Google Scholar. Cf. also Plummer, A., A Critical and Exegetical Commentary on the Gospel according to S. Luke (5th ed. 3rd imp., Edinburgh: T & T Clark, 1942) 496 for a discussion of the use of the article in these passages.Google Scholar

¹⁹ Cf. also Kilpatrick, , Luke 22. 19b-20, 51Google Scholar, who calls this an unusual expression in Luke, cf. also Kilpatrick, , The Eucharist, 31.Google Scholar

²⁰ It also occurs in Lk 13. 3 as a variant for όμοίως in some manuscripts. In 13. 5 όμοίως occurs as a variant in some manuscripts.

²¹ Kilpatrick, , The Eucharist, 31.Google Scholar

²² Cf. Jeremias, , Abendmahlsworte, 77Google Scholar; Chadwick, , Shorter Text, 252Google Scholar; Kilpatrick, , Luke 22. 19b-20, 51Google Scholar; The Eucharist, 31Google Scholar; Greijdanus, , Lucas, 149Google Scholar; Burkitt, , On Luke 22, 180Google Scholar; Taylor, , Passion Narrative, 55.Google Scholar

²³ Cf. Taylor, , Passion Narrative, 55Google Scholar; Zahn, , Lucas, 676, 677Google Scholar; Jeremias, , Abendmahlsworte, 76Google Scholar; Kilpatrick, , The Eucharist, 32Google Scholar; Chadwick, , Shorter Text, 252Google Scholar, who incidentally points out that this argument carries little weight, since ‘… one can pour out either a cup or its contents’.

²⁴ If the occurrence of the word in Mk 2. 22 proves to be original, Luke could in this instance of course also be dependent upon Mark or a manuscript of Mark which contained this word.

²⁵ This, the determination of what exactly a non-Lucan feature is, is actually one of the methodological problems in the use of this criterion. There seems to be no pattern in the determination of a non-Lucan feature, and what might be non-Lucan to one person will not necessarily be non-Lucan to another. This makes the whole criterion very arbitrary and it is another aspect of this criterion which needs improvement.

²⁶ The analysis is limited to that part of the narrative which might have liturgical significance and not the whole of the narrative. This is done because the disputed passage stands in a highly liturgical context and in order to compare apples to apples, it was thought best to limit the context to that part which might also be of liturgical significance. Therefore, not the whole of the narrative (Lk 22. 14–23), but only the actual eucharistic narrative is analysed.

²⁷ See discussion on p. 117 above.

²⁸ In both Matthew and Mark's narratives some manuscripts have the word added to διαθήκη in a similar construction as the Lucan and Pauline narratives. This is probably due to harmonisation to Luke and Paul's story. The fact of harmonisation here seems to imply that the Lucan/Pauline expression became the dominant expression in the early Christian liturgical eucharist language.

²⁹ It also occurs in certain manuscripts in Mk 2. 22, the parallel for Lk 5. 37.

³⁰ Also in Mk 16. 17, which is not considered part of Mark for the purpose of this article.

³¹ The closeness of the words of Lk 22. 19b-20 to 1 Cor 11. 24–25 has the result that the critics in favour of the short reading think that the latter might be the source of the interpolation of these disputed words and that Lk 22. 19b-20 is a harmonisation to the narrative in 1 Corinthians. Against this it has been shown by Schürmann (amongst others) that there are a number of important linguistic differences between the Lucan and Pauline narratives which rule harmonisation out of the question. The language of these two narratives are consequently related to an earlier ecclesiastical source of some kind (written or oral), used by both Luke and Paul. Cf. e.g. Schürmann, Untersuchungen, 177; E. Schweizer, Das Evangelium nach Lukas (Göttingen: Vandenhoeck & Ruprecht, 1982, Das Neue Testament Deutsch); Marshall, Gospel of Luke, 800; Taylor, Passion Narrative, 56–7; Jeremias, Abendmahlsworte, 77; also G. D. Fee, The First Epistle to the Corinthians (Grand Rapids: Eerdmans, 1987, The New International Commentary on the New Testament) 546–7. This notion is strengthened by the fact that the Matthean and Marcan narratives also contain language which might point to a pre-Matthean and pre-Marcan source for those two narratives. The only way in which it could be explained how it is possible that all four of the narratives show signs of such archaic language and that some of the same archaic features recur in each narrative, is to relate it to the same kind of source, which can only be a liturgical source, wherein the basic aspects of the narrative will be the same. The differences in the four narratives could be accounted for by the fact that they probably go back to different liturgical traditions, wherein the basic facts were the same, but which differed as to some minor aspects of the liturgy.