[1] For the sake of simplicity the unknown author of the Epistle of Barrabas is just called Barnabas.

[2] Cfr. von Müller, I. (hg.), Handbuch der klassischen Altertums-Wissenschaft I, 2 Aufl. (München, 1892) 544.Google Scholar

[3] On this and other examples, we Deissmann, A., Light from the Ancient East (Rev. ed., London, 1927Google Scholar [reprinted Grand Rapids, 1978]) 277. Cfr. also Zahn, Th., Die Offenbarung Johannes (Kommentar zum Neuen Testament, Band 18) (1.–3. Aufl., Leipzig/Erlangen, 1926) 458 ff.Google Scholar

[4] From Anthologia Graeca 11, 334. – The numerical value of letters was also used for stylistic purposes, cfr. the collection of epigrams by Leonidas of Alexandria (first century C.E.) in Book 6, 321–329, so-called isopsepha, i.e. poems in which the sum of the numerical values of the letters is identical in each couplet (see Leonidas' explanation in Book 9, 356).

[5] The English translation is quoted from Artemidorus, The Interpretation of Dreams (Oneirocritica), transl. by White, R. J. (Park Ridge, N.J., 1975).Google Scholar

[6] See examples in notes 14 and 27 (from Irenaeus).

[7] Ρωμη = 948 (p = 100, ω = 800, μ = 40, η = 8). The text is quoted from Hennecke, E./Schneemelcher, W. (eds.), New Testament Apocrypha, vol. 1 (London, 1974) 730.Google Scholar – For other examples, we also Perdrizet, P., ‘Isopséphie’, Revue des étudesgrecques 17 (1904) 350–60.Google Scholar

[8] Hebrew from Greek γεωμετρία. On the origin of the word, we Weisskopf, R., Gematria. Buchstabenberechnung, Toro und Schöpfung im rabbinischen Judentum. Dissertation (type-written) (Tübingen, 1975/1978) 7–14 and 276Google Scholar; Sambursky, S., ‘On the Origin and Significance of the Term Gematria’, JJS 29 (1978) 35–8.CrossRefGoogle Scholar

[9] On the method, we Weisskopf (cfr. n. 8) passim; Dornseiff, F., Das Alphabet in Mystik und Magie (Stoicheia, Heft VII) (Leipzig/Berlin, 1922) 91 ff.Google Scholar; Lieberman, S., Hellenism in Jewish Palestine (New York, 1950) 69 ff.Google Scholar; Strack, H. L./Stemberger, G., Einleitung in Talmud und Midrasch (Siebente, völlig neu bearbeitete Auflage, München, 1982) 39.Google Scholar

[10] Cfr. also bTaan 17a; bSanh 22b.

[11] This interpretation occurs several times in Talmud and Midrash, e.g. bNed32a; GenR 43,2; 44,9; LevR 28,4. It is also found in Targum Ps.Jonathan. – It first occurs in the name of Bar Kappara.

[12] So Helm, L., Studien zur typologischen Schriftauslegung im zweiten Jahrhundert. Barnabas und Justin. Dissertation (typewritten) (Frankfurt a.M., 1970) 45.Google Scholar

[13] Bowker, J., The Targums and Rabbinic Literature. An Introduction to Jewish Interpretation of Scripture (Cambridge, 1969) 195.CrossRefGoogle Scholar

[14] ι = 10, η = 8, σ = 200, ο = 70, ν = 400, ς = 200. This interpretation is found e.g. in Irenaeus, Adv.haer. 1,15,2 (referring to the gnostics) and Or.Sib. 1, 324 ff. – Notice that not even the Hebrew name for Jesus has the value 318.

[15] Cfr. Rabinowitz, L. I., ‘The Study of a Midrash’, JQR 58 (1967–1968) 143–61.CrossRefGoogle Scholar

[16] So the article on ‘Gematria’ in Encyclopaedia Judaica 7, col. 370.

[17] On this combination, we below note 37.

[18] So Weisskopf, op. cit. (cfr. n. 8) 194; Helm, op. cit. (cfr. n. 12) 45; Strack/Stemberger, op. cit. (cfr. n. 9) 39; Metzger, B. M., Manuscripts of the Greek Bible. An Introduction to Greek Palaeography (New York/Oxford, 1981) 9Google Scholar; Kraft, R. A., Barnabas and the Didache (The Apostolic Fathers, vol. 3) (New York, 1965) 107Google Scholar; Wengst, K., Didache (Apostellehre), Barnabasbrief, Zweiter Klemensbrief, Schrift an Diognet (Schriften des Urchristentums II) (Darmstadt 1984) 130.Google Scholar

[19] This method resembles notarikon another hermeneutical rule found in the thirty-two middoth. Notarikonhas no interest in the numerical value of the letters, but each letter in a word can be treated as an abbreviation for another word. See Strack/Stemberger, op. cit. (cfr. n. 9) 39 and the article on ‘Notarikon’ in Encyclopaedia Judaica 12, cols. 1231 f.

[20] This interpretation of the name Isaac is found e.g. in BerR 53,7.

[21] Weisskopf, op. cit. (cfr. n. 8), 179. And he goes on: ‘So wird ’ in BerR 53, 7 auf die 10 Gebote, die 90-jährige Sara, den 100-jährige Abraham und die nach 7 Tagen erfolgende erstmalige Beschneidung eines Abrahamsnachkommen bezogen, alles Daten, die unmittelbar mit der Gestalt Isaaks zusammenhängen. Die 10 Gebote als Grundlage des Wandels vor Gott in der Welt sind dabei keine Ausnahme, denn sie waren nach rabbinischer Meinung auch schon Abraham bekannt. Sie stehen am Anfang der Aufzählung und wollen gewissermassen als Vorzeichen dem folgenden voran-stehen.’ (ibid.) – The use of gematria as an exegetical method is well summarized in these words: ‘Jedoch fällt auf, dass an keiner Stelle wo Gematrie auftritt, ein grundsätzlich neuer Gedanke in die Haggada eingeführt wird. Die Gematrie hat durchweg bestätigende und verfeinernde Funktion.’ (The report on Weisskopf±s dissertation, ThLZ 105 (1980) 637.)Google Scholar

[22] This connection is of course noted before, e.g. by Roberts, C. H., Manuscript, Society and Belief in Early Christian Egypt (London, 1979) 35 f.Google Scholar

[23] On this subject, see Traube, L., Nomina Sacra. Versuch einer Geschichte der christlichen Kürzung (München, 1907)Google Scholar; Paap, A. H. R. E., Nomina Sacra in the Greek papyri of the first five centuries A.D. (Leiden, 1959).Google Scholar A brief survey is given in Metzger, op. cit. (cfr. n. 18) 36 f.

[24] The examples given are the most common abbreviations of the nominative form.

[25] Paap, op. cit. (cfr. n. 23) 107 mentions 7 sources with 48 instances. It occurs for the first time in P. Lond. Christ. 1 (Egerton P. 2), in the unknown gospel from Oxyrhynchus (?), dated ± 150 A.D.

[26] See C. H. Roberts, op. cit. (cfr. n. 22) 26–48 and Brown, S., ‘Concerning the Origin of the Nomina Sacra’, Studia Papyrologica 9 (1970) 7–19.Google Scholar

[27] On the number 18 (ιη) as a reference to the name Jesus, we Irenaeus, , Adv. haer. 1Google Scholar, 3, 2 and Hippolytus, , Comm. Dan. 2, 27.Google Scholar

[28] Cfr. e.g. Lucian of Samosata, , Judicium vocalium 12Google Scholar and Pseudo-Cyprian, , De pascha computus 10Google Scholar, 18, 20, 22. See also Dinkier, E., Signum Crucis (Tübingen, 1967) especially 26 ff.Google Scholar

[29] Hatch, E., Essays in Biblical Greek (Amsterdam, 1970) (first printed 1889) 155.Google Scholar

[30] According to Wevers, J. W. (ed.), Genesis (Septuaginta, vol. I) (Göttingen, 1974) 164Google Scholar, there are 8 manuscripts with the reading τιη, but only the two mentioned in the text are of any interest. The other six manuscripts date from 12.–15. century.

[31] Published in Oates, J. F./Samuel, A. E./Welles, C. B., Yale Papyri in the Beinecke Rare Book and Manuscript Library 1 (New Haven/Toronto, 1967) 3–8.Google Scholar See also Dinkier, E., ‘Papyrus Yalensis 1 als ältest bekannter christlicher Genesistext. Zur Frühgeschichte des Kreuz-Symbols’, ZNW 73 (1982) 281–5.Google Scholar

[32] According to the editor, C. B. Welles, it dates from ‘Ca. A.D. 90’ (op. cit. (cfr. n. 31) 3). C. H. Roberts, op. cit. (cfr. n. 22) 13 ‘would be reluctant to date the papyrus before the middle of the [second] century’.

[33] This reading is only found in a late LXX-manuscript from the 12. century, but the manuscript D (Cotton Genesis from the 5. century) is very close to Barnabas' reading: δέκα καί όκτὼ καί τριακοίους. – The common assertion that Barnabas argues from the Hebrew text is well refuted by Wengst, K., Tradition und Theologie des Barnabasbriefes (Berlin/New York, 1971) 68.Google Scholar

[34] Windisch, H., Die apostolischen Väter III: Der Barnabasbrief (Handbuch zum Neuen Testament. Ergänzungs-Band) (Tübingen, 1920) 357.Google Scholar

[35] Kraft, op. cit. (cfr. n. 18) 107 f. Cfr. also Wengst, op. cit. (cfr. n. 18) 125.

[36] Kraft, R. A., The Epistle of Barnabas, its Quotations and their Sources, Dissertation (micro-film) (Harvard, 1961) 196.Google Scholar – That Clement quotes from Gen 14 is not surprising, since his sole interest is the number 318 – related to Abraham's knowledge of arithmetic. – It is also interesting to see that in both the passages (chap. 10 and 22) where Pseudo-Cyprian mentions the number 318, he stresses that Abraham was a hundred years old – but this is an information from Gen 17(!) – I think both these observations weaken Kraft's argument.

[37] What was the reason for this combination? This is difficult to know, and my answer is nothing more than a tentative solution: The combination was part of an argumentation against carnal circumcision in one of Barnabas's sources. This source is reflected in 9. 6 where Barnabas downgrades the importance of circumcision by referring to the fact that also Syrians, Arabs and Egyptians are circumcised. And maybe someone had put Gen 17. 23 and 14. 14 together to show that a lot of people were circumcised without belonging to God's own people – an argument along the lines of Romans 9. 7 f.: Not all who are circumcised on the body are the children of God, but only those who are circumcised on the heart. Even if Abraham's 318 men were circumcised, the Scripture says that only Abraham was counted righteous.