¹ The Oxyrhynchus Papyri [= P.Ox.] have been published in many parts since 1898 (part 25 in 1959) by B. P. Grenfell and A. S. Hunt, et al. An enormous literature has grown up around the material which is directly relevant to the study of Jesus’ “unwritten” sayings (agrapha); see the extensive bibliography in J. A. Fitzmyer, “The Oxyrhynchus Logoi of Jesus and the Coptic Gospel According to Thomas,” Theological Studies 20 (1959), pp. 555–560.

² Photographic plates of the Gospel [= G.Thom.] were published byLabib, P., Coptic Gnostic Papyri in the Coptic Museum at Old Cairo I (1956)Google Scholar, plates 80:10–99:28. Unless otherwise indicated, the numbering of the logia and their translation (with slight adaptations) are taken from A. Guillaumont, et al., The Gospel according to Thomas (1959).

³ In the following discussion, reference will be made to several authors who have contributed in one way or another to the study of P.Ox. 655. For the convenience of the reader, we will list here (in alphabetical order according to author) that literature; henceforth only the name of the author (or initials in the case of Grenfell and Hunt) will be cited to indicate the source in which the contribution is found:

Prof. Allen [of Oxford], Note in the Guardian of July 24, 1904 [quoted by P. Batiffol in Revue Biblique 1 (1904), p. 491];

Fitzmyer, J. A., “The Oxyrhynchus Logoi of Jesus and the Coptic Gospel According to Thomas,” Theological Studies 20 (1959), pp. 543–551CrossRefGoogle Scholar;

R. M. Grant and D. N. Freedman, The Secret Sayings of Jesus (1960), pp. 51–52;

Grenfell, B. P. and Hunt, A. S.[ = G-H], Oxyrhynchus Papyri IV (1904), pp. 22–28Google Scholar; G.

Heinrici, , “Die neuen Herrensprueche,” Theologische Studien und Kritiken 78 (1905), pp. 205–209Google Scholar;

Hilgenfeld, A., “Neue gnostische Logia Jesu,” Zeitschrift fuer Wissenschaftliche Theologie 47 (1904), pp. 567–573Google Scholar;

O. Hofius, “Das Koptische Thomasevangelium und die Oxyrhynchus-Papyri Nr. 1, 654 und 655; II,” Evangelische Theologie 20 (1960), pp. 189–192;

Kasser, R., “Les Manuscrits de Nag ‘Hammâdi,” Revue de Théologie et de Philosophie 9 (1959), 351–370Google Scholar;

Katz, P., “ΠΩΣ ΑϒΞANOϒΣIN. Matt. VI. 28,” Journal of Theological Studies 5 (1954), pp. 207–209Google Scholar;

E. Klostermann, Apocrypha II: Evangelien (Kleine Texte 6, 1929³,) pp. 23–24;

Michelsen, J. A. H., “Nieuw-ontdekte fragmenten van evangeliën,” Teyler's Theologisch Tijdschrift 3(1905), pp. 161–164Google Scholar, and “Uittreksels uit het Evangelie volgens Thomas,” ibid. 7(1909), pp. 214–233.

Osborn, G., [short note], Journal of Theological Studies 32 (1930/1931), p. 179Google Scholar;

H. C. Puech, “Gnostische Evangelien und verwandte Dokumente: Das Thomas-Evangelium,” in Neutestamentliche Apokryphen I: Evangelien, ed. E. Hennecke and W. Schneemelcher (1959³), pp. 199–223;

Skeat, T. C., “The Lilies of the Field,” Zeitschrift fuer die Neutestamentliche Wissenschaft 37 (1938), pp. 211–214Google Scholar;

A. de Santos Otero [ = Santos], Los Evangelios Apocrifos (1956), pp. 81–83;

W. Schneemelcher, “Papyrusfragmente apokrypher Evangelien: Oxyrhynchos-Papyrus 655,” in Neutestamentliche Apokryphen I [see above under Puech], pp. 70–72;

C. Taylor, The Oxyrhynchus Sayings of Jesus (1905), pp. 18–23;

Zahn, T., “Neue Funde aus der alten Kirche,” Neue Kirchliche Zeitschrift 16 (1905), pp. 96–101Google Scholar.

The relationship between P.Ox. 655 and G.Thom. is discussed by Fitzmyer, Grant-Freedman, Hofius, Kasser, Puech, and Schneemelcher. Most recently, G. Garitte, “Les ‘Logoi’ d'Oxyrhynque et l'apocryphe copte dit ‘Évangile de Thomas',” Le Muséon 73 (1960), pp. 151–172, has challenged the assumption that G.Thom. basically is a translation from a Greek original which was roughly identical with the P.Ox. parallels, and suggests, per contra, that P.Ox. 654, 1, and 655 are translations from a Coptic Vorlage like G.Thom. (he leaves open the question as to whether G.Thom. originally was written in Greek or in Coptic). A. Guillaumont, “Les Logoi d'Oxyrhynchos sont traduits du copte?” ibid., pp. 325–333, considers Garitte's hypothesis to be improbable, especially in the light of the complex textual situation involved (note that the P.Ox. fragments are from three different manuscripts and do not parallel G.Thom. exactly at every point, etc.), but Garitte's answer (”Les ‘Logoi’ d'Oxyrhynque sont traduits du copte,” ibid., pp. 335–349) attempts to reaffirm and strengthen his earlier claims.

⁴ Grenfell and Hunt [ = G–H]. It should be noted, however, that already in 1909, Michelsen had suggested that all three of the P.Ox. fragments had come from a lost G.Thom. P.Ox. 655 now is in the possession of the Semitic Museum at Harvard University, and is housed at Houghton Library in Cambridge, Mass., under the catalogue number SM 4367.

⁵ Fragments (f) and (h) apparently have been lost in subsequent years, for they were not with the other portions of P.Ox. 655 when Mr. Roy B. Ward and the author recently remounted the papyrus for the Semitic Museum. The content of these smaller scraps is, according to the transcriptions of G–H, the letters KA in (f), K and AI in the two lines of (g), and Є in (h).

⁶ This is a common orthographical variant and helps to lengthen lines 3–6, which otherwise would be unusually short in the reconstruction. The difference between ЄI and H is very difficult to detect in P.Ox. 655, since the vertical stroke of the Є has very little curvature.

⁷ See Klostermann, p. 23; Santos, p. 82; and Grant-Freedman, p. 51. Other attempts at filling the lacunae left by G-H in lines 10 and 12 include Hilgenfeld's ǫτι … ὲνδὐεσθε, Heinrici's άλλ’ … ένδεῖσθε, and Michelsen's τò ἔν ἔχοντων … ένδεές (1909).

⁸ See also the discussion of Katz. Actually, Michelsen (1909) already had suggested the same correction on the basis of the “western” variant to Luke 12:27, although he could not know of the support offered by the original hand of Sinaiticus.

⁹ In lines 13–16, Fitzmyer follows the parallel from the Ps-Clem. Homily 18 for the verbs άπέλαβoν … άπέκρυψαν (see also Michelsen in 1905); but com pare Codex Bezae (et al.) in Luk e 11:52.

¹⁰ Hofius does not attempt to restore col. i, lines 24–30, and Fitzmyer differs from our reconstruction of that section in the following particulars:

lines 24 ff.; τοὑς χιτῶνας … αὐτοὐς … αὐτοὐς.

Other alternatives are ὶμάτια and ένδὐματα. We have chosen the latter (in the singular as collective, although the Coptic is plural; for a similar use of the collective singular, see W. Dittenberger, Sylloge Inscriptionum Graecarum III [1920³], p. 170 [1023:29 ff.] — παραγένωνται … τò ὄνομα πατριαστί) because it is more likely as a Stichwort and it occurs in the parallel from Clement of Alexandria.

lines 27 f.; τóτε γενήσεσθε υἱοὶ τοῦ ζῶντς καὶ οὐ μὴ ϕοβηθήσεσθε.

The Coptic has a lacuna here for the first verb, but the word “son” is definitely in the singular, and the context of the disciples’ question seems to support the translation adopted in this paper (despite the agreement of Schoedel's translation in Grant-Freedman with Fitzmyer's).

¹¹ Hofius also errs in reading col. ii, line 2 as ΛO (it is unmistakably ΛЄ, and there is no similarity between O and Є in the handwriting of the papyrus), nor can the second letter of line 6 be O (it is almost positively A). Throughout his restorations of the three P.Ox., Hofius seems to pay little attention to the average number of letters per line; certainly there can be some flexibility here, but not the extremes found in his article!

¹² In fragment (b), the last lines of each column correspond to each other, and thus one would expect that the same number of lines preceded in each column. Nevertheless, G-H have 23 lines in the transcription of the extant fragments from col. i, but only allow for 22 lines in col. ii.

¹³ Grant-Freedman, p. 52 n. 4. This identification is possible, but it is by no means positive. The main problem is how to construct reasonable Greek around the dative/locative adjective, ϕωτεινῷ (assuming that the copulative verb occurs at the end of the preceding line), on the basis of the Coptic context. Kasser, p. 367 n. 1, makes the following attempt: ЄΙ ΦΩC ЄΤСΙΝ / ЄΝ ΑΝΘΡΩΠΩ ΦΩΤЄΝΩ / ЄΝ ΟΛΩ ΤΩ ΚΟСМΩ / ΦΩΤΙΖЄΙ. ЄΙ ΔЄ ΜΗ / СΚΟΤЄΙΝΟС ЄСΤΙΝ.

¹⁴ G-H thought that the letters of (e) came from the beginning of a line, but the lacuna to their left is close enough to allow some letters to precede them. (They also might possibly be from the end of a line which extended into the right margin, and thus not from the bottom of a column at all!) From an examination of the fiber grains of the papyrus fragments themselves, there is no objection to the possibility that (d) and (e) represent the bottom of the same column. At the end of the next-to-the-last line of (d) there is a perplexing little apostrophe-like hook for which no satisfactory explanation has been found.

¹⁵ Possibly through misreading G-H, Santos has mistakenly included fragment (f) after (d) at the bottom of col. i, and has placed (h) after (e) at the bottom of col. ii. Since (d) and (e) themselves appear to be from the bottom of a column (or columns), this is impossible. Apparently Puech (p. 213) based his comments concerning the external relationship between G.Thom. and P.Ox. 655 on the erroneous order into which Santos had placed the fragments; Schneemelcher's comments on p. 72 betray a similarly misplaced confidence in Santos’ order, and Hofius (p. 189) also seems to have been misled in this regard.