¹ For the attempts to analyze the plan and structure of Celsus's work, see Pichler, Karl, Streit um das Christentum: Der Angriffdes Kelsos und die Antwort des Origenes (Regensburger Studien zur Theologie 23; Frankfurt a. M./Bern: Lang, 1980) 15–24Google Scholar . On this section and its context in Celsus's polemic, see Andresen, Carl, Logos und Nomos: Die Polemik des Kelsos wider das Christentum (Arbeiten zur Kirchengeschichte 30; Berlin: de Gruyter, 1955) 221–23CrossRefGoogle Scholar ; Borret, Marcel, Origène, Contre Celse V (SC 227; Paris: Cerf, 1976) 105, 108–10Google Scholar ; , Pichler, Streit um das Christentum, 170—76Google Scholar.

² I cite the edition of Borret, Marcel, Origène, Centre Celse IV (SC 150; Paris: Cerf, 1969) 206.4–8Google Scholar; it differs slightly from that of Koetschau, Paul, Origenes Werke II (GCS 3; Leipzig: Hinrichs, 1899) 232.12–16Google Scholar. The translation is my own, reflecting an emendation which I shall argue below.

³ Celsus's date is usually set ca. 180 CE. For literature about this fragment, see, e.g., Chadwick, Henry, Origen: Contra Celsum (Cambridge: Cambridge University Press, 1965) xxvi–xxviiGoogle Scholar; Rosenbaum, H.-U., “Zur Datierung von Celsus' Λληθής Λγος” VC 26 (1972) 102–11Google Scholar; , Borret, Origène, Centre Celse V, 128–29Google Scholar; , Pichler, Streir um das Christentum, 94–97Google Scholar; Burke, Gary Tapp, “Celsus and Late Second-Century Christianity” (Ph.D. diss., University of Iowa, 1981) 49-60, 197–98Google Scholar.

⁴ Contra Celsum 6.19 ( Borret, Marcel, Origèn, Contre Celse III [SC 147; Paris: Cerf, 1969] 224.1–3Google Scholar; , Koetschau, Origenes Werke II, 89.18–20)Google Scholar.

⁵ Theodor Keim is consequently wrong in asserting that Celsus has misunderstood the excerpt in equating its κρατν θες with the ὐπερουράνιος θεός. See , Keim, Kelsosl Celsus: Wahres Wort. Älteste Streitschrift antiker Weltanschauung gegen das Christentum vom Jahr 178 n. Chr. (1873Google Scholar; reprinted Aalen: Scientia, 1969) 123 n. 1, followed by Rohm, Johann, Des Kirchenschriftstellers Origenes acht Bücker gegen Celsus (Bibliothek der Kirchenväter 60; 3 vols.; Kempten: Kösel, 1874-1877) 3. 426 n. 4Google Scholar. They make the same assertion of Origen, but in fact his “misunderstanding” (Cels. 8.15 [Borret, Origène, Contre Celse IV, 206.22-208.32; Koetschau, Origenes Werke II, 233.6-15]) is as purposeful and polemically driven as Celsus's, for it enables him to reject the relevance of the document. Cels. 8.16 ( , Borret, Origène, Contre Celse IV, 208.8–15Google Scholar; , Koetschau, Origenes Werke II, 234.2–9Google Scholar) shows that Origen understands full well what the excerpt actually intends. On Celsus's philosophical conception of God, heavily influenced by Plato and shared by a majority of contemporary intellectuals, see, e.g., Glðckner, Otto, “Die Gottes- und Weltanschauung des Celsus,” Philologus 82 (1927) 329–52Google Scholar; Festugière, A.-J., La révélation d'Hermès Trismégiste IV: Le Dieu inconnu et la Gnose (3d ed.; Paris: Gabalda, 1954) 92–140Google Scholar; Dörrie, Heinrich, “Die platonische Theologie des Kelsos in ihrer Auseinandersetzung mil der christlichen Theologie auf Grund von Origenes c. Celsum 7, 42ff,” Nachrichten von der Akademie der Wissenschafien zu Göttingen, phiiologisch-historische Klasse (1967) 19–55Google Scholar(reprinted in idem, Platonica Minora, 229-62).

⁶ Already Mosheim, Johann Lorenz (Origenes, Vorstehers der christlichen Schule :u Alexandrien und Aeltestens, acht Bücher von der Wahrheit der christlichen Religion wider den Weltweisen Celsus [Hamburg: Bohn, 1745] 825 n. 4)Google Scholar senses something fishy in Origen's claim of ignorance of the doctrines and sect represented by the “Celestial Dialogue.” The section of Origen's response summarized here (still Ceis. 8.15) is , Borret, Origène, Centre Celse IV, 206.17–21Google Scholar; , Koetschau, Origenes Werke II, 233.1–5Google Scholar. The other passages alluded to are Cels. 5.61, 62, 64; 6.11, 24, 26, 28, 29, 30, 34. On Origen's purpose in these and other passages cited in the next two notes see , Pichler, Streit um das Christentum, 206–9Google Scholar.

⁷ Origen Cels. 8.15 ( , Borret, Origène, Comre Celse IV, 206.21–22Google Scholar; , Koetschau, Origenes Werke II, 233.5–6)Google Scholar. Thomas Whittaker (“Celsus and Origen,” in idem, Apollonius ofTyana and Other Essays [London: Swan Sonnenschain, 1906] 112) rightly brands the charge of forgery “absurd,” although he supports it with the equally absurd statement that “there is nothing whatever in the character of Celsus as revealed in the fragments of his work to justify the ascription to him of fraudulence or indifference to truth.” The half-heartedness with which Origen presents the charge is inherent in the way it is parenthetically and suppositionally introduced (εἵ γε); his first charge assumed the opposite. It is equally inherent in the fact that he appends ἤ κολοθησιν προσθηκε to it. As κλουθα αντώ λγειν shows further on (in Origen Cels. 8.16 [, Borret, Origéne, Contre Celse IV, 206.6–8Google Scholar; , Koetschau, Origenes Werke II, 234.1–2]), the words imply that the excerpt is genuine but that Celsus is reading erroneous inferences into itGoogle Scholar.

⁸ The earlier passages are in Origen Cels. 6.27, 35, 38.

⁹ Th charge as leveled at the “Celestial Dialogue” begins at Origen Cels. 8.16 (, Borret, Origène, Contre Clse IV, 208.1–2Google Scholar; , Oelschau, Origenes Werke II, 233.25–26Google Scholar); elsewhere note, e.g., , OrigenCels. 5.54, 64, 65Google Scholar; 6.24, 34-35, 53, 74. , Mosheim (Origenes achr Bücher, 825 n. 4) expresses as much wonderment about this charge as he does about Origen's ignorance (see above n. 6), observing that the second part of the excerpt offers no doctrinal content different from that of the first partGoogle Scholar.

¹⁰ The editions are those of Charles Delarue, most readily available in PG 11.1537; and the revised edition of Delarue by Lommatzsch, Karl Heinrich Eduard (Origenis opera omnia [Berlin: n.p., 1846]) 19. 129Google Scholar. The translations dependent on Delarue or on Lommatzsch are the Latin by Vincent Thuillier in PG ! 1 and those of , Keim, KelsoslCelsus: Wahres Wort, 123Google Scholar; , Rohm, Origenes acht Bücher, 427Google Scholar; Aubé, B., Histoire des persécutions de l'église: La polémique païenne a la fin du II^e siècle (2d ed.; Paris: Didier, 1878) 374Google Scholar; Pélagaud, Élysée, Un conservateur au second siècle: Etude sur Celse et la premiere escarmouche entre la philosophic et le christianisme naissant (Lyons: Georg, 1878) 371Google Scholar; and Crombie, Frederick in ANF 4. 644—45Google Scholar. In his Celse ou le conflit de la civilisation antique et du christianisme primitif (Paris: Siècle, 1925) 420Google Scholar (= idem, Celse: Discours vrai centre les Chretiens [Utrecht: Bosch, 1965] 147–48)Google Scholar, Louis Rougier continues to follow Delarue's text.

¹¹ In his critical apparatus , Koetschau (Origenes Werkes II, 232Google Scholar) gives no grounds for objection to the earlier texts' syntactical format. He implies them, however, with a reference to Franz Overbeck's review of Aubé, whose translation of the excerpt follows Delarue's text. Overbeck himself does not offer any grounds for objection either. He contents himself merely with remarking ( TLZ 3 [1878] 534Google Scholar): “Both [viz. Aubé and Keim] are also surely mistaken when they make the words od και τίς ἄλλος etc. in 8.15 the apodosis of the sentence.”

¹² See, for example, BDF § 442.(8, a special case of 7); and BAG, s.v. καί, I.2.h.

¹³ Harnack, Adolf, Geschichte der altchristlichen Literatur bis Euseus I: Die Überlieferung und der Bestand (2d ed.; Leipzig: Hinrichs, 1958) 201Google Scholar. In his Agrapha: Aussercanonische Schriftfragmente ([2d ed.; Leipzig: Hinrichs, 1906] 286–87Google Scholar) Alfred Resch seems to hold Celsus's citation even less integral than Origen, for he divides it into four separate “Dicta” (apocrypha 83-86). This is even clearer in the first edition ( Resch, Alfred, Agrapha: Aussercanonische Evangelienfragmente [Leipzig: Hinrichs, 1889] 441—45)Google Scholar, where he speaks of four “Citate” and they are not all consecutive (apocrypha 75, 79-81).

¹⁴ Koetschau encloses the clause in parentheses; Borret, whose text I cited, sets the clause off with dashes. The later editors include Glòckner, Otto, Άληθής Λγος (Bonn: Marcus und Weber, 1924) 65.16–17Google Scholar; Schroder, Heinrich Otto, “Der Alethes Logos des Celsus. Untersuchungen zum Werk und seinem Verfasser mil einer Wiederherstellung des griechischen Textes und Kommentar” (Ph.D. diss., Giessen, 1939) 80.11–12Google Scholar; and Bader, Robert, Άληθς Λγοςdes Kelsos (Stuttgart: Kohlhammer, 1940) 197.4–5Google Scholar. All three of these set the clause off with dashes.

¹⁵ Norden, Eduard, Die antike Kunsttprosa (1898Google Scholar; reprinted Stuttgart: Teubner, 1958) 505-10; BDF §§ 489-90.

¹⁶ See, for example, BDF § 372.(Ic) citing Matt 26:33 // Mark 14:29 and 1 Cor 9:11.

¹⁷ Koetschau's, PaulKritische Bemerkungen :u meiner Ausgabe von Origenes (Leipzig: Hinrichs, 1899) gives no hint of his having considered this text, nor does what prompted this work, the review of his edition of Origen byGoogle ScholarWendland, Paul in Göttingische Gelehrie Anzeigen (1899) 276—304Google Scholar. None of Koetschau's successors, cited above in n. 13 and including Borret, seems to have considered this text either. There is nothing among Albert Wifstrand's conjectures in his EIKOTA: Emendationen und Interpretationen zu griechischen Prosaikern de Kaiserzeit IV: Zu christlichen Schriftsîeilern (Lund: Gleerup, 1939)Google Scholar, or his Die wahre Lehre des Kelsos (Lund: Gleerup, 1942)Google Scholar.

¹⁸ Bouhéreau's, Élie work (Traité d'Origène contre Celse [Amsterdam: Desbordes, 1700]Google Scholar) was only accessible to me in the form of notes cited by , Lommatzsch (Origenis opera omnia, 19. 129 n. 8)Google Scholar.

¹⁹ , Mosheim (Origenes acht Backer, 825 n, 4) followed byGoogle Scholar, Keim (Kelsos/Celsus: Wahres Won, 123 n. 1)Google Scholar, and he, in turn, by , Rohm (Origenes acht Backer, 426 n. 4)Google Scholar. Keim is still cited by , Schroder, Alethes Logos des Celsus, 175Google Scholar.

²⁰ It is all the more unprecedented now as we have polemic against orthodox Christianity in Gnostic tracts from Nag Hammadi, and its tone is distinctly different from that evident in our fragment, read this way. See, e.g., Koschorke, Klaus, Die Polemik der Gnostiker gegen das kirchliche Christentum (NHS 12; Leiden: Brill, 1978) on NHC 7, 3 and 9, 3; andGoogle ScholarPainchaud, Louis, “La polémique anti-ecclésiale et lexégèse de la passion dans le Deuxième Traité du Grand Seth,” in Bare, Bernard, ed., Colloque international sur les textes de Nag Hammadi (Quebec: Université Laval and Louvain: Peelers, 1981) 340–51Google Scholar, on NHC 7, 2. The picture is equally unprecedented, if not also improbable, for non-Gnostic early Christian dialogue literature. See Hoffmann, Manfred, Der Dialog ei den christlichen Schriflstellern der ersten vier Jahrhunderte (TU 96; Berlin: Akademie-Verlag, 1966)Google Scholar; and Voss, Bernd Reiner, Der Dialog in derfrühchristlichen Literatur (Studia et Testimonia Antiqua 9; Munich: Fink, 1970)Google Scholar.

²¹ , Mosheim, Origenes Acht Bücher, 825 n. 4Google Scholar.

²² writes, Chadwick (Origen: Contra Celsum, 462 n. 6)Google Scholar, “Is this a Gnostic version of Jesus' agony in Gethsemane?” This approach was already suggested by , Resch, Agrapha (2d ed.) 287 (apocryphon 86)Google Scholar: “reminiscent of Peter's statement: κύριε ίδοὺ μάΧαιραι ὦδε δύο. Luke 22, 38.” Hoffmann, R. Joseph (Celsus On The True Doctrine: A Discourse Against the Christians [New York/Oxford: Oxford University Press, 1987] 143 n. 199) cites Chadwick's view, but he makes Chadwick much more certain than he actually is, and dependence on the Lukan passage has become a certainty. Hoffmann calls the excerpt a “mangled quotation”; by whom (Celsus?), and why he thinks so, he does not sayGoogle Scholar.

²³ See Völker, Walther, Das Bild vom nichtgnostischen Christentum ei Celsus (Halle: Waisenhaus, 1928) esp. 23–27Google Scholar; and, with corrections and refinements of Völker, Ullmann, Wolfgang, “Gnostische und politische Hãresie bei Celsus. Zum Bild der Kirche bei Celsus,” Theologische Versuche 2 (1970) 153–58Google Scholar, esp. 155; Pichler, further, Streit urn das Christentum, 54–59Google Scholar.

²⁴ For example, in Cels. 5.54 (so Origen, rightly), 62 (where he is named explicitly); 6.52, 53 (so again Origen, rightly), 74 (where Origen notes the frequency of Celsus's allusion to Marcionite teaching and reports an expose of its pros and cons, which Celsus seems to have offered at this point); finally 7.18, though perhaps 2.27 (so Origen); 6.29, 73; and 7.25 are also to be included. On the Marcionite origin of these passages, see Harnack, Adolf von, Marcion: Das Evangelium vom fremden Gott (2d ed.; Leipzig: Hinrichs, 1924) 325*-27* andGoogle Scholar, further, 275* on 5.54 (seconded by , Chadwick, Origen: Contra Celsum, 306 n. 3)Google Scholar; 126*, 265*, 295*, 312* on 6.52-53 (so too Chadwick, 368 n. 5, 369 nn. 3-4 and the rightly cautionary n. 5); 284* on 6.74 (Chadwick, 388 n. 1); 173, 254* on 2.27 (Chadwick, 90 n. 2); 281* on 7.25; with Chadwick, 386 n. 7 and 409 n. 7 on 6.73 and 7.18, respectively. Note further , Burke, “Celsus and Late Second-Century Christianity,” 151–52Google Scholar.

²⁵ On the ancient use of the term “Gnostic” as a sectarian self-designation and its partial or total equivalence with the heresiological Ophites, , Chadwick (Origen: Contra Celsum, 306 n. 3) refers toGoogle ScholarCasey, R. P., “The Study of Gnosticism,” JTS 36 (1935) 45–61CrossRefGoogle Scholar. See further Edwards, M. J., “Gnostics and Valentinians in the Church Fathers,” JTS 40 (1989) 26–47CrossRefGoogle Scholar, in response to Smith, Morton, “The History of the Term Gnostikos” in Layton, Bentley, ed., The Rediscovery of Gnosticism, vol. 2: Sethian Gnosticism (Leiden: Brill, 1981) 796–807Google Scholar. On the “Ophite Diagram(s)” and the problems involved, see Welburn, A. J., “Reconstructing the Ophite Diagram,” Nov T 23 (1981) 261–87CrossRefGoogle Scholar.

²⁶ Those who make a Marcionite attribution include Aubé, Histoire des persecutions, 374 n.l, followed by Patrick, John, The Apology of Origen in Reply to Celsus: A Chapter in the History of Apologetics (Edinburgh/London: Blackwood, 1892) 76 n. 2Google Scholar; and more hesitantly, , Burke, “Celsus and Late Second-Century Christianity,” 152Google Scholar. Marcionite affiliation is equally hinted at with allusion to Marcionite references both within and without Contra Celsum by , Bader, Άληθής Λγος des Kelsos, 197 n. 3 referring to Cels. 8.15; and by others, such as Lommatzsch (Origenis opera omnia 19.129 n. 6) andGoogle ScholarKoetschau, Paul (Des Origenes alhr Bücher gegen Celsus [Bibliothek der Kirchenväter 53; 2 vols.; Munich: Kösel & Rustet, 1926] 2. 315 n. 3)Google Scholar, with simple allusion to the Marcionite sections, Cels. 5.54 and 6.53. Ophite affiliation is asserted somewhat waveringly by , Keim, Kelsos/Celsus: Wahres Wort, 123 n. 1Google Scholar, followed as usual by , Rohm, Origenes achi Bücher, 426 n. 4Google Scholar; then , Resch, Agrapha (2d ed.) 286Google Scholar; , Schröder, Alethes Logos des Celsus, 175Google Scholar; and perhaps , Chadwick, Origen: Contra Celsum, 462 n. 6Google Scholar, with his reference to the Ophite section, Cels. 6.27. , Harnack (Geschichte tier altchristlichen Literatur, 201)Google Scholar refuses to decide.

²⁷ , Chadwick, Origen: Contra Celsum, 462Google Scholar; , Borret, Origène, Centre Celse IV, 207Google Scholar.

²⁸ , Burke, “Celsus and Late Second-Century Christianity,” 152 n. 117 (pp. 174–75)Google Scholar.

²⁹ Celsus's rephrasing Τν νί νθρώπον. ὃν ποοαίνουσιν ίσχυρότερον καί κύριον τοὓ κρατούντος θεο[”the Son of man, whom they represent as stronger than and lord of the ruling god”]) is Borret, Origène, Contre Celse IV, 206.12-13 and , Koetschau, Origenes Werke II, 232.20–21Google Scholar; Origen's rephrasing Φαμέν τν υίόν ούκ ίσκυρότερον το πατρς λλ ποδεστερον[”we say that the Son is not stronger than the Father; we say that he is inferior”]) is , Borret, Origène, Contre Celse IV, 206.24–25 andGoogle Scholar, Koetschau, Origenes Werke II, 237.7–8Google Scholar. In both cases Chadwick and Borret naturally take the genitives that follow ίσχυρότερον as comparative. , Hoffmann (Celsus On The True Doctrine, 117)Google Scholar offers “mightier than the lord of the almighty God” (sic) for Celsus's words. Keim (Kelsos Celsus: Wahres Wort, 123) and Aubé (Histoire des persecutions, 374) have a predecessor in Bouhéreau (as cited by Lommatzsch, 129 n. 7), who says of ίσχυρτερος: “Supply to? τούπατρς [”than the Father”] from what follows,” and by “from what follows” Origen's rephrasing is meant. Aubé actually inserts “[than his Father]” into his translation; so too do Pélagaud (Etude sur Celse, 371, without the brackets), and , Rougier (Celse ou ie conflit, 420Google Scholar; idem, Celse Discours vrai, 147). In his Italian translation Aristide Colonna (Contra Celso di Origene [Torino: Unione Tipografico-Editrice Torinese, 1971] 670) similarly construes θεοῡ with νίς. Those who render the clause as I think it must be include , Mosheim (Origenes acht Bücher, 825 n. 4)Google Scholar; Thuillier (PG 11. 1538); , Rohm (Origenes achl Bücher, 427)Google Scholar; Crombie (ANF 4. 644); , Resch (Agrapha [2d ed.] 286 in remarks on apocryphon 83)Google Scholar; , Patrick (Apology of Origen, 76)Google Scholar; Koetschau's updated version of , Rohm (Origenes achr Bücher, 2. 315)Google Scholar; and now the Spanish translation by Bueno, Daniel Ruiz (Origenes contra Celso [Madrid: Biblioteca de Autores Cristianos, 1967] 532)Google Scholar. Schroder's insertion of before υίς (Alethes Logos des Celsus, 80 line 10) seems intended to cement this syntactical interpretation of the clause.

³⁰ See Legasse, S., “L”Homme fort'de Luc xi 21–22Google Scholar, ”NovT 5 (1962) 5–9Google Scholar; Grundmann, Walter, “ίσχύω,” TDNT 3 (1965) 399–401Google Scholar; and the commentaries. On the verses' Marcionite interpretation, see , Harnack, Marcion, 120, 129, 2O8*-9*, 275*, 301*Google Scholar.

³¹ Mum 24:4; 2 Sam 22:31-33, 48; 23:5; Job 22:13; 33:29; 34:31; 36:22, 26; 37:5, 10; Pss 7:11; 4l(42):2; Isa 9:6(5); Neh 1:5; 9:31-32, always rendering, “God/god,” taken, evidently by way of circumlocution, for its homonym , “strength.”

³² See Zahn, Theodor, Geschichte des Neutestamentlichen Kanons, vol. 1: Das Neue Testament vor Origenes (Erlangen: Deichert, 1888) 620—22Google Scholar; , Harnack, Marcion, 40—42Google Scholar; Knox, John, Marcion and the New Testament (Chicago: University of Chicago Press, 1942) 77–113Google Scholar; Hoffmann, R. Joseph, Marcion: On the Restitution of Christianity (American Academy of Religion. Academy Series 46; Chico, CA: Scholars Press, 1984) 113-24, 133-34, 135-41Google Scholar.

³³ Tertullian Marc. 1.6.2; 1.9.9; 2.2.3; 2.16.4-5; 3.15.1; 5.7.9.

³⁴ See, e.g., Schenke, Hans-Martin, Der Gott “Mensch” in der Gnosis (Göttingen: Vandenhoeck Ruprecht, 1962) particularly 34—43 on the Apocryphon of JohnGoogle Scholar.

³⁵ See , Harnack, Marcion, 105—6Google Scholar; , Hoffmann, Marcion, 176Google Scholar; and especially Aland, Barbara, “Marcion. Versuch einer neuen Interpretation,” ZTK 70 (1973) 433-35Google Scholar. Aland argues that although in other respects influenced by Gnosticism, this facet of Marcion's theology is anti-Gnostic in tendency and soteriologically driven. Marcion's disciple Apelles is said to have parted company with his master over this issue, claiming a heavenly origin for human souls; see , Harnack, Marcion, 177, 188-89, with 194-95 in assessmentGoogle Scholar; , Aland (”Marcion,” 442—43) is willing to admit a hint of this idea for Marcion himselfGoogle Scholar. On Marcion's relationship to Gnosticism and the question of whether he is to be classified as a Gnostic, which Harnack generally denied, see Blackman, Edwin Cyril, Marcion and his Influence (London: SPCK, 1948) 82–87Google Scholar; Jonas, Hans, The Gnostic Religion. The Message of the Alien God and the Beginnings of Christianity (2d ed.; Boston: Beacon, 1958) 137-39Google Scholar; Bianchi, Ugo, “Marcion: Theologien biblique ou docteur gnostique?” VC 21 (1967) 141-49Google Scholar; , Aland, “Marcion,” 423, 424, 429-35Google Scholar; and , Hoffmann, Marcion, 155-84Google Scholar.

³⁶ See , Harnack, Marcion, 123-24 with n. 1Google Scholar.

³⁷ Note particularly Tertullian Marc. 4.12.1-2. On the Marcionite use of the pericope see , Harnack, Marcion, 88, 91, 298*Google Scholar.

³⁸ Compare Kũhner, Raphael and Gerth, Bernhard, Ausfiihriiche Grammatik tier griechischen Sprache II.I (1898Google Scholar; reprinted Hannover: Harm, 1976) § 470.3, and Smyth, Herbert Weir, Greek Grammar (rev. Gordon M. Messing; Cambridge, MA: Harvard University Press, 1956) § 1268Google Scholar.

³⁹ See, e.g., Puech, Henri-Charles, “Fragments retrouvés de I'Apocaìypse d'Aìh>i>ène,'i>ène,'>Google Scholar in idem, En tluête de la Gnose, vol. 1: La Gnose el le temps (Paris: Gallimard, 1978) 271-300, esp. 281-84 = Melanges Franz Cumont (2 vols.; Brussels: Institut de Philologie et d'Histoire Orientales et Slaves, 1936) 2. 935-62; idem, “Les nouveaux écrits gnostiques découverts en Haute-Égypte,” in Coptic Studies in honor of Walter Ewig Crum (Boston: Byzantine Institute. 1950) 126-28.

⁴⁰ See , Harnack, Mercian, 119-20, 124, 265*-68*, 274* (citing Justin Martyr, Celsus's contemporary and a personal opponent of Marcion)Google Scholar; Mitchell, C. W., Ephraim's Prose Refuiaiions of Mani, Marcion. and Bardaisan (2 vols.; London/Oxford: Williams and Norgate, 1921) 2. xxv, xxvi, passimGoogle Scholar.

⁴¹ , Mosheim (Origenes acht Bücher, 825 n. 4)Google Scholar and , Colonna (Conlro Celso di Origene, 670 n. 17)Google Scholar deserve credit for recognizing this fact, however dimly or partially. With Mosheim it seems only to have been an intuition, for he thinks the proverbial tradition is Oriental (”couched in the figurative language of the Orientals, or in one of the proverbs common among the Orientals”). Colonna's remarks (”A proverbial expression to indicate someone who wishes to stay far away from actual dangers, or not to risk much in them”) may reflect some knowledge of the actual tradition, but, as we shall see, his appraisal of the point here is far from accurate.

⁴² See Jardé, A., “puteus,” in , Daremberg-Saglio, Dictionnaire des antiatés grecques ei romaines 4.i (n.d.) 779-81Google Scholar.

⁴³ For Thales see Diels, Herman, Die Fragmente der Vorsokratiker (ed. Kranz, Walther; 6th ed.; 3 vols.; Dublin/Zurich: Weidmann, 1969-1973) 1. 75 (A9)Google Scholar, citing , PlatoTheael. 174A. In Al (p. 70 line 31)Google Scholar, from Diogenes Laertius, it is a βθρος (“pit”) into which he falls. Other cases of accidental falls into a well include Plato Gorg. 471C; Luke 14:5; and the little boys in the sepulchral epigrams Anth. Pal. 7.170 (ascribed to Poseidippos or Callimachus) and SEG 4.573 (from Notium, second century BCE). In the epigram from Notium an uncle (?) jumps into a well in a vain attempt to rescue the child. In his comments on this text, Robert, Louis (”Etudes dépigraphie grecque,” Revue de Philologie, de Littérature eld'Histoire Anciennes 60 [1934] 49—52)Google Scholar cites another example from Mark the , Deacon'sLife of PorphyryGoogle Scholar. Yet another example is Kaibel, Georg, Epigrammata graeca ex lapidibus conlecta (1878Google Scholar; reprinted Hildesheim: Olms, 1965) no. 571 (from Rome, first or second century CE), where the fall is into a spring. In Diodorus Siculus 5.50.5; 12.58.5; and , LucianDial. Meretr. 12.312, the fall is intentional, that is, a suicideGoogle Scholar.

⁴⁴ Thus in the generalization drawn for philosophers from Thales' experience a little later in Theaet. 174C, είς Φρατα becomes είς πᾱσαν πυραν, “into dire straits.”

⁴⁵ , PlutarchMoralia 68B. Diogenes Laertius 6.52 is a probable allusionGoogle Scholar.

⁴⁶ See Mylonas, George E., Eleusis and the Eleusinian Mysteries (Princeton: Princeton University Press, 1961) 44–47 and plate 33Google Scholar; Richardson, N. J., The Homeric Hymn to Demeter (Oxford: Clarendon, 1974) 181, 250, 326-28 (on lines 99 and 272 of the hymn)Google Scholar. The conjunction of maidens dancing around a well with falling into it is exemplified, for example, in a myth preserved in the Geoponica ( 11.4.2), where the maidens are the Charites. For Persephone, see , Richardson on line 42 of The Homeric Hymn to Demeter (p. 163)Google Scholar; the cult involved was centered on a Syracusan spring.

⁴⁷ Zenobius 4.100 and Diogenianus 6.21 (Leutsch, E. L. and Schneidewin, F. G., eds., Corpus paroemiographicorum graecorum [ 2 vols.; Göttingen: Vandenhoeck & Ruprecht, 1839-1851] 1. 114 and 273, respectively)Google Scholar, whence it is taken up, with the gloss garbled by curtailment, into the Lexica of Hesychius, Photius, and the Suda. The translation of the gloss is from that given by Zenobius. The text itself runs: Λκος περί Φραρ χορευει παροιμία έπΦραρ μηδν νων. Leutsch and Schneidewin (p. 114, in the comments on Zenobius 4.100) think that Plutarch's ἂρχησις περί τ Φρέαρ (Moralia 68B) is an adaptation of this same proverb. This may perhaps be so, but its point is nevertheless quite different.

⁴⁸ Hesychius s.v. κολυμβηις, “divers.” The fact that Hesychius gives the word an entry shows that this is a peculiar or technical usage and does not necessarily mean that such men literally “dove” into wells. The usage probably derives from the passages in Plato cited in what follows, where it is evidently a colloquialism. Hesychius's gloss reads: τοὺς κ τών Φρετων ναπμποντας τοὺς κδους(“men who fetch vessels out of wells”).

⁴⁹ Note further the parallel between Plato's είς τ Φρατα κολυμβάν (Prot. 349E-35OA) and κολυμβάν είς τν Τρταρον, “diving into Tartarus,” in a fragment of the fifth-century comic playwright Pherekrates preserved by Athenaeus (6.268E; John Maxwell Edmonds, The Fragments of Attic Comedy [3 vols.; Leiden: Brill, 1957-61] 1. 246 [fragment 108, line 21] with Edmonds's note on p. 247).

⁵⁰ See Heintz, Jean-Georges, “bar,” TOOT 1 (1974) 463–66Google Scholar; Tomp, Nicholas J., Primitive Conceptions of Death and the Nether World in the Old Testament (BibOr 21; Rome: Pontifical Biblical Institute, 1969) 66–69, esp. 69Google Scholar. The fact that the Israelites shared the conception of the underworld as a subterranean body of water with other ancient Near Eastern peoples helped the Israelite tradition to assimilate a well to Sheol. On this see , Tromp, Primitive Conceptions, 59–66Google Scholar.

⁵¹ Konstantin von Tischendorf, Apocalypses apocryphae (1866; reprinted Hildesheim: Olms, 1966) 62 (Φρέαρ [”a well… full of darkness and gloom”]) and 90 (σκτους καί ζότους πεπληρωμνον [”a seven-mouthed well of chastisement”]), respectively.

⁵² It may not be an accident that not only the Greek verb employed by the “Celestial Dialogue” here (νειν) but its particular form as well (νων) are precisely those used by Zenobius in his gloss on the proverb about the wolf dancing around a well (see above n. 46). It may rather indicate direct dependence of the former upon the latter, as well as additional support for allusion to the proverb with περί τ Φραρ in the “Celestial Dialogue.”

⁵³ , Resch (Agrapha [2d ed.] 286)Google Scholar hints at this with “no one is willing to go down to rescue the poor unfortunate” as a comment on οὐδείς είς τ Φρέαρ, but he does not pursue the matter any further.

⁵⁴ The most famous tradent of this motif is the “Hymn of the Pearl” in chaps. 108—13 of the Acts of Thomas; see Poirier, Paul-Hubert, L'Hymne de la Perle des Acies de Thomas (Louvain-la-Neuve: Institut orientaliste, 1981) 318–20Google Scholar. Note, further, Jacques-É, Ménard, “Le ‘Descensus ad Inferos,’” in Ex Orbe Religionum: Studio Geo Widengren Oata (Studies in the History of Religion 21-22; ed. Bleeker, C. J., et al.; 2 vols.; Leiden: Brill, 1972) 2. 304–6Google Scholar; Peel, Malcolm L., “The 'Decensus [sic] ad Inferos' in ‘The Teachings of Silvanus’ (CG VII, 4)” Numen 26 (1979) 27–36CrossRefGoogle Scholar, with other texts, including other new ones from Nag Hammadi (add 7, 1, Paraphrase of Shem), cited p. 33 n. 24. Attridge, Harold W. (“Liberating Death's Captives. Reconsideration of an Early Christian Myth,” in Goehring, James E., et al., eds., Gnosticism and the Early Christian World [Sonoma, CA: Polebridge, 1990] 105, 106Google Scholar) rightly argues that the identification of Hades with the world is not confined to Gnostic texts and that in the Epistle to the Hebrews the motif emerges earlier than is usually accepted (see pp. 108-9). Another early context in which the motif emerges is baptismal, in particular in the baptism of Christ, where the descent of the Christ-dove from heaven occurs in conjunction with Jesus' descent into the waters of death. This form of the motif is all the more interesting for the “Celestial Dialogue” because in it too the descent is into the waters of a well (or cistern) and there is common assimilation of reference to Luke 11:21—22, with the “strong man” being the devil and his demons, who are defeated in the ritual of baptism. See Rousseau, Dom Olivier, “La descente aux enfers, fondement sotériologique du baptême chrétien,” Rech SR 40 (1951-1952) 274–97, esp. 280-87Google Scholar; Daniélou, Jean, A History of Early Christian Doctrine before the Council of Nicea, vol. 1: The Theology of Jewish Christianity (trans. and ed. Baker, John A.; London: Darton, Longman & Todd and Philadelphia: Westminster, 1964) 224-31, 233–48Google Scholar; , Ménard, “Le ‘Descensus ad Inferos’” 303—4Google Scholar; Charlesworth, James H., ed. and trans., The Odes of Solomon (SBLTT 13; Pseudepigrapha Series 7; Chico, CA: Scholars Press, 1977) 99–100Google Scholar in his notes on Ode 24. The application of the tradition of descensus ad inferos to Christ's descent into appearance on earth doubtless owes something, in Gnostic texts at least, to the widespread equation of Hades with earth in contemporary theories of the descent of souls from their sidereal homeland into terrestrial embodiment; see, e.g., Cumont, Franz, Luxperpetua (Paris: Geuthner, 1949) 189–96Google Scholar.

⁵⁵ As Marcion does, for example. See , Harnack, Marcion, 130—31, 294*-95*, and furtherGoogle Scholar, Hoffmann, Marcion, 195 n. 44Google Scholar, noting that the descensus ad inferos, whether from earth to underworld or heaven to earth, is the only mythologem known to have been expounded by Marcion; see also pp. 197-98. Marcion accepted a regular descensus ad inferos more easily since, however docetic his christology was otherwise, his soteriology was firmly based on Paul and entailed a central focus on the reality of the Passion.

⁵⁶ See, e.g., Quinn, Esther Casie, The Quest of Seth for the Oil of Life (Chicago/London: University of Chicago Press, 1962)Google Scholar; Klijn, Albertus Frederik Johannes, Seth in Jewish, Christian and Gnostic Literature (Nov T Sup 46; Leiden: Brill, 1977)Google Scholar; and, specifically on the passage from Origen, , Burke, “Celsus and Late Second-Century Christianity,” 159Google Scholar.

⁵⁷ On this motif see, e.g., Beyschlag, Karlmann, Simon Magus und die christliche Gnosis (WUNT 16; Tübingen: Mohr/Siebeck, 1974) 171–78Google Scholar; Orbe, Antonio, Crisloìogia Gnóstica: Introducción a la soteriología de los siglos II y III (2 vols.; Madrid: Biblioteca de Autores Cristianos, 1976) 1. 230–38Google Scholar; , Peel, “The ‘Decensus ad Inferos,’” 36–39Google Scholar.

⁵⁸ , Harack, Marcion, 219*-20*, 260*, 302*-3*Google Scholar.

⁵⁹ I cite the English translation by Williams, C. S. C. following his edition of the text in “Eznik's Résumé of Marcionite Doctrine,” JTS 45 (1944) 70–73CrossRefGoogle Scholar. Another rendition is offered by Norman McLean at the end of his entry “Marcionism” in ERE 8 (1980) 409Google Scholar. German translations are those by Schmid, Joh. Michael, Des Wardapet Eznik von Kolh wider die Sekten (Vienna: Mechitharisten-Congregation, 1900) 172–78Google Scholar; and Weber, Simon, AusgewählteSchriften der armenischen Kirchenväter (Bibliothek der Kirchenväter 57; 2d ed; 2 vols.; Munich: Kösel & Pustet, 1927) 1. 152–55Google Scholar. French translations are thoseby Rivière, Jean in “Un expose marcionite de la redemption I. Texte et date du document,” RevScRel 1 (1921) 189–92Google Scholar; and Lariès, Louis and Ch. , Mercier, Eznik de Kolb De Deo (PO 28.3-4; Paris: Firmin-Didot, 1959) 662–65 (§358; text pp. 514—17 with notes on pp. 760-61)Google Scholar.

⁶⁰ For these features of Marcion's christology and eschatology see , Harnack, Marcion, 119–20Google Scholar; , Hoffmann, Marcion, 194—96, 202–4Google Scholar. For military imagery in the descensus ad inferos, note its presence in stanza 8 of the “Hymn of the Pearl,” where the descent is, however metaphorically, from heaven to earth, as in the “Celestial Dialogue.” See further Gschwind, Karl, Die Niederfahrt Christi in die Unterwelt (Münster: Aschendorff, 1911) 234–42Google Scholar; , Peel, “The ‘Decensus ad Inferos,’” 39–47Google Scholar; , Attridge, “Liberating Death's Captives,” 104, 105—6; and generallyGoogle Scholar, Orbe, Cristología Gnóstica, 1. 134–53Google Scholar. For the ancient roots of this tradition see Kroll, Josef, Gott und Hölle (Studien der Bibliothek Warburg 20; Leipzig/Berlin: Teubner, 1932)Google Scholar; and especially Widengren, Geo, Mesopotamian Elements in Manichaeism (Uppsala: Lundequist and Leipzig: Harrassowitz, 1946) 41–73Google Scholar.

⁶¹ On the issues of Eznik's sources and their date see Harnack, Adolf, “Beiträge zur Geschichte der marcionitischen Kirchen,” ZWT 19 (1876) 84–102, esp. 85, 86, 100Google Scholar; idem, Marcion. 372*-73*, 380*; Mariès, Louis, “Le De Deo d'Eznik de Kolth connu sous le nom de ‘Centre les sectes.’ Études de critique littéraire et textuelle,” Revue des Études Arméniennes 4 (1924) 171–97Google Scholar; and , Riviere, “Un expose marcionite I,” 186, 192–207Google Scholar. In the second part of his article ( “Un expose marcionite de la redemption II. Utilisation du document,” Rev Sc Rel 1 [1921] 297–323Google Scholar, esp. 312-18) Riviere maintains that the second descent of which our section speaks is a second descent into Hades, although he is fully aware of the inconsistencies of such a view. Harnack and others, whom Riviere takes to task, are correct: what occurs is a second descent to the Creator's heaven and then a further descent as far as earth for a visionary appearance in the conversion of Paul.

⁶² Henri-Charles Puech, “Appndice” to “Une collection de paroles de Jesus récemment retrouvé: LÉvangile selon Thomas,” Comptes rendus des seances de I'Académie des Inscriptions el Belles-lettres (1957) 166Google Scholar= idem, En quête de la Gnose, vol. 2: Sur l'Évangile selon Thomas (Paris: Gallimard, 1978) 56, and then in NTA poc 1 (1963) 302–3Google Scholar. Puech is followed by Doresse, Jean, Les livres secrets des Gnostiques d'Égypte, vol. 2: L'Évangile selon Thomas ou Les paroles secrètes de Jésus (Paris: Plon, 1959) 61, 187Google Scholar; Quecke, Hans, “Das Thomas-Evangelium,” in Unnik, Willem Cornells van, Evangelien aus dem Nilsand (Frankfurt a. M.: Scheffler, 1960) 173 n. 13Google Scholar; Bauer, Johannes B., “Echte Jesusworte?” in Unnik, Evangelien, 112Google Scholar; Grant, Robert M. and Freedman, David Noel, The Secret Sayings of Jesus (Garden City, NY: Doueday, 1960) 176Google Scholar; Wilson, R. McL., Studies in the Gospel of Thomas (London: Mowbray, 1960) 33Google Scholar; Kasser, Rodolphe, LÉvangile selon Thomas (Neuchatel: Delachaux & Niestlé, 1961) 97 and n. 3Google Scholar; Turner, H. E. W. and Montefiore, Hugh, Thomas and the Evangelists (Naperville, IL: Allenson, 1962) 19, 81Google Scholar; Leipoldt, Johannes, Das Evangelium nach Thomas (TU 101; Berlin: Akademie-Verlag, 1967) 70Google Scholar; Ménard, Jacques-É, L'Évangile selon Thomas (NHS 5; Leiden: Brill, 1975) 175Google Scholar; and Fieger, lately Michael, Das Thomasevangelium (NTAbh 22; Münster: Aschendorff, 1991) 207Google Scholar. Puech adds “(ophique?)” to his reference to the “Celestial Dialogue,” and among other commentators all but Quecke, Leipoldt, and Fieger, who express no view, follow him in this sectarian assessment. With the exception of Ménard (“perhaps”) they dispense with Puech's cautious question mark. As Bauer explicitly shows (“Echte Jesusworte?” 146 n. 19), this assessment of the sectarian affiliation of the “Celestial Dialogue” may simply be due to Resch, whose view this was (see above n. 25). In both cases where the Greek text has οραρ the Coptic text requires emendation. In the first instance it has Φραρ, “penetration,” and in the second , “sickness,” both of which are not only nonsensical in the context but, equally impossibly, are nouns of masculine gender accompanied by the feminine article. The correct reading in both cases is , “well, cistern, pit,” which is feminine and regularly used to render Φραρ. , Leipoldt (Evangelium nach Thomas, 70) remarks that “the Coptic scribe has miswritten the word for ‘well’ twice right after each other, and, what is more, each time in a completely different way,” adding, “That is unlikely.” But, as Kasser indicates with one example (L'Évangile selon Thomas, 97 n. 2) andGoogle ScholarCrum, (A Coptic Dictionary [Oxford: Clarendon, 1939] s.v. and , 54la and 745a) with many othersGoogle Scholar, and regularly interchangeable, so that may be a mere variant of . It is possible, consequently, that only is a case of scribal error. Leipoldt, in an earlier work ( “Ein neues Evanelium?” ThLZ 83 [1958] 490Google Scholar), and , Kasser (L'Évangile selon Thomas, 96, 97 n. 3)Google Scholar attempt a translation of the text as it stands, though Kasser immediately rejects an explanation of its meaning as improbable.

⁶³ Many translators adhere closely to this literal rendering. The translation in the editio princeps ( Guillaumont, A., et al,, eds., The Gospel according to Thomas [Leiden: Brill and New York/Evanston: Harper & Row, 1959] 41Google Scholar), followed by , Turner and , Montefiore (Thomas and the Evangelists, 81Google Scholar) and now by , Fieger (Thomasevangelium, 207Google Scholar), opts for “cistern” for .

⁶⁴ See BDF §§ 205-6. The reverse interchange is also attested, although much more rarely (BDF § 218).

⁶⁵ , Crum, A Coptic Dictionary, s.v. , 683aGoogle Scholar.

⁶⁶ , Leipoldt, Evangelium nach Thomas, 45 (“am †Brunnen†”)Google Scholar; Doresse, Jean, The Secret Books of the Egyptian Gnostics: An Introduction to the Gnostic Coptic Manuscripts Discovered at Chenoboskion (trans. Mairet, Philip; 1958Google Scholar; reprinted London: Hollis & Carter, 1960) 366, where the translation (Doresse's apocryphon 78) is “in the well,” and then p. 376, where the explanation refers to drawing water; see also , Grant and , Freedman, Secret Sayings of Jesus, 176Google Scholar. After noting Resch's view (Agrapha [2d ed.] 286), correct for the “Celestial Dialogue,” that a descent into the well to rescue a victim is behind the saying, and after a nod at Luke 14:5, Bauer (“Echte Jesusworte?” 113—15) embarks on an elaborate explanation based on the Valentinian Herakleon's interpretation of Jesus' meeting with the Samaritan woman at Jacob's well in John 4. Despite its ingenuity and attractiveness it still runs afoul of because it presumes drawing water from the well. The same water-drawing tack is taken up by , Hoffmann (Celsus On The True Doctrine, 116)Google Scholar for the “Celestial Dialogue,” where it is equally misguided.

⁶⁷ , Doresse, L'Évangile selon Thomas, 187Google Scholar; , Kasser, L'Évangile selon Thomas, 97Google Scholar; , Turner and , Montefiore, Thomas and the Evangelists, 80–81Google Scholar; , Ménard, L'Évangile selon Thomas, 175Google Scholar. Ménards reference is actually in the context of reference to the “Celestial Dialogue” as cited by Resch, although Resch does not himself hold to this interpretation of the descent into the well.

⁶⁸ , Lambdin's translation appears first in NHLE (1977) 126Google Scholar. It is then carried on into the revised edition of NHLE (1988) 134Google Scholar, and into Nag Hammadi Codex 11, 2-7 together with Kill, 2*. Brit. Lib. Or. 4926(1) and P. Oxy. 1, 654, 655 (ed. Layton, Bentley; NHS 20; Leiden: Brill, 1989) 1. 81Google Scholar.

⁶⁹ Only Doresse (L'Évangile selon Thomas, 61), , Kasser (L'Évangile selon Thomas, 97)Google Scholar, and , Fieger (Thomasevangelium, 207)Google Scholar voice an opinion, and they all opt for Jesus. Kasser would emend the text so as to have it read: “The Lord said, ‘There are many….’” That is not only unlikely, as Bentley Layton avers in his critical apparatus on this line in Nag Hammadi Codex II, 2-7 1. 80, but for the reason just adduced it is also unnecessary.

⁷⁰ This was observed first by , Doresse, L'Évangile selon Thomas, 187Google Scholar; followed by , Grant and , Freedman, Secret Sayings of Jesus, 176 n. 11Google Scholar; then in fuller form and detail by Bauer. “Echte Jesusworte?” 111–13Google Scholar; , Kasser, L'Évangile selon Thomas, 97Google Scholar; , Turner and , Montefiore, Thomas and the Evangelists, 23-24, 80–81Google Scholar; , Leipoldt, Evangelium nach Thomas, 70Google Scholar; , Ménard. L'Évangile selon Thomas, 175Google Scholar; and Suarez, Philippe de, L'Évangile selon Thomas (2d ed.; Marsanne: Métanola, 1975) 296Google Scholar.

^7l For the date of the Gospel of Thomas and its earliest attestations see the summary provided by Koester, Helmut in , Layton, Nag Hammadi Codex II, 2-7 1. 38–39Google Scholar.