Jesus said: “Behold, the sower went out, he filled his hand, he cast. And some fell on the path; the birds came, they gathered them. Others fell on the rock, and did not take root down into the earth and produced no ear up to heaven. And others fell on the thorns; [ ] they choked the seed and the worm ate them. And others fell upon the good soil and gave good fruit up to heaven; it gave sixty per measure and a hundred and twenty per measure.
As in the synoptic versions, the parable consists in a two-part introductory statement about the sower, the second part of which leads to a sequence of four further two-part statements which describe (i) where the seed fell, and (ii) what happened in consequence. It is striking that the divergences between GTh and the synoptic versions are concentrated in the second member of each of the five main statements, whereas the first member is in each case very similar.
Thomas provides no equivalent to the detailed, point-by-point interpretation that follows in all three synoptists, which requires not only a general distinction between the fruitful and unfruitful seeds but also clear differentiations among the unfruitful ones. In combination with the various types of soil, the birds, the sun, and the thorns all represent unfruitfulness, but, for the synoptic interpretation, they stand for different threats operating at different times and in different ways. This sober analysis of the fate of the word in a threatening environment has helped to shape the framing of the parable itself. On the other hand, the Thomas version seems to intend nothing more than the general contrast between fruitfulness and unfruitfulness. In each of the first three instances, the failure of the seed is total; there is no initial moment of fruitfulness that only later comes to nothing. In GTh, the parable not only lacks the synoptic interpretation, it also excludes it.178
We may suppose, then, that some precursor of the Markan evangelist had at his disposal an SC that included the Parable of the Sower without an accompanying interpretation. While it is not clear how far the Thomas version preserves actual pre-Markan wording, GTh does suggest the parable’s original independence.179 Yet the Markan provision of an interpretation is in keeping with the SC’s basic intention, which is precisely that Jesus’ sayings be interpreted. Thomas promises that “whoever finds the hermeneia of these sayings shall not taste death.”180 Similarly, the author of 2 Clement cites sayings in order to interpret them. Since that text is actually a homily and not a letter, the primary context of sayings interpretation would seem to be preaching. If so, then the interpretation of the Parable of the Sower has a homiletic origin. It is “authentic” not because it represents ipsissima verba of Jesus but because it realizes the intention of the parable in its written form, which is to generate the supplementary discourse of interpretation.181
In Mark 4, the originally contextless parable is provided with a setting in Jesus’ ministry (4.1-2), a conclusion (4.9), and an interpretation (4.10*, 13-20); and it is later supplemented again with a passage about the rationale for parables in general (4.10-12). In addition, two composite sayings are inserted between the interpretation of the Sower and the Parable of the Seed growing secretly, marked off from each other and the surrounding context by the use of introductory formulae. These sayings serve to outline a parable theory differing from that of 4.10-12. The first of these sayings comprises three main elements, a question, a supporting statement, and an appeal:
And he said to them: “Does a light come so as to be put under the measure or under the bed, rather than on the lampstand? For there is nothing secret except in order to be revealed, nor is anything hidden except to come to light. Whoever has ears to hear, let him hear!” (Mk. 4.21-23)
The lampstand saying was transferred by Matthew to another context (Mt. 5.15). Luke, encountering the saying in both Mark and Matthew, retains it in the Markan context but relocates the Matthean version (Lk. 8.16; 11.33). With the exception of a minor transposition, the Thomas version is especially close to the second Lukan one (as underlining indicates):
Jesus said: “[What you [s.] hear in your ears, proclaim from your rooftops.] For no one lights a lamp and puts it <under a measure>, nor does he put it <in a secret place,> but he puts it on the lampstand so that everyone who enters and leaves will see its light.” (GTh 33.1-3)182
It is likely that the lampstand saying developed in a linear manner, from Mark to Matthew to Luke to Thomas. It is variously linked with the “nothing secret” saying (Mk. 4.22; Lk. 8.17), the “city set on a hill” saying and the application to good works (Mt. 5.14-16), the saying about the eye as “lamp of the body” (Lk. 11.34-36), and the “rooftops” saying (GTh 33.1). Thus the linear development of the lamp saying is accompanied by a set of nonlinear juxtapositions, and it is this nonlinear development that is of the greater interpretative significance. Mark and Thomas confirm again that juxtaposition is a fundamental characteristic of the sayings tradition, both in SCs and in narrative gospels. It is a primarily literary phenomenon, in which free combinations and recombinations occur between relatively stable literary units.
The “nothing secret” saying recurs in Thomas in wording indistinguishable from Matthew 10.26, but, once again, in a different context. In Mark, the saying is a promise that the inner meaning of the parables will come to light; in Matthew, it is incorporated into a passage on fearless proclamation. For Thomas, its theme is truthful speech:
. . . Jesus said: “Do not tell a lie, and do not do what you hate, for everything is disclosed in the sight of Heaven. For there is nothing hidden that will not become manifest, and there is nothing covered that will remain undisclosed.” (GTh 6.2-6)
In Thomas, disclosure is the fate of hiddenness: thus, no secret falsehood will remain unexposed for ever. In Mark, disclosure is the purpose of hiddenness — an adaptation that accommodates the saying to the evangelist’s parable theory. The wording preserved in Thomas could represent the original written form of this saying, on which other versions are dependent. The more important point is that both sayings incorporated in Mark 4.21-22 may have been made available to the evangelist by a Thomas-like SC. Even if the wording of a saying in Thomas shows synoptic influence, its SC context is independent of narrative gospels.
Mark’s second composite saying comprises (i) an exhortation to hear, a mirror image of the exhortation that concludes the previous saying (4.23, 24a); (ii) a saying on reciprocity (4.24b); and (iii) a saying contrasting haves with have-nots (4.25). As we have seen, the second of these recurs in 1 Clement in a form suggesting a nonsynoptic SC origin; the third has a parallel in GTh.
And he said to them: “Take heed what you hear! In what measure you measure it shall be measured to you, and it shall be added to you. For whoever has, it will be given to him, and whoever does not have, what he has will be taken from him.” (Mk. 4.24-25)
“[. . . As you give, so it shall be given to you. As you judge, so shall you be judged. As you are kind, so shall you receive kindness.] In what measure you measure, with it shall it be measured to you.” (1 Clem. 13.2)
Jesus said: “Whoever has in his hand, it will be given to him, and whoever does not have, the little that he has will be taken from him.” (GTh 41.1-2)
Apart from minor expansions (italicized), the wording of these sayings is relatively stable. Clement and Thomas suggest that, even prior to Mark, the sayings may have existed separately in a literary form distinct from the canonical gospels.
The connection between the first and the second of Mark’s parables of growth is interrupted by the interpretation of the Parable of the Sower, and by the material comprising Mark’s parable theory (4.10-12, 21-25). In the original SC, it is possible that the Sower was followed immediately by the Parable of the Seed sown secretly, just as the latter is followed immediately by the Parable of the Mustard Seed.183 GTh 63-65 forms a collection of consecutive parables (the Rich Fool, the Great Feast, and the Wicked Tenants). Vestiges of a connection between the second and third parables may still be visible in GTh 20-21, though the order is reversed and the second parable survives only in fragmentary form in a composite context.
And he said: “The kingdom of God is as if a man should cast seed upon the ground. . . . And when the fruit ripened, immediately he sends the sickle, because the harvest has come.” (Mk. 4.26-29)
[Mariam said to Jesus: “Whom are your disciples like?” He said: “They are like children. . . .] Let there be among you a man of understanding. When the fruit ripened, he came immediately with his sickle in his hand, he harvested it. Whoever has ears to hear, let him hear!” (GTh 21.1-2, 8-10)
The conclusion of the Markan parable alludes to Joel 4.13 LXX;184 the allusion is absent in Thomas. Thomas’s concluding exhortation recurs in five other parables: the Fisherman (GTh 8), the Man of Light (24), the Rich Fool (63), the Wicked Tenants (65), and the Leaven (96). The original function of this exhortation was evidently to mark the conclusion of a parable. Mark uses it to conclude the Parable of the Sower, but extends it to the first of the composite sayings, and perhaps also to the later saying about defilement, where the text is uncertain (Mk. 4.9, 23; 7.16?). Matthew too can transfer the exhortation to hear out of a parabolic context (Mt. 11.15: the Baptist as Elijah), and also uses it to conclude his interpretation of the Wheat and the Tares (13.43). In Luke, it is attached to a short parable about salt (Lk. 14.35).
In the Parable of the Mustard Seed, Mark’s version may represent an expansion of a form resembling Thomas’s version. Here, if anywhere, Thomas has a strong claim to represent the oldest accessible form of a saying or parable of Jesus (see Figure 5.1).185
Luke places this parable in a non-Markan context, pairing it with the Parable of the Leaven. He is primarily dependent here on Matthew, who retains the Markan context but inserts the Parable of the Leaven (thus Mt. 13.31-33 = Lk. 13.18-21). In itself, the Matthean passage shows no knowledge of a second, Q version. At 1 Luke has material in common with Mark; in 2-6, with Matthew; Luke’s dependence on both Mark and Matthew is the most economical explanation. It is the Q hypothesis that generates a Q version of this parable, and not the texts as they stand.
Matthew and Luke present versions of the parable that are briefer than Mark’s, but they retain his hyperbolic insistence that the smallest of all seeds produces the greatest of all plants — a great tree, indeed, in whose branches the birds of the sky can make their nests (Mt. 13.31-32 = Lk. 13.18-19). The conclusion of the synoptic parable again alludes to scripture, in this case to Daniel 4.12.186 Both the hyperbole and the allusion are absent in Thomas, where the image is of individual birds finding shade under a plant rather than flocks of them nesting in a great tree.187 This conclusion results in a more coherent parable than the synoptic versions, and it may approximate more closely than they do to its earliest written form.188
| Lk. 13.18-19 | Mt. 13.31-32 | Mk. 4.30-32 | GTh 20 |
| 1a So he said: | 1 Another parable he set before them, saying: | 1a And he said: | 1 The disciples said to Jesus: “Tell us what the kingdom of heaven is like.” He said to them: |
| 1b “What is the kingdom of God like, | 1b “With what can we compare the kingdom of God, | ||
| 1c and to what shall I compare it? | 1c or what parable shall we use for it? | ||
| 2 “It is like a grain of mustard, which a man took and cast into his garden, | 2“The kingdom of heaven is like a grain of mustard, which a man took and sowed in his field. | 2 It is like a grain of mustard, which, when sown upon the ground, | 2 “It is like a grain of mustard, which |
| [3] | 3 It is the smallest of all the seeds, | 3 is the smallest of all the seeds on the earth; | 3 is the smallest of all seeds, |
| 4 and it grew | 4 but when it is grown becomes greater than all plants | 4 but when it is sown it springs up and becomes greater than all plants, | 4 and when it falls upon tilled soil it grows great |
| 5 and it became a tree | 5 and becomes a tree | 5 and produces large branches | [5] |
| 6 and the birds of the sky dwelt in its branches.” | 6 so that the birds of the sky come and dwell in its branches. | 6 so that the birds of the sky can dwell under its shadow.” | 6 and gives shade to the birds of the sky.” |
Bold: common enumeration
Italics: Lk. = Mt.
Underlining: Markan elements absent from GTh.
Mark’s compositional process in his parable chapter makes excellent sense if he has at his disposal an SC containing eight distinct items. This SC would overlap with GTh at all but one of these points. In Markan sequence and structure, these items are as follows:
1. The Parable of the Sower (= GTh 9; Mk. 4.3-8)
2. The Exhortation to hear (= GTh 21.10; Mk. 4.9, cf. v. 23)
3. The Lamp saying (= GTh 33.2-3; Mk. 4.21)
4. The Disclosure saying (= GTh 6.4-6; Mk. 4.22)
5. The Measure saying (= 1 Clem. 13.2; Mk. 4.24)
6. Haves and have-nots (= GTh 41.1-2; Mk. 4.25)
7. The Parable of the Seed sown secretly (= GTh 21.9; Mk. 4.26-29)
8. The Parable of the Mustard Seed (= GTh 20.1-4; Mk. 4.30-32)
Items 1, 7, and 8 may already have been grouped together in the evangelist’s SC. Item 1 is provided with a narrative introduction (Mk. 4.1-2), and item 8 is followed by a narrative conclusion (4.33-34). Item 2 has been transferred from the conclusion of another parable, perhaps the seed grown secretly, as the separate introductory formula indicates. Items 3-6 have been rearranged into two composite sayings, which the evangelist inserts into the parable collection in order to encourage receptive hearing. They complement the interpretation of the Parable of the Sower (4.13-20), which may be understood as a primitive Christian sermon outline. The interpretation was not present in the SC, and neither was the saying on the mystery of the kingdom, which disturbs the earlier link between the parable and its interpretation (4.11-12). The presentation of Jesus’ parables and sayings is an ongoing interpretative process.
Interpretation proceeds, first, by locating the parabolic teaching in a credible and vivid situation within the historical ministry of Jesus. The effect is not to “historicize” Jesus’ teaching, relegating it to the past, but rather to emphasize that the sayings are inseparable from the one who speaks in them. Second, the Parable of the Sower is marked out for special attention. The interpretative decision is made to regard this as a parable in four parts rather than two, and a key is provided that will also unlock the meaning of the other parables. The parable and its interpretation are carefully coordinated with one another, and the interpretation makes it impossible for hearers or readers to evade the existential question implicit within the parable. Third, originally unconnected sayings are juxtaposed with one another in order to clarify further the situation of parabolic address. In them, Jesus teaches that the aim of his parables is not to obscure the truth but to bring it to light; the basis for their reception lies in the hearer. Fourth, a passage is subsequently added which reflects on the distinction between the crowds, who hear the parables, and the disciples, who are taught what they mean (4.11-12; cf. vv. 33-34). This complex reception process takes its starting point from the initial inscription of discrete sayings and parables of Jesus in SC format.
This attempt to install the Gospel of Thomas at the heart of the so-called “synoptic problem” might be extended to the supposedly “Q” material of Matthew and Luke. In particular, the SC hypothesis would help to account for Matthew’s expansion of Mark. Matthew’s gospel would draw not only on Mark but also on one or more SC. Luke’s use of Matthew as well as Mark would make this evangelist less dependent on SCs than his predecessors — although, as we have seen, he seems to have compiled his own SC in the process of editing his Matthean source material.
Thomas, then, is the point of departure for an “SC hypothesis” that should replace the implausible Q hypothesis. GTh itself is best understood as a descendant of the early SCs employed by Mark and Matthew, and as a relation of SCs known from second-century sources. The early SCs cannot be reconstructed as a whole; unlike Q, there can be no “critical edition.” Yet items in GTh with synoptic parallels can plausibly be traced back to presynoptic collections. Even where a saying in its present form betrays the influence of synoptic redaction, that may be the result of secondary assimilation. In other cases, GTh may preserve sayings material in the closest available approximation to its earliest written form; in particular, it seems to incorporate a remarkably conservative transmission of some of Jesus’ parables. And whatever the relation of a given saying to its synoptic equivalents, the SC genre itself remains essentially independent of the synoptics.
The SC hypothesis undermines the assumption that Jesus’ sayings passed through an extended period of oral transmission during which there was no perceived need for inscription. It is true that, at least in the case of the oldest material, an interval between production and inscription is plausible, as there is no evidence that Jesus’ sayings were committed to writing during his ministry. What is less plausible is the supposition that the Parable of the Sower was transmitted orally over three or four decades, acquiring a detailed interpretation along the way, before being finally committed to writing by the evangelist Mark. Such a view rests on a questionable analogy with processes of cultural transmission in non- or preliterate societies. It also fails to take into account the evidence of Thomas.
The SC hypothesis outlines a mechanism by means of which a saying of Jesus can pass into wider circulation. The inscription of such a saying is already an interpretative act even if Jesus’ ipsissima verba are faithfully preserved, and it enables an unlimited series of further interpretative acts, including those that gave rise to the canonical gospels. From the outset reception entails an active shaping of the primary material, and the originary moment of the reception process is the event in which the word became text.
1. H. Koester, Ancient Christian Gospels: Their History and Development (Philadelphia: Trinity Press International; London: SCM Press, 1990), p. 80.
2. Koester, Ancient Christian Gospels, pp. 81-82.
3. Koester, Ancient Christian Gospels, p. 83.
4. Koester, Ancient Christian Gospels, p. 87.
5. Koester, Ancient Christian Gospels, pp. 87-99.
6. James M. Robinson, “ΛΟΓΟΙ ΣΟΦΩΝ: On the Gattung of Q” (1971), in his The Sayings Gospel Q: Collected Essays (Leuven: Leuven University Press, 2005), pp. 37-74; pp. 63n, 65-72.
7. Robinson, “ΛΟΓΟΙ ΣΟΦΩΝ,” p. 74.
8. Robinson, “On Bridging the Gulf between Q and the Gospel of Thomas” (1986), in The Sayings Gospel Q, pp. 203-58; pp. 219-45.
9. Robinson, “On Bridging the Gulf,” pp. 243-44.
10. Robinson, “On Bridging the Gulf,” pp. 244, 245.
11. Robinson, “On Bridging the Gulf,” p. 249.
12. Robinson, “On Bridging the Gulf,” p. 250.
13. See April D. DeConick, Recovering the Original Gospel of Thomas: A History of the Gospel and Its Growth (London & New York: T. & T. Clark International, 2005).
14. For texts and introductions, see volumes XXXIII and XX-XXI in the series Nag Hammadi and Manichean Studies (ed. J. M. Robinson and H. J. Klimkeit) and Nag Hammadi Studies (ed. Martin Krause, James M. Robinson, and Frederik Wisse). These volumes are referred to henceforth as NHMS XXXIII and NHS XX or XXI. NHMS XXXIII contains The Apocryphon of John: Synopsis of Nag Hammadi Codices II,1; III,1; and IV,1 with BG 8502,2, ed. Michael Waldstein and Frederik Wisse (Leiden: Brill, 1995). NHS XX-XXI contain Nag Hammadi Codex II,2-7, ed. Bentley Layton, 2 vols. (Leiden: Brill, 1989).
15. For refutation of claims that GTh was originally composed in Syriac or Aramaic, see Simon J. Gathercole, The Composition of the Gospel of Thomas: Original Language and Influences, SNTSMS (Cambridge: Cambridge University Press, 2012), pp. 17-125.
16. Waldstein and Wisse, NHMS XXXIII, p. 7.
17. Waldstein and Wisse, NHMS XXXIII, p. 6. In contrast, Bentley Layton understands Sahidic as “the prestigious and orthodox dialect of the greater Nile valley and monasticism” (NHS XX, p. 7).
18. Waldstein and Wisse, NHMS XXXIII, p. 6.
19. See NHS XX, pp. 6-16. Layton here identifies the language of Codex II as “Crypto-Subachmimic,” and draws attention to a fragmentary version of one of its tractates (On the Origin of the World) in pure Subachmimic (p. 7).
20. NHS XX, p. 4; Nag Hammadi Codices XI, XII, XIII, ed. Charles W. Hedrick (NHS XXVIII) (Leiden: Brill, 1990), p. 362.
21. Waldstein and Wisse, NHMS XXXIII, p. 6.
22. NHS XXI, p. 205.
23. GTh 16.4; 49.1; 75; cf. 23.1-2.
24. So James M. Robinson, “Introduction,” in The Nag Hammadi Library in English (Leiden: Brill, 19842) (henceforth NHL), pp. 16-21. “The headquarters monastery of the Pachomian order at Pabau, where the Basilica of Saint Pachomius was located, as well as the third Pachomian monastery at Chenoboskeia, where Pachomius himself began his Christian life as a hermit, are only 8.7 and 5.3 kilometers . . . respectively from the place where the library was buried” (p. 16).
25. Robinson, “Introduction,” p. 18.
26. So Alastair H. B. Logan, The Gnostics: Identifying an Early Christian Cult (London: T. & T. Clark, 2006), pp. 12-29. Logan appeals especially to evidence from Epiphanius.
27. Enno Popkes argues that the texts included in Codex II have undergone a common redaction: see his Das Menschenbild des Thomasevangeliums: Studien zu seiner religionsgeschichtlichen und theologischen Einordnung, WUNT (Tübingen: Mohr Siebeck, 2007). For discussion of this thesis, see also Jutta Leonhardt-Balzer, “On the Redactional and Theological Relationship between the Gospel of Thomas and the Apocryphon of John,” in Das Thomasevangelium: Entstehung — Rezeption — Theologie, ed. J. Frey, E. E. Popkes, and J. Schröter, BZNW (Berlin & New York: De Gruyter, 2008), pp. 251-71. Criteria for distinguishing redactional assimilation from common motifs or terminology are, however, difficult to establish.
28. Robert M. Grant with David Noel Freedman, The Secret Sayings of Jesus according to the Gospel of Thomas (London: Collins, 1960), p. 16.
29. Cf. the remarkable claim of Michael Fieger, writing as recently as 1991: “Der Thomasjünger ist aufgrund seines Erkenntnisstrebens und seiner Glaubensabwehr Gnostiker und kein Christ” (Das Thomasevangelium: Einleitung, Kommentar, und Systematik [Münster: Aschendorff, 1991], p. 290).
30. See Michael A. Williams, Rethinking “Gnosticism”: An Argument for Dismantling a Dubious Category (Princeton: Princeton University Press, 1996); Karen King, What Is Gnosticism? (Cambridge, MA: Harvard University Press, 2003). For a defence of the traditional view, see the chapter “Gnosticism as a Religion,” in Birger A. Pearson, Gnosticism and Christianity in Roman and Coptic Egypt (New York & London: T. & T. Clark International, 2004), pp. 201-23.
31. Irenaeus, Adv. Haer. i.11.1, 29.1; iii.4.23. On Irenaeus’s usage see Alastair Logan, Gnostic Truth and Christian Heresy: A Study in the History of Gnosticism (Edinburgh: T. & T. Clark, 1996), pp. 1-13.
32. Thus the aim of Book 7 of the Stromateis is “to show the Greeks that only the Gnostic [τὸν γνωστικόν] is truly religious” (Clement of Alexandria, Strom. vii.1.1.1).
33. Irenaeus, Adv. Haer. i.23.4. Irenaeus’s work was known to Eusebius under the title Refutation and Overthrow of Knowledge Falsely So-called [ἐλέγχου καὶ ἀνατροπῆς τῆς ψευδωνύμου γνώσεως], alluding to τὰς βεβήλους κενοϕωνίας και ἀντιθέσεις τῆς ψευδωνύμου γνώσεως (1 Tm. 6.20). This title may be original (Eusebius, HE v.7.1).
34. If the Apocryphon may be said to be characteristically “Gnostic,” that does not mean that all themes normally associated with Gnosticism are necessarily to be found there. Thus Karen L. King argues convincingly that “[i]t would be wrong to suppose that the Secret Revelation of John envisages two types of humanity: those formed after the spiritual image of the first Human and those formed after the psychic likeness of the demiurge and his minions. Rather, all human beings participate in both genealogies” (The Secret Revelation of John [Cambridge, MA: Harvard University Press, 2006], p. 121).
35. The synoptic edition in NHMS XXXIII is indispensable, as the English translation in NHL is based only on the longer version. (See also the synoptic presentation of the BG and NHC II versions in King, Secret Revelation, pp. 26-81.) The longer version includes, and the shorter versions lack, an introduction (Synopsis 1.2-5 = II 1.1-4); a passage on the creator deity, Yaltabaoth (29.10–30.8 = II 11.7-22); additional detail on the primal revelation of the First Man (37.16–38.8 = II 14.20-30); a list of angels responsible for different parts of the body and attributed to the Book of Zoroaster (42.1–50.10 = II 15.29–19.10); and an extended passage of first-person address by “Pronoia” (79.4–82.1 = II 30.12–31.25). Obscurities in the longer versions are frequently clarified by comparison with the shorter one.
36. NHMS XXXIII provides the Latin text only of Irenaeus, Adv. Haer. i.29 (pp. 188-93). In fact, the parallel with the Apocryphon of John resumes at i.30.4, where “her son” refers back to i.29.4. Thus i.30.1-3 is an interpolation, providing a simplified variant of the more complex account of heavenly origins in i.29.1-4. The variant interrupts the parallel with the Apocryphon, which continues up to i.30.10. Throughout, Irenaeus probably follows an earlier heresiological source, perhaps one based on the Syntagma contra omnes Haereses of Justin (1 Apol. 26.8; cf. Tertullian, Adversus Valentinianos 5; Irenaeus, Adv. Haer. iv.6.2; v.26.2; Eusebius, HE iv.11.8-10). On this see F. Wisse, “The Nag Hammadi Library and the Heresiologists,” VC 25 (1971), pp. 205-23; pp. 213-18; for counterarguments, see P. Perkins, “Irenaeus and the Gnostics: Rhetoric and Composition in Adversus Haereses Book One,” VC 30 (1976), pp. 193-200.
37. For a summary of the story told by the Apocryphon, see Gerard P. Luttikhuizen, “The Creation of Man and Woman in The Secret Book of John,” in The Creation of Man and Woman: Interpretations of the Biblical Narratives in Jewish and Christian Traditions, ed. G. P. Luttikhuizen (Leiden: Brill, 2000), pp. 140-55.
38. ApJn II 1.1-4. The Christian frame story is present in both longer and shorter versions, but is not mentioned by Irenaeus. According to H.-M. Schenke, it represents a christianizing of a pre-Christian “Sethian” gnosticism (“The Phenomenon and Significance of Gnostic Sethianism,” in The Rediscovery of Gnosticism: Proceedings of the International Conference at Yale, New Haven, Connecticut, March 28-31, 1978, vol. 2, Sethian Gnosticism, ed. B. Layton [Leiden: Brill, 1981], pp. 588-616; p. 612). The figure of Seth is important because he is the son of Adam, that is, “the Son of Man,” who corresponds to a heavenly prototype identified with Christ (Irenaeus, Adv. Haer. i.30.1, 6, 9, 13; cf. NHC II 24.36–25.2 and parallels). The prominence and literalistic understanding of the “son of man” concept suggest a Gentile Christian origin.
39. ApJn II 31.27-31.
40. ApJn II 31.34-37.
41. ApJn II 32.4-5.
42. ApJn II 13.19-20. John has requested an interpretation of the reference to the Spirit in Gn. 1.2 (LXX). The parallel in BG 45.7-11 reads as follows: “Do you think it is as Moses said, ‘above the water’?”
43. ApJn II 22.22-23; BG 58.16-17, “It is not as Moses said . . .”; III.28.13, “Are you thinking it is as Moses said . . . ?” The reference is to the “trance” (ἔκστασις) of Gn. 2.21a LXX. Cf. also II 23.3-4 (= BG 59.17-19; III 29.21-24), with reference to Gn. 2.21b; II 29.6-7 (= BG 73.4-6; III 37.22-24), with reference to Gn. 7.7.
44. P. Oxy. 654.1; GTh, introduction.
45. The reference to Simon Peter in GTh 13 may also allude to his traditional link with the Gospel of Mark, as attested by Papias (on this see my “The Fourfold Gospel,” in The Cambridge Companion to the Gospels, ed. Stephen C. Barton [Cambridge: Cambridge University Press, 2006], pp. 34-52; pp. 38-39; Richard Bauckham, Jesus and the Eyewitnesses: The Gospels as Eyewitness Testimony [Grand Rapids: Eerdmans, 2006], pp. 235-37). But the main point seems to be the parody of Peter’s confession as attested by Matthew.
46. Individual disciples are referred to on only three further occasions (Mary in Saying 21.1, Salome in 61.2, Simon Peter in 114.1). In none of these cases is Jesus’ teaching withheld from the others.
47. Sayings 6.1; 12.1; 18.1; 20.1; 24.1; 37.1; 51.1; 52.1; 53.1; 99.1; 113.1; cf. Sayings 14.1; 22.2-4; 60.1-6; 72.3; 91.1-2; 100.1-2; 104.1.
48. J. D. Crossan finds in the tension between GTh 12 and 13 the basis for his theory of two strata, “one . . . composed by the fifties C.E., possibly in Jerusalem, under the aegis of James’s authority,” the other “possibly as early as the sixties or seventies, under the aegis of the Thomas authority” (The Historical Jesus: The Life of a Mediterranean Jewish Peasant [Edinburgh: T. & T. Clark, 1991], p. 427). According to Crossan, “the earlier James-layer is now discernible primarily in those units with independent attestation,” whereas “the Thomas-layer is now discernible primarily in that which is unique to this collection” (pp. 427-48). While Crossan here makes the questionable assumption that Thomas is wholly independent of the synoptic gospels, he is right to note the significance of the James/Thomas polarity.
49. The attribution to Thomas would be a further instance of the “secondary authorial fiction” that assigns identities to “the originally unknown authors of the canonical gospels” (Ismo Dunderberg, The Beloved Disciple in Conflict? Revisiting the Gospels of John and Thomas [Oxford: Oxford University Press, 2006], p. 202).
50. ⲡⲧⲏⲣϥ = τὰ πάντα, as in Jn. 1.3.
51. For the Greek text, see Harold W. Attridge (ed.), “The Greek Fragments,” in NHS XX, pp. 95-128.
52. Elsewhere, “He said” at the start of a saying is attested only in Sayings 8.1; 65.1.
53. Other indications of textual fluidity in the existing texts of Thomas include (1) the different locations of the “split the wood saying” (P. Oxy. 1.29-30; GTh 77.2-3); (2) longer and shorter versions of the teaching about anxiety (P. Oxy. 655.1-17; GTh 36); (3) the displacement of Jesus’ answer to the disciples’ questions about religious practices (GTh 6.1; 14.1-5). Nicholas Perrin overlooks this evidence of fluidity in arguing that GTh was “a carefully worked piece of literature, brought together at one place and at one time by an industrious Syriac-speaking editor,” who structured the collection around “multiple catchword collections” (Thomas, the Other Gospel [London: SPCK, 2007], pp. 93, 94; see also his Thomas and Tatian: The Relationship between the Gospel of Thomas and the Diatessaron [Atlanta: SBL, 2002], pp. 49-170).
54. II 138.1-4.
55. II 139.21.
56. For the relationship between these texts, together with the ActsTh, see John D. Turner’s introduction to ThCont, NHS XXI, pp. 173-78; p. 177. Turner proposes a “trajectory,” from sayings collection to dialogue to romance, in which increasing significance is assigned to the figure of Thomas himself. Thomas is Jesus’ twin brother in ThCont (II 138.5-12) and ActsTh (11, 31, 39), though not in GTh. The suggestion that ThCont contains two sources (one in question-and-answer format, the other a monologue), and that the incipit was originally attached to the second of these (pp. 174-76), seems questionable. The “discrepancy” between the opening reference to secret sayings and “the actual genre of the work” (p. 173) is explained by the intertextual connection with the incipit of GTh. Similarly, the emphasis on self-knowledge in ThCont (II 138.8-18, immediately following the incipit) is dependent on GTh 3 (II 33.1-5).
57. Clement of Alexandria, Strom. ii.9.45.5. Clement’s wording, ὁ θαυμάσας βασιλεύσει, assimilates the original θαμβηθεὶς δὲ βασιλεύσ¬ to the context. Greek texts of this and the following passage in Synopsis Quattuor Evangeliorum, ed. K. Aland (Stuttgart: Württembergische Bibelanstalt Stuttgart, 19674), p. 94; O. Stählin, Clemens von Alexandria, II, GCS (Berlin: Akademie-Verlag, 19603).
58. Clement of Alexandria, Strom. v.14.96.3. Clement has just cited a passage from the end of Timaeus (90D), in which the assimilation of perceiving subject to perceived object is seen as the divinely ordained goal of human life.
59. The Coptic version includes two pairings (finds/troubled and troubled/amazed) where the Greek and Clement have just one (finds/amazed). It omits the final pairings in the Greek and Clement (amazed/reign and reign/rest), replacing them with “and he will reign over all.”
60. (1) In P. Oxy. 654, the saying opens with the imperative μὴ παυσάσθω, in Clement with οὐ παύσεται. (2) P. Oxy. 654 probably contained an infinitive absent from the Clement version: . . . ὁ ζη[τῶν τοῦ ζητεῖν ἕως ἄν] εὕρ¬, rather than . . . ὁ ζητῶν ἕως ἄν εὕρη. (3) There are minor differences in the construction of connections: καὶ ὅταν εὕρ¬ and thereafter καί + participle in P. Oxy. 654, εὕρων δέ in Clement and thereafter participle + δέ. At all three points the Coptic version is closer to P. Oxy. 654 than to Clement: (1) ⲙⲛⲧⲣⲉϥⲗⲟ, (2) ⲛϭⲓ ⲡⲉⲧϣⲓⲛⲉ ⲉϥϣⲓⲛⲉ, (3) ⲁⲩⲱ ⲹⲟⲧⲁⲛ ⲉϥϣⲁⲛϭⲓⲛⲉ.
61. The suggestion that GTh incorporates material from GHeb was first made by G. Quispel (“Some Remarks on the Gospel of Thomas,” NTS 5 [1958-59], pp. 276-90; “‘The Gospel of Thomas’ and the ‘Gospel of the Hebrews,’” NTS 12 [1965-66], pp. 371-82; Makarius, das Thomasevangelium und das Lied von der Perle, NovTSupp [Leiden: Brill, 1967], pp. 75-111). For Quispel, GTh represents a conflation of the Jewish encratism of GHeb and the gnosticizing tendencies of GEgy; readings from it are also preserved in the diatessaronic tradition. For a recent discussion, see P. Luomanen, “The Jewish-Christian Gospels and the Gospel of Thomas,” in Das Thomasevangelium: Entstehung-Rezeption-Theologie, ed. J. Frey, E. Popkes, and J. Schröter, BZNW (Berlin & New York: De Gruyter, 2008), pp. 119-53.
62. The same is true of the Gospel according to the Egyptians, to which Clement also refers (Strom. iii.9.63.1, 9.93.1). This may also have been a Thomas-like Sayings Collection. The dialogue with Salome that Clement cites has affinities with GTh 22, 37, 61. For further discussion of these texts, see Chapter 8.
63. Origen knows both of a “Gospel according to Thomas” (Homiliae in Lucam 1; the reference may be to the infancy gospel) and a “Gospel according to the Hebrews” (In Ioan. ii.12). In the latter, according to Origen, “the Saviour himself says: ‘My mother the Holy Spirit took me just now by one of my hairs and carried me away to the great Mount Tabor.’” This passage might have been found in a sayings collection rather than a narrative gospel; the tradition about the Holy Spirit as mother may be echoed in GTh 101.3; 105. Later, Eusebius, Didymus, and Jerome refer to a Gospel according to the Hebrews in the form of an Aramaic expansion of Matthew, identifying this with the original Hebrew Matthew of which Papias speaks. There is no trace of either identification in Clement or Origen. We may therefore differentiate a Greek GHeb attested by Clement and Origen from an Aramaic GHeb (the so-called “Gospel of the Nazareans”) attested by the three later writers influenced by Papias. For the patristic texts, see Elliott (ed.), The Apocryphal New Testament (Oxford: Clarendon Press, 19992), pp. 3-16, to which should be added (1) Eusebius, HE iii.25.4, where GHeb is excluded from the canon and connected especially with “those of the Hebrews who have accepted Christ”; (2) HE iv.22.7, where Hegesippus is said to have quoted from GHeb in Syriac and/or Greek; (3) Didymus the Blind, ad Ps. 33.1, where GHeb is said to correct the synoptic equation of Matthew and Levi (cf Mt. 9.9 = Lk. 5.27), showing that Levi is actually the Matthias of Acts 1. (On this last passage, see Dieter Lührmann, Die apokryph gewordenen Evangelien, NovTSupp [Leiden: Brill, 2004], pp. 182-91.) In spite of Jerome’s explicit references to “the Gospel . . . which I lately translated into Greek and Latin” (De Vir. Ill. 2) and to “the Gospel written in Hebrew which the Nazarenes read” (ad Is. 11.2), Elliott and others assign non-Matthean narratives to the Greek GHeb attested by Clement and Origen: (1) “the story of a woman accused of many sins before the Lord” (Eusebius, HE iii.39.17); (2) an expanded, harmonizing passage about the descent of the Holy Spirit, who addresses Jesus as “my son” (Jerome, ad Is. 11.2); and (3) a post-Easter appearance to James (Jerome, De Vir. Ill. 2). If these stories belonged to the Greek GHeb, it must have been a narrative gospel rather than a sayings collection. But there is no reason why an expanded Aramaic Matthew should not contain additional non-Matthean narrative material. If the narrative passages are reassigned to a Matthew-like Aramaic GHeb, it is possible to view the Greek GHeb as a Thomas-like sayings collection, and perhaps even as an earlier form of GTh. If the Aramaic/Greek distinction is downplayed, however, the references in Eusebius and Didymus may be assigned to the text known to Clement and Origen, resulting probably in a narrative gospel (so Lührmann, pp. 229-58; Andrew Gregory, “Jewish-Christian Gospels,” in The Non-Canonical Gospels, ed. Paul Foster [London & New York: T. & T. Clark, 2008], pp. 54-67).
64. GTh 49.
65. In what follows, I attempt to reconstruct the interpretative logic underlying the Apocryphon of John. On this, see King, Secret Revelation, pp. 215-24; and, for the related material in Irenaeus, Thomas Holsinger-Friesen, Irenaeus and Genesis: A Study of Competition in Early Christian Hermeneutics, JTISupp (Winona Lake, IN: Eisenbrauns, 2009), pp. 56-103. The Apocryphon will be referred to either in its shorter Berlin Gnostic codex version or in the longer Codex II version. Occasional references to Codex III are to the alternative translation of the shorter version in Nag Hammadi Codex III.
66. The interpretation of the image of God as a divine hypostasis is already present in Philo, for whom it is the archetypal pattern for the whole created world, the νοητὸς κόσμος which is also ὁ θεοῦ λόγος (De Opificio Mundi 25). In the Apocryphon, the specific link with humanity is evident in the revelation of the heavenly world recounted in BG 47.14–48.14 (II 14.13–15.4).
67. BG 50.15–51.1, II 19.13-14; cf. Irenaeus, Adv. Haer. i.30.6. The difference between the incorporeal androgynous human of Gn. 1 and the embodied male figure of Gn. 2 is already noted by Philo (De Opificio Mundi 134).
68. This deity claims to be “God,” but his claim is false (II 11.19-22, cf. II 12.8-9; AH i.29.4). In Gn. 1.27, however, “in the image of God” is taken to refer to the true God rather than the creator (cf. BG 47.20–48.10, II 14.18–15.4). The creator’s subordinate status also reflects the Demiurge of Plato’s Timaeus (cf. BG 44.7-9, II 12.33–13.5). Yet, “[r]ather than declare, as Plato did, that mimesis ensures that the mundane world is the best possible, the Secret Revelation of John exposes these likenesses as fundamental deception” (King, Secret Revelation, p. 94).
69. BG 38.10-14, II 15.1-4; cf. Philo, De Opificio Mundi 72-76; Quaestiones et Solutiones in Genesin i.4.
70. BG 27.1–28.4; II 4.22–5.11.
71. On this intertextual connection with the Wisdom of Solomon, see King, Secret Revelation, pp. 149-50. As King notes, however, the later text “split[s] the figure of Wisdom into two higher and lower characters: Pronoia and Sophia” (p. 226).
72. II 5.5-7.
73. BG 44.19–45.19; II 13.13-26.
74. BG 47.15–48.14; II 14.14–15.4; Irenaeus, Adv. Haer. i.30.6.
75. BG 51.1–52.1; II 19.15-33; Irenaeus, Adv. Haer. i.30.6.
76. Karen King suggests that this text “was plausible and persuasive to some people because it offered solutions to difficulties they had in reading Genesis” — difficulties with an anthropomorphic deity, for example (Secret Revelation, p. 221). That may underestimate the extent to which the Apocryphon’s reading of Genesis stems from genuine exegetical observations.
77. HypAr 87.23-27. This passage conflates Gn. 1.26 and 2.7. In consequence of the “let us create” of the former passage, the plural ⲁⲩⲣⲡⲗⲁⲥⲥⲉ replaces the singular ἔπλασεν of the latter. “Come, let us . . .” is also influenced by Gn. 11.7.
78. HypAr 89.3-7.
79. HypAr 92.4-10. “The Powers” (ⲛⲇⲩⲛⲁⲙⲓⲥ) are probably identical to “the Rulers.”
80. The opening of the Hypostasis of the Archons explicitly refers to this teaching of “the great Apostle” (HypAr 86.1-27). The assumption that the Christian frame of this text is secondary (so Roger A. Bullard, NHS XX, p. 222) seems to reflect the old view of “Gnosticism” as an independent pre-Christian religion.
81. Most of the passages that April DeConick assembles under the heading “The Primordial Adam and the Encratic Ideal” do not require an Adam-related interpretation (Recovering the Original Gospel of Thomas, pp. 166-67; see also her Seek to See Him: Ascent and Vision Mysticism in the Gospel of Thomas, VCSupp [Leiden: Brill, 1996], pp. 16-21). Is there any clear evidence that “Thomasine Christians sought to recreate the youth of Adam within the present experience of their community” (Recovering, p. 189)?
82. A connection between the motifs of greatness, the kingdom, and childlikeness also occurs in Mt. 18.1-3.
83. Cf. Mt. 18.2-6; Mk. 9.37, 42.
84. In contrast, April DeConick regards the whole saying as part of the “kernel” of GTh, with the exception of “this person will know the kingdom” (Recovering the Original Gospel of Thomas, p. 70).
85. Mt. 11.9-10 = Lk. 7.26-27, citing Ex. 23.20 and/or Mal. 3.1. This passage might have been composed to connect Mt. 11.7-8 with 11.11, using a composite scriptural citation derived from Mk. 1.2.
86. The individual elements would then be (i) the reply to the Baptist’s messengers (Mt. 11.1-6), followed by (ii) the saying on the reed shaken by the wind (11.7-8) and (iii) its application to the Baptist (11.9-10); together with the Baptist-related sayings on (iv) greatness (11.11), (v) violence (11.12), (vi) the law and the prophets (11.13), (vii) Elijah (11.14-15), and (viii) this generation (11.16-19).
87. For discussion of the relationship of Sayings 46 and 52, see M. Moreland, “Thomas 52 as Critique of Early Christian Hermeneutics,” in Thomasine Traditions in Antiquity: The Social and Cultural World of the Gospel of Thomas, ed. Jon M. A. Asgeirsson, April D. DeConick, and Risto Uro (Leiden & Boston: Brill, 2006), pp. 75-91; pp. 86-88.
88. GTh 85 has also been seen as the third of a trilogy of sayings interpreting key themes in Gn. 1–2 (so Stevan L. Davies, The Gospel of Thomas and Christian Wisdom [California: Bardic Press, 20052], pp. 63-69). But in GTh 83-84 the difficult references to “the images,” “your likeness,” and “your images” may be only indirectly related to Genesis. See however Enno Popkes, “The Image Character of Human Existence: GThom 83 and GThom 84 as Core Texts of the Anthropology of the Gospel of Thomas,” in Das Thomasevangelium, pp. 416-34.
89. Nicholas Perrin finds here one of a number of parallels between GTh and Tatian, who is criticized by Irenaeus (Adv. Haer. i.28.1; iii.23.8) for denying the salvation of Adam (Thomas, the Other Gospel, pp. 121-22). Other parallels are found in Thomas’s asceticism (vegetarianism, GTh 11; celibacy, 49, 75), and in verbal links with the Diatessaron, which are said to show that Thomas is dependent on Tatian throughout (pp. 81-106). In view of the fundamental differences between Thomas and the Diatessaron, it seems more likely that the parallels represent common traditions, textual or otherwise. In the case of the Adam saying, the common origin would be the Pauline claim that “in Adam all die” (1 Cor. 15.22; Irenaeus, Adv. Haer. iii.23.8).
90. τὸν ἄνθρωπον ὅν ἔπλασεν (Gn. 2.15) becomes Αδαμ thereafter, either with the article (2.16, 19, 20, 21, 22 [2x], 25) or without it (2.19, 20, 23); ἄνθρωπος is retained in 2.18. The MT reads 
in each case; only at 4.25–5.3 is the article omitted (although there are ambiguous cases at 2.20; 3.17, 21).
91. See Philo, De Opificio Mundi 134; Quaestiones et Solutiones in Genesin i.4.
92. As Oscar Cullmann notes, v. 46 “is meaningful only if Paul thinks of a doctrine which asserts just what he denies here”; at this point, “Paul deliberately abandons — in fact, expressly attacks — the Philonic doctrine” (The Christology of the New Testament, Eng. trans. [London: SCM Press, 1959], p. 177). Arguments against this interpretation (e.g. C. Wolff, Der erste Brief des Paulus an die Korinther, ThHKNT 7/II [Berlin: Evangelische Verlagsanstalt, 1982], pp. 202-3) seem to me to be beside the point. Given the clear links between Philo, GTh 85, ApJn, and other Genesis-related texts, it is hardly possible to speak of a “Spärlichkeit ausserphilonischer Zeugnisse” (p. 202). The fact that Paul, unlike Philo, “geht . . . auf Gen. 1,27 gar nicht ein” (p. 202) is precisely the point: Paul here rejects the Genesis exegesis that produces a primal man in the image of God and focuses instead on an eschatological “last Adam” (1 Cor. 15.45). For Philo, it is human origin that provides the key to human nature; for Paul, human destiny.
93. The primary allusion in 1 Cor. 15.49 is to Gn. 5.3 rather than 1.26-27.
94. April DeConick rightly notes that this passage locates the original error not in the godhead but in the human (Seek to See Him, p. 16). Her suggestion that “a Great Power” is a reference to Christ seems unlikely, however, since the power and wealth must belong to Adam himself (pp. 16-17).
95. Compare the conclusion of A. Marjanen, that GTh is more closely related to the Gospel of John than to the Apocryphon of John or to the Gospel of Philip (“Is Thomas a Gnostic Gospel?” in Thomas at the Crossroads: Essays on the Gospel of Thomas, ed. R. Uro, SNTW [Edinburgh: T. & T. Clark, 1998], pp. 107-39; p. 138).
96. II 32.10-19.
97. On “secondary orality,” see Risto Uro, “Thomas and Oral Gospel Tradition,” in Thomas at the Crossroads, pp. 8-32; Thomas: Seeking the Historical Context of the Gospel of Thomas (London & New York: T. & T. Clark, 2003), pp. 106-33.
98. Uro rightly points to GTh’s “mixture of early-looking traditions with features that very probably derive from the canonical gospels” (Thomas: Seeking the Historical Context, pp. 130-31). A theory of composition must therefore explain “both the influence of the canonical gospels and Thomas’ access to traditions that are clearly independent of the canonical gospels” (p. 132). See too the balanced comments of Jörg Frey, “Die Lilien und das Gewand: EvThom 36 und 37 als Paradigma für das Verhältnis des Thomasevangeliums zur synoptischen Überlieferung,” in Das Thomasevangelium, pp. 122-80; pp. 176-80.
99. This distinction between independence of format and of content is rightly drawn by Jens Schröter, for whom the presentation of sayings without contexts indicates “daß dieses [das EvThom] ein ganz eigenes Konzept verfolgt (hier ist der Terminus ‘unabhängig’ sicher angebracht), daß dies jedoch andererseits kein ausreichendes Indiz ist, um auch eine sprachliche Beeinflussung auszuschliessen” (Erinnerung an Jesu Worte: Studien zur Rezeption der Logienüberlieferung in Markus, Q und Thomas, WMANT [Neukirchen-Vluyn: Neukirchener Verlag, 1997], p. 139). But it should not be too quickly concluded that the simple format of GTh is original to this work.
100. For an attempt along these lines, see DeConick, Recovering the Original Gospel of Thomas.
101. For “Sayings Collections” as an early Christian genre, see Helmut Koester, From Jesus to the Gospels: Interpreting the New Testament in Its Context (Minneapolis: Fortress, 2007), pp. 26-28. Koester appeals to Q, Thomas, Justin, and 2 Clement, but concedes that “[e]vidence for the continued existence of sayings collections is not easy to obtain” (p. 27). Paradoxically, such evidence as there is comes more sharply into focus in the absence of Q, since criteria for identifying SC material can then be derived from GTh alone.
102. A typical example of the overblown rhetoric accompanying this claim, complete with multiple mixed metaphors: “The objectifying, controlling power of the written medium, while taking the life out of spoken language, can freeze oral forms and preserve them in fossilized profiles” (The Oral and the Written Gospel: The Hermeneutics of Speaking and Writing in the Synoptic Tradition, Mark, Paul, and Q [Philadelphia: Fortress, 19831, 19972], p. 44). In a more recent interview, Kelber partially retracts this claim, seeing a “conflictual relationship” between orality and literacy only as one possibility among their many potential modes of interaction (Werner H. Kelber and Tom Thatcher, “‘It’s Not Easy to Take a Fresh Approach’: Reflections on The Oral and the Written Gospel,” in Jesus, the Voice, and the Text: Beyond the Oral and the Written Gospel, ed. Tom Thatcher [Waco, TX: Baylor University Press, 2008], pp. 27-43; p. 30). For Kelber’s account of GTh as an “interface” between orality and textuality, see his “In the Beginning Were the Words: The Apotheosis and Narrative Displacement of the Logos,” JAAR 58 (1990), pp. 69-98, esp. pp. 78-82.
103. On this see Helmut Koester, Synoptische Überlieferung bei den Apostolischen Vätern (Berlin: Akademie-Verlag, 1957), pp. 62-70; Ancient Christian Gospels, pp. 349-59; Andrew F. Gregory and Christopher Tuckett, “2 Clement and the Writings That Later Formed the New Testament,” in The Reception of the New Testament in the Apostolic Fathers (Oxford: Oxford University Press, 2005), pp. 251-92.
104. With reference to the repeated λέγει of the Greek fragments, Lührmann speaks of “die damit signalisierte Zeitlosigkeit der einzelnen Worte Jesu” (Lührmann, Die apokryph gewordenen Evangelien, p. 178). The Coptic translation decision may make this less likely.
105. So Gregory and Tuckett, “2 Clement,” p. 255.
106. 2 Clem. 8.5.
107. As argued initially by H. Koester (Synoptische Überlieferungen, pp. 101-2). Later, Koester rightly sees in this passage “possibly the earliest evidence for the designation of a sayings collection as ‘gospel’” (Ancient Christian Gospels, p. 355).
108. 2 Clem. 12.2.
109. 2 Clem. 12.3.
110. 2 Clem. 12.4.
111. The author addresses his hearers as ἀδελϕοὶ καὶ ἀδελϕαί in 19.1; 20.2.
112. 2 Clem. 12.5.
113. Clement of Alexandria, Strom. iii.13.92.2.
114. Clement of Alexandria, Strom. iii.13.93.1.
115. Clement of Alexandria, Strom. iii.13.95.2.
116. In his early work Koester argued that the authors of 2 Clement and GEgy are both here “von einem frei umlaufenden Apophthegma abhängig” (Synoptische Überlieferungen, p. 104). Later, he states that this passage “must have been circulating in the free tradition of sayings from which the author of the sayings collection that 2 Clement knew and used obtained also other non-canonical materials” (Ancient Christian Gospels, pp. 359-60). Both statements presuppose the form-critical assumption of a homogeneous ongoing oral tradition whose individual components may, sooner or later, be independently reduced to writing. Koester does not see that his own sayings collection hypothesis — adopted no doubt under the influence of GTh, not yet available to him in the earlier work — actually represents an alternative mechanism for the transmission of Jesus’ sayings, making the “free” oral tradition hypothesis redundant.
117. Clement of Alexandria, Strom. iii.9.64.1 (cf. iii.6.45.3); iii.9.66.1-2. Clement provides two versions of the first half of this dialogue. My rendering prefers the second version of Salome’s question to the first (where she asks, “How long shall death prevail?”) and the first version of Jesus’ response to the second (“you women” rather than “women”).
118. Clement of Alexandria, Strom. iii.17.104.1.
119. Clement of Alexandria, Strom. iii.9.66.3–67.1.
120. Clement of Alexandria, Strom. iii.9.63.2.
121. Clement of Alexandria, Strom. iii.13.92.2.
122. GTh 37.1-3; cf. P. Oxy. 655, i.17-23.
123. GTh 22.1-7.
124. 2 Clem. 2.4 (= Mt. 9.13).
125. 2 Clem. 3.2. Cf. Mt. 10.32: “Everyone who will confess [in] me before people, I too will confess in him before my Father who is in heaven.”
126. 2 Clem. 4.2. Cf. Mt. 7.21: “Not everyone who says to me, ‘Lord, Lord,’ will enter into the kingdom of heaven, but the one who does the will of my Father who is in heaven.”
127. 2 Clem. 4.5. Cf. Mt. 7.23: “And then I will confess to them, ‘<I never knew you;> depart from me, workers of lawlessness.’” Lk. 13.27: “[ ] And he will say to you, ‘<I do not know you [or] where you are from;> depart from me, all workers of unrighteousness.’” According to Koester, “the source of 2 Clem. 4.2, 5 was probably the same collection of sayings that is used elsewhere in this writing, that is, a collection based on Matthew and Luke that also incorporated sayings from the free tradition” (Ancient Christian Gospels, p. 357).
128. 2 Clem. 5.2-4. Cf. Lk. 10.3: “Behold I send you as lambs [ἄρνας] in the midst of wolves” (Mt. 10.16a, πρόβατα). Lk. 12.4-5: “Do not fear those who kill the body and after that have nothing further to do. . . . Fear the one who after killing has authority . . . to cast into Gehenna.” P. Oxy. LX.4009: “[Be blame]less as the [doves a]nd wise [as the snakes]. You will be as [lambs in the midst] of wolves.” [I said to hi]m: “What then if we [are torn]?” [He answering] says to me: [ ] “The [wolves tear]ing the [lamb can no] longer do any[thing] to it. There[fore] I say to y]ou: Do [n]ot f[ear th]ose who k[ill yo]u and [after killing] are no lon[ger a]ble [to do anything].” On the relationship between the parallel passages from 2 Clement and P. Oxy. 4009, see Lührmann, Die apokryph gewordenen Evangelien, pp. 73-86.
129. 2 Clem. 6.1-2. Cf. Lk. 16.13: “No servant can serve two masters. For either he will hate the one and love the other, or he will be devoted to the one and despise the other. You cannot serve God and Mammon.” Mt. 16.26: “For what will it advantage [ὠϕεληθήσεται] a person, if he gain the whole world and lose his soul?”
130. 2 Clem. 8.5. Cf. Lk. 16.12: “And if you are not faithful in another’s, who will give you what is your own?” Lk. 16.10a: “The one who is faithful in little is also faithful in much.”
131. 2 Clem. 9.11. Cf. Mt. 12.50: “<For whoever does the will of my Father who is in heaven> is my brother and sister and mother.”
132. 2 Clem. 12.2.
133. 2 Clem. 13.4. The reference here to “God” as speaker, rather than “the Lord,” may reflect the conviction that “we must think about Jesus Christ as about God” (2 Clem. 1.1). Cf. Lk. 6.32: “And if you love those who love you, what credit is it to you?” Lk. 6.27 (cf. v. 35): “Love your enemies and do good to those who hate you.”
134. 2 Clem. 17.4.
135. According to Gregory and Tuckett, 2 Clement may have used “a post-synoptic harmony for some of his traditions, but other, apocryphal gospels for other traditions” (“2 Clement,” p. 278). Given that GTh contains both sayings dependent on Matthew and Luke and independent sayings, it is plausible to suppose a single SC in the case of 2 Clement.
136. Gregory and Tuckett conclude that 2 Clement presupposes Matthew’s finished gospel in the case of 2 and 3, and Matthew and Luke in the case of 8 (“2 Clement,” pp. 258-60, 270). Koester takes a similar view (Ancient Christian Gospels, pp. 349-53).
137. P. Oxy. LX.4009 (see above); GEgy.
138. 2 Clement includes seven citations from the LXX. (1) 2 Clem. 2.1 = Is. 54.1 LXX. (2) 2 Clem. 3.5; Is. 29.13: minor differences with LXX are mostly accounted for by Mt. 15.8. (3) 2 Clem. 6.8, summarizing Ez. 14.14, 18, 20. (4) 2 Clem. 7.6 = Is. 66.24b. (5) 2 Clem. 13.2 = Is. 52.5 (+ πᾶσιν); there is an additional unidentified citation here. (6) 2 Clem. 14.1; Jer. 7.11: minor differences to accommodate the passage to its new context. (7) 2 Clem. 14.2, summarizing Gn. 1.27. An unidentified prophetic word is cited in 2 Clem. 11.2, also occurring in 1 Clem. 23.3-4 where it is regarded as “scripture.”
139. Justin, 1 Apol. 15.1-4. On Justin’s gospel citations, see Koester, Ancient Christian Gospels, pp. 360-402. Koester argues that, like 2 Clement, Justin draws from “collections of sayings which were composed on the basis of harmonized gospel texts and which incorporated additional sayings from the non-canonical tradition” (p. 374). On this specific passage, see A. J. Bellinzoni, The Sayings of Jesus in the Writings of Justin Martyr (Leiden: Brill, 1967), pp. 57-61, 87-88, 96-97. According to Bellinzoni, “Justin did not quote the sayings of Jesus from memory but . . . used one or more written sources often quoted by other fathers in a form almost identical to Justin’s version . . . , most often and most strikingly in Clement of Alexandria, Pseudoclementine Homilies, and Origen” (p. 95). Bellinzoni’s claim is assessed positively by Annewies van den Hoek (“Divergent Gospel Traditions in Clement of Alexandria and Other Authors of the Second Century,” Apocrypha 7 [1996], pp. 43-62), but more sceptically by G. Strecker (“Eine Evangelienharmonie bei Justin und Pseudoklemens?” NTS 24 [1977-78], pp. 297-316).
140. “If your right eye cause you to stumble, cut it out [ ]; for it is better for you to enter one-eyed into the kingdom of heaven than with two to be sent into the eternal fire.” Underlining = substitution from Mt. 5.30; italics, from Mt. 18.8-9.
141. Cf. Lk. 11.52: “Woe to you lawyers, for you took the key of knowledge. You did not enter and those wishing to enter you prevented.” Mt. 23.13: “Woe to you, <scribes and Pharisees>, hypocrites! For you shut the kingdom of heaven. . . .” Mt. 10.16: “Behold, I send you as sheep in the midst of wolves. Be then wise [ϕρόνιμοι] as serpents and innocent [ἀκέραιοι] as doves.”
142. Clement of Alexandria, Strom. iii.13.92.2; cf. GTh 22, 37.
143. Cf. Lk. 16.13: “No servant can serve two masters. For either he will hate the one and love the other, or he will be devoted to the one and despise the other. You cannot serve God and Mammon.” Mt. 16.26: “For what will it advantage a person, if he gain the whole world and lose his soul?”
144. Mt. 6.19-20 is here juxtaposed with Mt. 16.26 (6.20 being repeated). Divergences from Matthean wording are slight. See Bellinzoni, Sayings, pp. 61-62, 89-90 (although Bellinzoni does not focus on the phenomenon of juxtaposition).
145. “If we wish to serve God and Mammon . . .” derives from the conclusion of the two-masters saying (cf. Mt. 6.24c = Lk. 16.13c), while “. . . it is unprofitable to us” serves to introduces the commercial imagery of the following saying.
146. Cf. Lk. 11.27-28: “And it came to pass as he said these things, a certain woman from the crowd raising her voice said to him, ‘Blessed are the womb that bore you and the breasts that fed you!’ And he said [ ]: ‘Blessed rather are those who hear the word of God and keep it [ ].’” Lk. 23.29: “. . . for behold, the days are coming when they will say: ‘Blessed are the barren and the wombs that have not conceived and the breasts that have not fed.’”
147. Cf. Mt. 26.24: “. . . but woe to that man by which the Son of man is betrayed. It were good for that man if he had not been born.” Mt. 18.6: “<Whoever offends one of these little ones who believe in me,> it would be well for him that a mill-wheel be hung about his neck and he be drowned in the depth of the sea.”
148. P. Oxy. 1r, ll. 2-9. The reading is uncertain, however.
149. As in Logion 2, ⲡⲧⲏⲣϥ probably represents τὰ πάντα (as in Jn. 1.3).
150. Questions: GTh 12.1; 21.1; 37.1; 43.1; 51.1; 53.1; 113.1. Requests: GTh 18.1; 20.1; 24.1; 72.1; 91.1; 114.1. Statements: GTh 52.1; 79.1; 99.1; 100.1. Suggestions: GTh 104.1. Requests that Jesus should “tell” or “show” are interchangeable with questions. Requests for other types of action are found only in GTh 72.1 (“Tell my brothers to divide my father’s possessions with me”) and 114.1 (“Let Mary leave us, for women are not worthy of life”). In GTh 72.2-3, two alternative responses are provided, one addressed to the man who has made the request, the other to the disciples.
151. Clement of Alexandria, Strom. iii.13.92.2. Koester notes that the 2 Clem. 12 version represents the “oldest and most original form” of this tradition (Ancient Christian Gospels, p. 359).
152. Clement of Alexandria, Strom. iii.9.64.1 (cf. iii.6.45.3); iii.9.66.1-2.
153. See Dunderberg, The Beloved Disciple in Conflict? pp. 89-101.
154. Compare the request + statement in GTh 24, and the statement + question in GTh 12.
155. Clement of Alexandria, Strom. v.14.93.3; cf. ii.9.45.5: “So also in the Gospel to the Hebrews it is written: ‘The one who wonders shall reign, and the one who has reigned shall rest.’”
156. Cf. Mt. 5.7: “Blessed are the merciful, for they shall receive mercy.”
157. Cf. Lk. 6.37c: “Forgive, and you will be forgiven.”
158. No parallel.
159. Cf. Lk. 6.38a: “Give, and it shall be given to you.”
160. Cf. Mt. 7.2a: “With what judgment you judge, you will be judged.”
161. No parallel.
162. Cf. Mt. 7.2b; Lk. 6.38b: “With the measure you measure, [ ] it will be measured to you.”
163. Against Koester, who claims that 1 Clem. 13.2 was “drawn from the oral tradition” (Ancient Christian Gospels, p. 20). Koester had earlier suggested that the passage may have been constructed by the author of 1 Clement “aus irgendeiner schriftlichen Herrenwortsammlung, die wir nicht mehr kennen, die aber älter sein mag als unsere Evangelien” (Synoptische Überlieferungen, p. 16).
164. It is also cited in the same form by Clement of Alexandria (Strom. ii.18.91), who derives it from 1 Clement, a text well known to him.
165. Cf. A. Gregory’s discussion of this passage (“1 Clement and the Writings That Later Formed the New Testament,” in The Reception of the New Testament, pp. 129-57; pp. 131-39). Gregory concludes “that Clement refers here to a collection of sayings that is independent of and earlier than the broadly similar sayings of Jesus that are preserved also in Matthew and/or Luke” (pp. 133-34).
166. According to Koester, Polycarp draws the first passage from 1 Clem. 13.2, altering the order of the imperatives and bringing the wording into line with Matthew and Luke respectively at the beginning and end (Synoptische Überlieferungen, pp. 115-18). A derivation of both passages from a Sayings Collection influenced by Luke and Matthew seems at least as likely.
167. Cf. Mt. 5.3, 10, 11; Lk. 6.20, 22.
168. That a number of sayings in GTh betray the influence of Matthew and Luke has been well demonstrated by Simon Gathercole, who gives extensive attention to the relevant methodological issues (The Composition of the Gospel of Thomas, pp. 127-224). It should be noted that GTh’s “dependence on the synoptics” would not establish this work’s secondary, postcanonical status, since both Matthew and Luke are also dependent on at least one of the synoptic gospels.
169. Kelber, The Oral and the Written Gospel, pp. 91, 95, 93.
170. This “anti-grammatological” tendency is a central concern in the earlier work of Jacques Derrida: see his Of Grammatology, Eng. trans. (Baltimore & London: Johns Hopkins University Press, 1976); “Plato’s Pharmacy,” in Dissemination, Eng. trans. (Chicago: University of Chicago Press, 1981), pp. 61-171; “The Linguistic Circle of Geneva,” in Margins of Philosophy, Eng. trans. (Chicago: University of Chicago Press, 1982), pp. 137-53.
171. Eusebius, HE iii.39.3.
172. Eusebius, HE iii.39.1.
173. In his preface Papias writes, “I also will not hesitate to compile, along with the interpretations [συγκατατάξαι ταῖς ἑρμηνείαις], whatever I have learned well and remembered well from the elders” (Eusebius, HE ii.39.3). These traditions of the elders are thus supplementary to the exegesis or interpretation of the Lord’s sayings, about which Papias must have spoken in the immediately preceding passage. On this see A. D. Baum, “Papias als Kommentator evangelischer Aussprüche Jesu: Erwägungen zur Art seines Werkes,” NovT 38 (1996), pp. 257-76; pp. 270-73. According to Baum, referring to HE iii.39.3, Papias’s work was in three parts: “Erstens dürfte Papias, wie in vergleichbaren Werken in der Regel üblich, die von ihm behandelten Herrensprüche (λόγια κυριακά) zitiert haben. Zweitens hat er diese Herrensprüche mit Erläuterungen (ἑρμηνείαι) versehen. Und drittens hat er in diese Auslegungen Mitteilungen der Presbyter (ὅσα ποτὲ παρὰ τῶν πρεσβυτέρων καλῶς ἔμαθον καὶ καλῶς ἐμνημόνευσα) eingefügt” (A. D. Baum, “Papias, der Vorzug der Viva Vox und die Evangelienschriften,” NTS 44 [1998], pp. 144-51; p. 147). Since Papias states that “Matthew compiled the sayings [τὰ λόγια συνετάξατο]” (Eusebius, HE iii.39.16), it is likely that Matthew is a primary source of the Sayings Collection that Papias himself compiled and interpreted.
174. For a thorough discussion of the “vormarkinische Sammlung” underlying Mk. 4.1-34, see Heinz-Wolfgang Kuhn, Ältere Sammlungen im Markusevangelium (Göttingen: Vandenhoeck & Ruprecht, 1971), pp. 127-46. Kuhn draws attention to (1) the catchword linkage between the three parables (σπείρειν, vv. 3, 31, 32, to which should be added γῆ, vv. 5, 8, 26, 28, 31 [2x]; καρπός, vv. 7, 8, 29; βασιλεία τοῦ θεοῦ, vv. 26, 30); (2) the non-Markan introductory formula καὶ ἔλεγεν (vv. 9, 26, 30); (3) the double conclusion (vv. 33, 34); (4) an inconsistency in relation to the addressees (vv. 10, 33); and (5) the shift from an original emphasis on comfort in adversity to the present concern with secrecy (pp. 129-136). Unfortunately Kuhn fails to engage with GTh and concludes from the catchwords that such a collection could have taken purely oral form (p. 144). On the general issue of pre-Markan sources and traditions, there is an extensive bibliography in W. R. Telford, Writing on the Gospel of Mark (Blandford Forum: Deo Publishing, 2009), pp. 338-41.
175. I here follow Bultmann, who sees Mk. 4.11-12 as “ganz sekundär” (Die Geschichte der synoptischen Tradition [Göttingen: Vandenhoeck & Ruprecht, 19677], p. 215). In v. 10, σὺν τοῖς δώδεκα is Mark’s addition to his source (p. 71), but more significant is his substitution of “parables” for “parable”: “In V.10 wird nach dem Sinn der Parabelrede überhaupt gefragt, und darauf antwortet V.11f. Aber V.13 setzt voraus, daß nach dem Sinn der eben erzählten Parabel gefragt worden ist. Die Frage in V.10 muß also in der Quelle etwa gelautet haben wie Lk 8,9” (p. 351n). Such an approach is sharply criticized on narrative-critical grounds by J. Schröter (Erinnerung, pp. 301-9). While narrative-critical analysis may well lead one to question a given literary-critical hypothesis, the two approaches are not mutually exclusive in principle.
176. GTh 24, 51, 52, 53; cf. 20, 99 (“The disciples said to him”), 91 (“They said to him”).
177. The possible derivation of the Markan καὶ ἒλεγεν (4.21, 24, 26, 30, etc.) from a “Spruchsammlung” such as P. Oxy. 1 is already noted by Bultmann (Geschichte, p. 349). Bultmann assumes that the introductory formula λέγει ὁ Ἰησοῦς would have been original here: “Wurde eine derartig angelegte Sammlung in ein Evangelium aufgenommen, so war es natürlich, daß ὁ Ἰησοῦς fortfiel und etwa das Präsens in ein Tempus der Vergangenheit umgesetzt wurde” (p. 349). For Bultmann, a format such as this represents a primitive, presynoptic stage in the collecting of Jesus’ sayings (p. 348).
178. Compare the analysis of J. D. Crossan, In Parables: The Challenge of the Historical Jesus (New York: Harper & Row, 1973), pp. 39-44. Crossan too believes that the Markan interpretation has affected the parable itself, and draws attention especially to the intrusive reference to the sun (Mk. 4.6a), which aims to show what tribulation or persecution (4.16-17) is really like: “not slow withering but instant scorching” (p. 42).
179. This would be the case even if one accepted J. Schröter’s claim that Thomas’s version “läßt sich . . . am ungezwungensten als eine nachsynoptische Rezeption dieses Gleichnisses verstehen” (Erinnerung, p. 318). The text as a whole does not appear to require any such conclusion, however.
180. GTh 1.
181. The familiar critical point about the secondariness of the interpretation must be differentiated from the negative evaluation that often accompanies it. Thus J. Wellhausen writes: “Er [d.h. der Kommentar] ist später als die Parabel und kann nicht von Jesus selbst herrühren. Das Wort im Sinne des Evangeliums, die Verfolgung wegen des Evangeliums und der Abfall davon liegen außerhalb des Gesichtskreises seiner Gegenwart; die apostolische Gemeinde wird vorausgesetzt und tritt an stelle des jüdischen Auditoriums” (Das Evangelium Marci Übersetzt und Erklärt [Berlin: Georg Reimer, 19092], p. 32). In a nutshell, the problem is that the interpretation belongs within a “christlich-kirchliche” context, not that of a simple “Lehrer der Juden” (p. 32). There is no sense here that a logical development might be traceable between a parable, its inscription, and its ongoing interpretation within early Christian communities.
182. The dependence of GTh 33 on Luke is rightly noted by Schröter, Erinnerung, pp. 374-75; Gathercole, The Composition of the Gospel of Thomas, pp. 194-96.
183. This would then constitute the core of “eine schon dem Mk vorliegende kleine Sammlung” (Bultmann, Geschichte, p. 351). Bultmann believes, questionably, that this grouping of thematically linked parables could have occurred at the oral stage (p. 348).
184. Jl. 4.13: ἐξαποστείλατε δρέπανα ὅτι παρέστηκεν ὁ τρύγητος. Mk. 4.29: εὐθὺς ἀποστέλλει τὸ δρέπανον ὅτι παρέστηκεν ὁ θερισμός. Note that in the Markan context the ὅτι-clause is redundant.
185. See Richard Bauckham, “The Parable of the Vine: Rediscovering a Lost Parable of Jesus,” NTS 33 (1987), pp. 84-101; pp. 93-94. “Like Jesus’ parables in general, it [the GTh Parable of the Mustard Seed] does not break the bounds of naturalism and introduce mythological features such as are found in the Old Testament texts about the world-tree. . . . The other versions sacrifice naturalistic accuracy to closer correspondence with the Old Testament texts” (p. 94).
186. Dn. 4.20-21 (Theodotion) reads as follows: “The great and strong tree [δένδρον = Mt., Lk.] that you saw . . . Under it the wild beasts made their home, and in its branches [ἐν τοῖς κλάδοις αὐτοῦ = Mt., Lk.] the birds of the sky dwelt [κατεσκήνουν τὰ ὄρνεα τοῦ οὐρανοῦ = Mk, Mt, Lk].”
187. As usual, the Q theory introduces unnecessary complications here. Thus John Dominic Crossan postulates a Q version of the parable, best preserved in Luke, which then provides supporting evidence for a reconstructed pre-Markan version (In Parables, pp. 44-49). Crossan is right, however, to point out that “the attempt to allude to the tree with the nesting birds of apocalyptic eschatological imagery” is secondary, and that GTh is here “much closer to the original,” i.e. presumably the original inscription (p. 49).
188. Note, however, Simon Gathercole’s critique of the assumption that an apparently simpler or more coherent version of a passage is necessarily earlier (The Composition of the Gospel of Thomas, pp. 132-38).