Why was an Egyptian-language fragment of Plato’s Republic buried in a jar, in the middle of the desert, five hundred miles south of Alexandria?

Working Title: The Lion Within: Plato’s Republic in Roman Egypt


The discovery, at Nag Hammadi in 1945, of a sealed jar filled with religious manuscripts reinvigorated the study of Gnosticism. One of the more confounding elements, found within the sixth codex, is a Coptic-language excerpt of the ninth book of Plato’s Republic. Building upon previous scholarship, this paper will contend that the excerpt was included as a necessary reference for understanding the seventh logion of The Gospel of Thomas, a non-canonical early Christian wisdom-text. Plato Rep. IX 588a-589b relates a parable that solves this difficult saying of the risen Jesus. When, in the 4th century, an audience member could not be guaranteed a background in classical Greek philosophy, the Coptic excerpt of Rep. IX could fill in the gap. It is not a perfect translation, however, with shifted emphasis and tone. This study’s additional goal is to demonstrate that this shift is explicable within the history of Roman Egypt, and analogous to broader changes in Christian hermeneutics and canonicity between the 2nd and 4th centuries.


Books XIII and IX of Plato’s Republic are concerned chiefly with the decline of his ideal city from injustice, mirroring the rising action of books II through IV.1 The Nag Hammadi Codex (NHC) VI excerpt begins right when (588a) Plato’s dialectical figure Socrates has just finished evaluating different forms of governance and reprises the motivating question first raised in Book II: Is it more profitable for a man to be thoroughly just, unjust, or to do unjust things while appearing just to his peers? As answer and also summary, he presents a parable of man as a soul in three parts: the many-headed beast (appetitive), the lion (spirited), and the human being (rational).

“Fashioning an image of the soul in words” (εἰκόνα πλάσαντες τῆς ψυχῆς λόγῳ)2 may not be familiar to a modern reader, nor might this moment of the ninth book seem a crucial turning-point in the dialogue. The motif carried weight in ancient metaphysics, though, especially among the Middle and Neo-Platonists. The tripartite soul, divided passions, or animal spirits appear to varying degrees in Philo, Alcinous, Plotinus, Porphyry, Proclus, Eusebius, and Paul. Each one of these thinkers warrants his own lengthy treatment, and his debt to the broader Platonic tradition would be a crucial piece. This study’s interest is rather in The Gospel of Thomas, which includes in its seventh logion an apparent reference to the lion-man Platonic motif.

The Gospel of Thomas was a non-canonical Christian tractate and, in modern times, one of the most influential texts found in Coptic, in its entirety, near Nag Hammadi, Egypt. Thomas does not read like a canonical Christian gospel. It is structured into clustered sayings (‘logia’), thus sharing a genre with wisdom-texts of the 2nd century, such as the Chaldean Oracles and the Hermetica. These works are contemporaries of Thomas, too, as the majority of scholars place its composition in Greek or Syriac between 40 and 140 AD, the latest being Porter’s claim at 250 AD.3 The Thomas sayings are declarations of the risen savior-figure Jesus, and the stakes of the tractate are established in its opening invocation: “Whoever finds the meaning of these sayings will not taste death.” There is much debate, discussed below at some length, as to why Thomas was excluded from the synoptic canon: Was this a function primarily of its content (world-denying in places), its emphasis (soteriological instead of historical), or its style (discursive instead of narrative)? The question is certainly relevant for scholars of the Early Church, interested in relating Thomas to a proposed sayings-gospel precursor ‘Q’ for Matthew and Luke. This study’s focal point, however, is in an apparent reference to the Platonic lion-man motif within the work’s seventh logion:

Jesus said: Blessed is the lion whom the man shall eat and the lion becomes man; but foul is the man whom the lion shall eat and the lion shall become man.

The stakes for understanding are high; per the incipit, each Thomas saying has within it the key to eternal salvation. If the Thomas logia are intentionally confounding or obscure, then the seventh would seem one of the most difficult. Why does the lion become man in both cases, whether eating or being eaten? The man is cursed when the lion eats him. Why, if the man eats the lion and humanizes it, does the lion still receive the blessing? For decades after the Nag Hammadi discovery, several scholars maintained that the second half of the logion was corrupted, switching the subject and object to achieve chiasmus.4 Without compelling evidence of scribal error, though, the Coptic text calls for treatment verbatim.

Logion 7 has no obvious parallels with synoptic Christian sources, unlike those sayings (nearly half of Thomas) that do overlap with dialogue in Luke or Matthew.5 Nor does familiarity with the wider Judeo-Christian milieu solve this puzzle. There is a lion-deity trope found more generally outside the canonical Christian tradition, in some Jewish and Gnostic sources. Israel and Yahweh are identified with or compared to a lion.6 Cosmogenic works, such as the Pistis Sophia (discovered in late 18th century) and Apocryphon of John (NHC II), imagine the Old Testament demiurge Yaldabaoth to have the face of a lion (leontoeides).7 Yet these negative portrayals of the lion as an Old Testament, devilish, and Earth-bound figure do not agree with the balanced (if confusing) portrayal in Thomas 7. Why would one bless the Hebrew deity or false creator-demigod for being eaten by man? Working through these possibilities, Howard M. Jackson shows convincingly that Plato Rep. IX 588a-589b is the necessary point of reference for one wishing to unravel the mystery of Thomas 7. If the lion represents the spirited passion (θυμοειδές) within a human being, and ‘the man within’ represents his reason (λογιστικὸν), one can ‘rewrite’ Thomas 7 as follows:

Jesus said: Blessed (harmonious) is the leonine passion which the inner-man reason shall eat, and the passion becomes (subset of) outer man; but foul (disharmonious) is the reason which the passion shall eat and the passion shall become (ruler over) outer man.

There is now a compelling case to make that the excerpt of Plato in NH Codex VI is related significantly to the seventh logion of Thomas in NH Codex II. This is an argument advanced by both Jackson (1985) and Lanzillotta (2013), the two most thorough investigators. Stances taken by other scholars will receive more discussion below, but their offerings are not as much alternative hypotheses as healthy skepticism. Would the 4th-century Coptic-writer have known the excerpt to be Plato? Could they have unknowingly copied down the Republic fragment, thinking it a Hermetic text or just an interesting passage? It is impossible to disprove these scenarios entirely, but the specificity of inclusion suggests an intentionality. The excerpt begins with Socrates’ question and ends with the conclusion of the relevant parable, at a significant moment.8 Also, some choices of redaction and shifts of emphasis, verbal mood, and interrogative syntax can be linked to purposive work—perhaps a ‘gnosticizing’ agenda of the translator, but almost certainly not idle commonplace copying. It is also worth remembering the expense of papyrus, the immense time and effort of scribal work. On these bases, this paper proceeds from the premise that the Republic excerpt’s inclusion was not an accident or coincidence.

This paper’s second major argument, and its departure from previous scholarship, requires firm grounding in the historical playing-field. As mentioned above, the evidence and scholarly consensus put Thomas’ original composition in Greek or Syriac between 40 and 140 AD, and certainly no later than 250. The full version of Thomas available to modern readers is in Coptic, from Nag Hammadi. The entire NH collection is dated no earlier than the middle of the 4th century. The manuscripts may have been buried after Athanasius, Archbishop of Alexandria, condemned heretical religious texts in 367,9 although scholars are careful to note that this is “mere speculation.”10 This means that at least one century, and perhaps two or more, separated (1) the original cluster-composition of Thomas and (2) the translation, revision, and collection of Coptic texts that were eventually buried near this Pachomian monastery in Upper Egypt.

The geopolitical and socioeconomic differences between 2nd-century Alexandria and 4th-century Upper Egypt are considerable. In the larger empire, E.R. Dodds saw in the Marcus Aurelius’ reign (161-180) the start of an ‘Age of Anxiety’, a Rome obsessed by peripheral wars and backward-looking classicism. Speaking broadly about the latter centuries, so-called ‘Late Antiquity’, Peter Brown saw in the overtaxed and culturally relegated colonies of Egypt, the Levant, and North Africa the foment of Christianity. During and after the reign of Augustus, Naphtali Lewis describes, Roman Egypt was under the direct command of the emperor’s personal representative. It was a codified apartheid state, where “the touchstone of status was Hellenism”, with top-down legal designations accompanying ethnic ones: Roman citizens ranking above Jews and Greek dwellers of the major cities, and they in turn above ‘Egyptians’ (all others). Caste mobility was nearly impossible, and while “the lot of the humble and the poor was not enviable anywhere in the Roman Empire… the population of Egypt appears to have been singled out for exceptionally harsh treatment.”11 Even the enfranchisement in 212 AD of all inhabitants as Roman citizens did little to change the social structure and class stratifications, which remained de facto if not de jure. The privileged position of educated, literate, and urban Greeks was enforced by law and then by norm, and the claim to the Hellenistic purisms of the Second Sophistic was as strong in Alexandria as in Athens. The separateness of Egyptian-speakers in southern country towns was enforced, even as such villages would become strong bases of support for Christian traditions, canonical and otherwise.

Having postulated that the Platonic motif contained in Book IX of the Republic is crucial to understanding Thomas 7, it bears acknowledging that the Nag Hammadi corpus is only one static window onto a dynamic entity. Why is there a Coptic translation of these religious tractates, and the Republic excerpt in particular? Why does the excerpt begin and end in such (philosophically) precise moments? There is a story in the background: Greek-language works, those of the relatively privileged, had to be rendered comprehensible for Egyptian-speakers, outside the well-educated urban sphere. The first version of Thomas likely originated in a Hellenized population of the Levant—Meyer’s best guess is Edessa, Syria12—and circulated in Greek-speaking cultural hubs like Alexandria. Some Greek fragments of Thomas were found at Oxyrhynchus, in Middle Egypt, 260 miles south. The full Coptic version comes from Nag Hammadi, another 240 miles along the Nile and, in many respects, another world entirely.

This difference in milieu between original tractate composition and the NHC version goes relatively unmentioned in the scholarship, by Jackson and especially by Lanzillotta. Jackson assumes that the Plato translator, contributing to the Nag Hammadi corpus, must very well have been from the same ‘school’ as the original Thomas writers, since there is an undeniable relationship. Lanzillotta’s aim is the application of intertextual theory, de-emphasizing historical and social background (at best speculative) in favor of the words on the page. For him, Thomas and Plato exist in a flat universe, with a linear transformation between them. His conclusion is that Thomas, in its ‘gnosticizing’ zeal, insufficiently appreciates the philosophical intricacy of the Republic. This paper’s alternative proposal is to draw the following contrast:

The original readers of The Gospel of Thomas were classically-educated speakers of Greek in Egypt (or Syriac in the Levant), fluent in Hellenistic philosophy (including the lion-man motif) and embracing a multifaceted appreciation for its seventh logion. This first group would not require a Coptic translation of a Platonic commonplace. Instead, the Coptic Republic fragment makes sense for a second audience: semi-literate readers and hearers in 4th-century Upper Egypt, de facto excluded from the wealth, learning, and Hellenism of the Roman Near East. For them, perhaps practicing semi-canonical Christianity, the codices of Nag Hammadi could be an anthologized compendium of apocryphal knowledge, delivered as oral ministry.

What is the implication of this distinction? The third thrust of the paper includes an analogy to the hermeneutical divergence between Origen and Augustine: various versus linear, multifaceted versus straightforward, textual versus rhetorical, literary versus liturgical. In the first case, if The Gospel of Thomas existed within a broad Greco-Christian philosophical frame and was read by hermeneutical circles of 2nd- and 3rd-century Alexandrians not unlike those of Philo and Origen, Plato can be allegorized into Thomas 7 with layers of literal, metaphorical, and spiritual meaning. In the second case, if Thomas existed as apocryphal scripture for a semi-heretical, quasi-Christian movement of semi-literate persons, then the Rep. IX motif existed necessarily and sufficiently as a precise allegorical tool. The goal was linear progression, from fixed words to final revelation: ‘Apply directly to logion 7.’ This second interpretive frame is analogous to (and contemporaneous with) Augustine’s method in the 4th and 5th centuries, whereby an all-encompassing and omniscient Word solves itself by moving from ‘signs’ to ‘things’. Perhaps working from the notes of his forebears several generations earlier, the translator-redactor of Coptic Plato considered the Republic excerpt a necessary reference—a semiotic ‘thing’—in the quest for Gnostic salvation. Neither Origen nor Augustine endorsed a Gnostic agenda or The Gospel of Thomas, but they offer a useful comparative basis when thinking about biblical hermeneutics writ large.

Section 2 of this paper offers an answer to the first question: What has been said convincingly in previous scholarship about the Platonic motif, the Gospel of Thomas, and the nature of their relationship? Section 3 critiques that scholarship and provides social-historical grounding in Roman Upper Egypt, working toward a deeper explanation for the Coptic translation. Section 4 extends the boundaries of significance to allegorical practice and canonicity, musing on potential connections and implications. This three-part, five-act study offers explanatory power for the ultimate mystery—why Rep. IX 588a-589b, why in Coptic, why in Nag Hammadi?—by distinguishing between the two literary-historical contexts: (1) 2nd-century Greek-readers, with a broad hermeneutical frame that relies on fluency in Greek philosophy; (2) 4th-century pseudo-Christians, reading or hearing in Coptic, with a hermeneutic that privileges verbatim truth and linear progression toward revelation.


  1. Kenneth Dorter, The Transformation of Plato’s Republic (Lanham, MD: Lexington Books, 2006), 7. 

  2. Plat. Rep. 588b. Translations of Plato are my own or in consultation of G.M.A. Grube’s Republic, rev. C.D.C. Reeve (Hackett, 1992). 

  3. J. R. Porter, The Lost Bible (New York: Metro Books, 2010), 9. Greek fragments of Thomas, found in Oxyrhynchus, Egypt, are dated to between 130 and 250 AD. 

  4. Jean Doresse’s translation in The Secret Books of the Egyptian Gnostics (1960): “Jesus says: Blessed is the lion which a man eats so that the lion becomes a man. But cursed is the man whom a lion eats so that the man becomes a lion!” Doresse agrees with the general reading, however: “No doubt the lion here represents human passions, or more precisely, the lying spirit of evil.” (p. 371). He later makes explicit appeal to “Platonic origin”, e.g. for logion 83 (p. 377). 

  5. E.g. “For many of the first will be last” (Thomas 4); “If a blind person leads a blind person, both of them will fall into a hole” (34); “Give the emperor what belongs to the emperor, give God what belongs to God, and give me what is mine” (100) 

  6. As noted by Jackson, Lion Becomes Man (1985), there are several such expressions in the Old Testament (p. 13): “He couched, he lay down as a lion, and as a great lion: who shall stir him up? Blessed is he that blesseth thee, and cursed is he that curseth thee” (Num 24:9). “Behold, the people shall rise up as a great lion, and lift up himself as a young lion: he shall not lie down until he eat of the prey, and drink the blood of the slain” (Num 23:24). “For it increaseth. Thou huntest me as a fierce lion: and again thou shewest thyself marvellous upon me” (Job 10:16). “I reckoned till morning, that, as a lion, so will he break all my bones: from day even to night wilt thou make an end of me” (Isa 38:13). “For I will be unto Ephraim as a lion, and as a young lion to the house of Judah: I, even I, will tear and go away; I will take away, and none shall rescue him” (Hos 5:14). “Therefore I will be unto them as a lion: as a leopard by the way will I observe them” (Hos 13:7). “They shall walk after the Lord: he shall roar like a lion: when he shall roar, then the children shall tremble from the west” (Hos 11:10). “The lion hath roared, who will not fear? the Lord God hath spoken, who can but prophesy?” (Amos 3:8). This image is reflected occasionally in the New Testament: “And cried with a loud voice, as when a lion roareth: and when he had cried, seven thunders uttered their voices.” (Rev 10:3). 

  7. P. Sophia 31; A. John 10. 

  8. I use the word “excerpt”, not “fragment”, for the intentionality it connotes. 

  9. James M. Robinson, “Introduction”, in The Nag Hammadi Library in English 4th ed. (Leiden: Brill, 1996), 16-19. 

  10. David Brakke, “Canon Formation and Social Conflict in Fourth-Century Egypt: Athanasius of Alexandria’s Thirty-Ninth ‘Festal Letter’”, Harvard Theological Review 87, no. 4 (1994): 518. JSTOR: 1509966. 

  11. Naphtali Lewis, Life in Egypt Under Roman Rule (Oxford: Clarendon Press, 1983), 32-5. 

  12. Marvin Meyer, “Thomas Christianity,” in The Nag Hammadi Scriptures, ed. Marvin Meyer (HarperOne 2007), 780.