Meddling with the Gospel

Celsus, Early Christian Textuality, and the Politics of Reading

In: Novum Testamentum
Jeremiah Coogan Research fellow, University of Oxford Oxford UK
Jesuit School of Theology, Santa Clara University Berkeley, California USA

Search for other papers by Jeremiah Coogan in
Current site
Google Scholar
Open Access


The second-century philosopher Celsus disparaged Christians who “alter the original text of the Gospel three or four or many times” (Cels. 2.27). Scholars have understood this passage as a critique of multiple distinct Gospels, but Celsus’ invective is better explained by comparison with elite second-century polemics (e.g., Gellius, Lucian, Galen) against readers who lack discernment and arbitrarily alter manuscripts. For Celsus, Christians’ irresponsible textual practices reveal their cultural inferiority. The complaint is about varying copies of what Celsus thinks to be the same work: “the Gospel.” Christian thinkers in the second and third centuries—including Irenaeus, Tertullian, Origen, and the author(s) of the Little Labyrinth—also participate in this discourse about good and bad readers. This article thus illuminates the wider ancient Mediterranean politics of reading in which early Christian textuality emerged.

1 Introduction*

A first-century CE fresco from the southern Italian city of Pompeii depicts several common textual technologies (see fig. 1).1 A bookroll and two sets of wax tablets in the lower register of the fresco are juxtaposed with generous heaps of coins in the upper register. Assorted writing implements accompany the inscribed media. The fresco comes from a modest estate with extensive inscriptions that offer insight into the social standing of this household, owned by a well-to-do woman, Julia Felix, and including enslaved workers, dependents, and renters who leased retail space. The intermingling of money and writing signals Julia’s control of her business ventures and her ability to deploy the technologies of literacy.2 The depiction of a bookroll in addition to the tablets communicates educational attainment beyond the literacy and numeracy of bookkeeping. The fresco uses writing as a sign of professional competence, economic success, and cultural attainment.3

Figure 1
Figure 1

Fresco from the praedia of Julia Felix (ii.4.10), Pompeii, Museo Archeologico Nazionale di Napoli (mann), inv. nr. 8598d

Citation: Novum Testamentum 65, 3 (2023) ; 10.1163/15685365-bja10044

by permission of the minis tero DELLA cultura—museo archeologico nazionale di napoli

Recent scholarship has often located early Christian textuality within the social contexts evoked by this fresco. In the first three centuries CE, few Christian books and readers belonged to the cultural stratosphere populated by figures like Seneca, Plutarch, Suetonius, or Galen. Most Christians participated in a wider landscape of practical literacy, the mode of reading and writing advertised by Julia Felix’s fresco.4 Yet such literacy was contested. Roman elites polemicized against what they saw as the deficient education and incompetent textual practices of any but the most privileged readers. As William A. Johnson has argued, textual practices afforded an opportunity to display wealth, status, and virtue in the fiercely competitive social scene of the High Roman Empire.5 Investigating this contested space enriches our understanding of early Christian books and their readers.

Early Christian thinkers participated in these polemics about books and readers that characterized the vibrant intellectual milieu of the second and third centuries CE. Christian intellectuals like Justin of Rome, Tatian the Assyrian, Irenaeus of Lyons, Tertullian of Carthage, and Origen of Alexandria all engaged in debates about Gospel books and Gospel reading. In this article, however, I begin with one of Christianity’s cultured despisers, the second-century philosopher Celsus.6 By reading Celsus’ polemics alongside both early Christian intellectuals and elite Roman thinkers like Gellius, Lucian, and Galen, I locate his arguments about Gospel literature within the landscape of intellectual activity under the Antonine and Severan emperors.7 This article thus illuminates the social and intellectual contexts in which early Christian textuality emerged.

Celsus is the first non-Christian to demonstrate knowledge of Gospel books.8 This makes him a valuable historical source, even though our access to the second-century Celsus and his treatise True Logos (Ἀληθὴς Λόγος) is mediated through the third-century Christian scholar Origen, writing in the 240s CE.9 At the urging of his patron Ambrose, Origen composed the Contra Celsum, a lengthy refutation of Celsus’ treatise against Christianity. As Origen reports, Celsus objected to the textual instability of Gospel books. Among his many critiques of the nascent Christian movement, he mocked Christians who “alter the original text of the Gospel three or four or many times, and modify it in order to be able to reject criticisms” (μεταχαράττειν ἐκ τῆς πρώτης γραφῆς τὸ εὐαγγέλιον τριχῇ καὶ τετραχῇ καὶ πολλαχῇ καὶ μεταπλάττειν, ἵν’ ἔχοιεν πρὸς τοὺς ἐλέγχους ἀρνεῖσθαι, Cels. 2.27).10

Scholars have often understood Celsus’ statement as referring to multiple, distinct Gospels.11 If this were true, it would offer early evidence for a pluriform Gospel and would attest “pagan” critiques of the dissonance that such a collection of Gospel books occasioned. Tjitze Baarda and David Parker have influentially stated that early Christian thinking about a unified Gospel was an apologetic response to such critiques.12 Several second-century Christian projects might thus be read as responses to Celsus’ critique: The single Gospel of Marcion of Pontus, adapted from an earlier text resembling Luke; the painstaking synthesis of Tatian the Assyrian, who produced a single Gospel by interweaving material from multiple existing written narratives; and the innovative theory of Irenaeus of Lyons, who imagined a quadriform Gospel corresponding to the divinely ordained structure of the cosmos.13 Each, in its own way, articulates a unified Gospel corpus. Because of this reading of Celsus’ critique, the passage plays a major role in scholarly accounts of early Gospel reception and of the emergence of a fourfold Gospel.

This reading of Celsus’ statement invites two major challenges. First, the existence and use of multiple Gospel books did not offer a compelling objection to Christian book practices in Celsus’ Roman Mediterranean context. To have multiple books on identical or related topics was no reproach in the bibliophilic milieu of the second century CE. Intellectual culture under the Antonine and Severan emperors was characterized by what Karen ní Mheallaigh has called a “documentary” impulse, a focus on written records, books, libraries, and archives.14 As a manifestation of this dynamic, we find readers collecting parallel plays, parallel biographies, parallel histories, and parallel technical treatises.15 Owning numerous books on the same subject was a display of erudition and a source of pride. We thus need a better account of what Celsus found blameworthy about multiple Gospel books. Second, Celsus does not criticize contradictions between different works and offers only limited indication of knowing multiple different Gospels. Origen’s Contra Celsum preserves plentiful evidence that Celsus used the Gospel according to Matthew, but scant indication that Celsus knew other Gospels. For both reasons, I propose a different reading of Celsus’ critique.

Celsus’ critique of Gospel books participates in the politics of reading of the second- and third-century Roman Mediterranean, reflecting elite discourses about reading, education, and cultural capital. Celsus does not critique Gospel plurality or contradiction; rather, he polemicizes against the textual instability of a single work that he calls “Gospel” (εὐαγγέλιον). He blames this instability on uneducated and incompetent Christian readers.16 Like other second-century elites, Celsus criticizes the inadequate παιδεία of subaltern readers who are said to lack appropriate discernment and, therefore, to change texts arbitrarily.17

In what follows, I trace this widespread elite critique of bad readers and textual meddling by focusing on several examples from Celsus’ contemporary, the physician Galen of Pergamum (ca. 129–216 CE). Then, I re-read Celsus’ critique of Gospel difference within the second-century context of his polemical project. In this way, I re-evaluate its significance for understanding early Christian literature. Finally, I show how Christian thinkers used similar polemics for their own rhetorical ends. Celsus’ critique reflects hierarchies that were embraced by Christian thinkers as well. This reconsideration of Celsus’ critique illuminates the politics of reading that shaped how intellectuals—Christian and otherwise—engaged textual similarity and difference in the Roman Mediterranean.

2 Bad Reading and Textual Instability

I begin with bad readers.18 To critique Christians, Celsus draws upon an expansive repertoire of invective that contrasts the πεπαιδευμένος, the educated man, with the imposter who pretends at παιδεία.19 The πεπαιδευμένος possesses the discernment which the poseur lacks.20

An example of this polemic appears in the opening anecdote from the treatise On My Own Books by the physician Galen of Pergamum.21 Galen describes a dispute at a bookseller’s shop in Rome. A gullible man had purchased a bookroll labelled “The Doctor by Galen” (Γαληνοῦ ἰατρός). Yet the attribution is spurious. The title is, we are led to assume, a dishonest bookseller’s clever attempt to pass off an inferior medical work by connecting it to the renowned Galen. But a discerning reader can tell the difference: A “man of letters” (τις ἀνὴρ τῶν φιλολόγων) enters the scene. The educated man recognizes the deception after reading only the first two lines of the bookroll. He pronounces that the work does not match Galen’s style (λέξις) and must be falsely attributed (ψευδῶς ἐπιγέγραπται); he corrects the misattribution by ripping up the fraudulent label. As Galen informs us, this man’s superb παιδεία means that he can identify the fabrication. The man has been formed from childhood (ἐπεπαίδευτο τὴν πρώτην παιδείαν) in the appropriate literary disciplines (ἐπαιδεύοντο παρά τε γραμματικοῖς καὶ ῥήτορσιν). He has the right breeding and education. A less-elite reader, though, is at the mercy of an opportunistic bookseller. Only a properly educated man—like Galen himself—can recognize what he reads, interpret it well, and intervene with understanding.22 Or this is what Galen would have us believe.

Elite reading is characterized by the ability to discern good and bad, correct and incorrect. But many readers lack the appropriate παιδεία. Moreover, some incompetent readers have the temerity to intervene in their texts anyway. This is a double mark of inadequacy. Not only are these readers incompetent, but they fail even to recognize this fact: They don’t know their place.23 As Cat Lambert demonstrates, these polemics appear in a widespread topos whereby elite figures denigrate subaltern reading as “bookwormish,” constituted by “nibbling at” texts without adequate intellectual and cultural formation.24 According to this ancient exclusionary discourse, subaltern readers lack the discernment cultivated by appropriate education and thus damage the texts that they encounter.25

Galen and his ilk do not regard all textual intervention as bad. In a world of manuscripts, the first stage of scholarly reading was to establish a correct text (διορθοῦν, ἐπανορθοῦν, κατορθοῦν).26 In theory, a competent reader could discern and correct errors. By contrast, the arriviste lacked the right παιδεία. They might attempt corrections, but they inevitably fail. Like many of his elite contemporaries, Galen contrasts his own learned editorial intervention with the meddling of bad readers. I illustrate this with two examples from Galen’s expansive commentary on the medical treatise known as the Epidemics, a work attributed to the famous physician Hippocrates of Cos (fifth century BCE).

First, in his commentary on Book 3 of the Epidemics, Galen discusses the problems posed by unusual symbols (χαρακτῆρες) at the end of the text in some manuscripts. Galen narrates events that he situates in the second century BCE, but the purpose of this account is to position Galen’s own textual scholarship. His comments reflect the practices and prejudices of his second-century CE context. As Galen writes,27

ὡς γὰρ μηδ’ ἑρμηνεῦσαι δυνηθέντος αὐτοῦ καλῶς τὰ διεσκευασμένα πρὸς τοῦ Μνήμονος, ἀλλ’ ἐν οἷς ἠπόρει πιθανολογίας ὑπαλλάττοντος τοὺς χαρακτῆρας εἰς εὐπορίαν ἐξηγήσεως, οὕτω πεποίηται τὴν ἀντιλογίαν, [οὐδὲν βιβλίον] οὔτε τὸ κατὰ τὴν βασιλικὴν βιβλιοθήκην εὑρεθὲν οὔτε τὸ ἐκ τῶν πλοίων οὔτε τὸ κατὰ τὴν ὑπὸ Βακχείου γενομένην ἔκδοσιν ἔχειν φάσκων οὕτω τοὺς χαρακτῆρας, ὡς ὁ Ζήνων ἔγραψεν ἐπὶ τοῦ προκειμένου κατὰ τὸν λόγον μειρακίου.

Because Zeno was unable to interpret Mnemon’s revisions well but was at a loss for even a probable meaning, he changed the symbols to fit his exegesis, and thus [Apollonius Byblas] made the refutation that [no bookroll], neither that discovered in the royal library nor that from the ships nor that from the edition of Baccheius, has the symbols in the way that Zeno wrote in the account of the young man under discussion.

In Galen’s account, the Alexandrian physician Zeno (second century BCE) is accused of having changed (ὑπαλλάττοντος) the perplexing Hippocratic symbols to fit his own interpretation. If we could ask him, Zeno might have described his interventions as reasonable conjectural emendation. Yet, according to Galen, Zeno is blameworthy because he fails to recognize that the symbols are post-Hippocratic; he thus exhibits deficient critical judgment. Moreover, Zeno deserves ethical reproach because he excuses himself from the hard work of interpretation by modifying these perplexing symbols to fit his own understanding. Zeno’s actions contrast with Galen’s own approach, characterized by careful collation of (what Galen thinks to be) the oldest and best manuscripts. Note, moreover, similarities between what Galen reports and what Celsus says about Christians: To fix interpretive problems with a text, a reader irresponsibly modifies the content—and is subject to reproach as a result.

Second, at the beginning of his commentary on Book 6 of the Epidemics, Galen discusses his editorial response to the misguided interventions of previous readers.28 Unnamed readers had irresponsibly changed the text of Hippocrates (τὴν […] λέξιν ὑπαλλάττων). As in the previous example, the changes support the meddlers’ preferred interpretations (ὡς ἕκαστος ἤλπισε πιθανῶς ἐξηγήσασθαι). Because these readers had damaged (ἐλυμήναντο) the text, Galen must seek out ancient copies (ἀντίγραφα) and earlier commentaries (ὑπομνήματα) of Hippocrates to repair the injury. Emendation (ἐπανόρθωσις) is not rejected; it is exactly what Galen is doing. But only the rare reader (as Galen would have it) is qualified to undertake this task of textual repair. Galen characterizes his editorial process as one marked by restraint, while other, unqualified readers irresponsibly modify the text for their own ends.29

As these examples illustrate, complaints about failed textual intervention were part of the rhetorical boundary-marking that distinguished the idealized elite reader from those who have no business reading at all. On this model, bad readers meddle with texts and change them to fit their own interpretations. Good readers, by contrast, discern how to restore texts appropriately. This critique of readers who intervene without discernment is widespread.30

Yet this discourse of elite literary competence is a fiction. As scholars such as Nicholas Horsfall, Sarah Blake, Joseph Howley, and Candida Moss have demonstrated, Roman elites sometimes did their own reading and writing, but frequently relied upon non-elite (often enslaved) workers to read, write, and edit texts.31 Many intellectual tasks were performed by subaltern workers who are deliberately obscured by ancient sources. The rhetoric of elite reading thus distorts the complex landscape of textual labor and expertise in the Roman Mediterranean. This construction of elite παιδεία as the sine qua non of authoring, editing, and reading is what underwrites the authority of elite figures as the arbiters of intellectual activity, literary production, and cultural authority.

3 Celsus’ Critique of Gospel Difference

This Roman politics of reading illuminates Celsus’ critique of Gospel difference. Here we must avoid conflating Origen and Celsus in two respects. First, Origen makes little effort to distinguish what Celsus knows about Jesus from what Origen thinks Celsus should have known about Jesus. Origen repeatedly complains that Celsus should have discussed material from Luke or John. Yet this does not mean that Celsus had access to these texts, only that Origen thinks that he should have. Second, Origen is writing eighty years after Celsus and has his own priorities. For both reasons, we must be careful not to retroject Origen’s fourfold Gospel—or Origen’s approach to Gospel books— uncritically back onto Celsus.

As a survey of the Contra Celsum reveals, Origen attests only Celsus’ use of material found in the Gospel according to Matthew.32 Much of what Celsus knows about Jesus appears in multiple Gospels, both canonical and noncanonical. (Given the similarities between many early Gospel texts, this is not surprising.) Yet Celsus discusses several details that reflect distinctive Matthean material, from Jesus’ family’s sojourn in Egypt (Cels. 1.38) to the earthquake at Jesus’ crucifixion (Cels. 2.59). No distinctive material indicates Celsus’ knowledge of any other written Gospel. Nothing demonstrates that Celsus engages a text that looks like Mark, Luke, or John. Nor is there indication that Celsus is reading Marcion’s Gospel, the Gospel according to the Hebrews, the Gospel of Thomas, or any other extant second-century text about Jesus. Absence of evidence is not evidence of absence; Celsus might have read other Gospel books. Nonetheless, Celsus’ reliance on Matthew matters for how we understand his critique of unstable Gospel books.

When discussing Celsus’ engagement with Gospel difference, scholars rely upon two passages, Contra Celsum 2.27 and 5.52. These are the only extant passages where Celsus mentions variance in Gospel texts. Both have been read as evidence for Celsus’ objection to a pluriform Gospel, but they are open to a different reading.

I begin with Contra Celsum 5.52, which scholars often cite to demonstrate Celsus’ critique of differences between Gospels:

Ἣν δὲ βουλόμεθα ἐξετάσαι νῦν τοῦ Κέλσου λέξιν οὕτως ἔχει· […] Καὶ μὴν καὶ πρὸς τὸν αὐτοῦ τοῦδε τάφον ἐλθεῖν ἄγγελον, οἱ μὲν ἕνα, οἱ δὲ δύο, τοὺς ἀποκρινομένους ταῖς γυναιξὶν ὅτι ἀνέστη. Ὁ γὰρ τοῦ θεοῦ παῖς, ὡς ἔοικεν, οὐκ ἐδύνατο ἀνοῖξαι τὸν τάφον, ἀλλ’ ἐδεήθη ἄλλου ἀποκινήσοντος τὴν πέτραν. […]33

The statement of Celsus which we now want to examine reads as follows: “[…] Furthermore, they say that an angel came to the tomb of this very man (some say one angel, some two), who replied to the women that he was risen. The child of God, it seems, was not able to open the tomb, but needed someone else to move the stone. […]”

The question is the number of angels at Jesus’ tomb. Celsus observes that, according to Christians, “an angel came to the tomb of this very man (some say one angel, some two), who replied to the women that he was risen.” This comment may reflect Celsus’ knowledge of the difference between the single angel of Matthew and Mark and the two angels of Luke and John.34 If so, it would be the only passage in which Celsus discusses Gospel contradiction. Yet Celsus does not critique Gospel difference. Instead, he lists various angelic figures—from Gospel texts, from Enochic literature, from other Jewish scriptures—in order to assert that, even if Jesus was an angelic figure, he was not unique in that regard. Celsus does not describe this as a contradiction or even as a textual difference between Gospel books. That is simply not the point.

In this light, Origen’s response in Cels. 5.56 is striking:

Εἶτα ἑξῆς τὰ ἄμικτα καὶ ἀνόμοια μιγνὺς καὶ ἐξομοιῶν ἀλλήλοις ἐπιφέρει τῷ περὶ τῶν, ὥς φησι, καταβεβηκότων ἑξήκοντα ἢ ἑβδομήκοντα ἀγγέλων λόγῳ, πηγὰς θερμῶν κατὰ αὐτὸν δακρυσάντων, ὅτι καὶ πρὸς αὐτοῦ τοῦ Ἰησοῦ τὸν τάφον ἱστόρηνται ἐληλυθέναι ὑπό τινων μὲν ἄγγελοι δύο ὑπό τινων δὲ εἷς· οὐκ οἶμαι τηρήσας Ματθαῖον μὲν καὶ Μάρκον ἕνα ἱστορηκέναι, Λουκᾶν δὲ καὶ Ἰωάννην δύο, ἅπερ οὐκ ἦν ἐναντία. Οἱ μὲν γὰρ ἀναγράψαντες ἕνα τὸν ἀποκυλίσαντα “τὸν λίθον” ἀπὸ τοῦ μνημείου τοῦτόν φασιν εἶναι, οἱ δὲ τοὺς δύο τοὺς ἐπιστάντας “ἐν ἐσθῆτι ἀστραπτούσῃ” ταῖς γενομέναις ἐπὶ τὸ μνημεῖον γυναιξὶν ἢ τοὺς θεωρηθέντας ἔνδον “ἐν λευκοῖς καθεζομένους.” Ἕκαστον δὲ τούτων νῦν παραδεικνύναι δυνατὸν καὶ γεγενημένον καὶ δηλωτικόν τινος εἶναι τροπολογίας, τῆς περὶ τῶν προφαινομένων τοῖς τὴν ἀνάστασιν τοῦ λόγου θεωρεῖν παρεσκευασμένοις, οὐ τῆς παρούσης ἐστὶ πραγματείας ἀλλὰ μᾶλλον τῶν τοῦ εὐαγγελίου ἐξηγητικῶν.35

Then [Celsus] next combines incompatible and dissimilar statements and compares them to one another. For after his words about the sixty or seventy angels who, as he says, came down, and whose tears in his opinion are hot springs, he goes on to say that according to some two angels are related to have come to the tomb of Jesus himself, while others say only one. I do not think he noticed that Matthew and Mark narrate one, while Luke and John narrate two angels. But these statements are not contradictory. For the writers that have one angel say that this one was he who rolled back the stone from the tomb, while those that have two say they stood in shining raiment before the women who came to the tomb, or that they were seen “sitting in white robes” within it. However, while it would be possible to substantiate now each of these statements, both as historical events and as manifesting some allegorical meaning which concerns the truths which are made clear to people who have been prepared to see the resurrection of the Logos, it is not relevant to the present treatise but rather to commentaries on the Gospel.

Origen knows that Matthew and Mark describe one angel, while Luke and John describe two. But Origen does not think that Celsus is objecting to contradiction between Gospels. On the contrary, Origen asserts that this is not what Celsus was investigating. For this reason, Origen simply points the reader to “commentaries on the Gospel” (presumably his own Commentary on Matthew) for an explanation of the differences.36 Then he continues responding to Celsus’ argument about the relationship between Jesus and angels. This leaves several possibilities: One is that Celsus knows multiple distinct Gospels. Another is that he is aware of differences in how Christians describe Jesus’ resurrection but does not derive this knowledge from comparing Gospel texts. Yet another is that Celsus has observed textual variation among manuscripts of a single Gospel (presumably Matthew). Any of these options is possible, but—more importantly—neither the material cited from Celsus nor Origen’s response indicates that Celsus critiques a contradictory corpus of Gospel books.

This leads us back to Contra Celsum 2.27:

Μετὰ ταῦτά τινας τῶν πιστευόντων φησὶν ὡς ἐκ μέθης ἥκοντας εἰς τὸ ἐφεστάναι αὑτοῖς μεταχαράττειν ἐκ τῆς πρώτης γραφῆς τὸ εὐαγγέλιον τριχῇ καὶ τετραχῇ καὶ πολλαχῇ καὶ μεταπλάττειν, ἵν’ ἔχοιεν πρὸς τοὺς ἐλέγχους ἀρνεῖσθαι. Μεταχαράξαντας δὲ τὸ εὐαγγέλιον ἄλλους οὐκ οἶδα ἢ τοὺς ἀπὸ Μαρκίωνος καὶ τοὺς ἀπὸ Οὐαλεντίνου οἶμαι δὲ καὶ τοὺς ἀπὸ Λουκάνου. Τοῦτο δὲ λεγόμενον οὐ τοῦ λόγου ἐστὶν ἔγκλημα ἀλλὰ τῶν τολμησάντων ῥᾳδιουργῆσαι τὰ εὐαγγέλια. Καὶ ὥσπερ οὐ φιλοσοφίας ἔγκλημά εἰσιν οἱ σοφισταὶ ἢ οἱ Ἐπικούρειοι ἢ οἱ Περιπατητικοὶ ἢ οἵτινές ποτ’ ἂν ὦσιν οἱ ψευδοδοξοῦντες, οὕτως οὐ τοῦ ἀληθινοῦ χριστιανισμοῦ ἔγκλημα οἱ μεταχαράττοντες τὰ εὐαγγέλια καὶ αἱρέσεις ξένας ἐπεισάγοντες τῷ βουλήματι τῆς Ἰησοῦ διδασκαλίας.37

After this [Celsus] says that some believers, as though drunk, go so far as to attack themselves and alter the original text of the Gospel three or four or many times, and modify it in order to be able to reject criticisms. But I do not know of people who have altered the Gospel apart from the followers of Marcion and Valentinus, and I think also the followers of Lucan. But this statement is not a criticism of the logos, but only of those who have dared to tamper with the Gospels. And just as it is no criticism of philosophy that there are the Sophists or the Epicureans or the Peripatetics, or any others who hold false doctrines, so those who alter the Gospels and introduce heresies foreign to the meaning of Jesus’ teaching do not give ground for any criticism of true Christianity.

Celsus describes textual chaos analogous to the results of an intoxicated brawl (ὡς ἐκ μέθης ἥκοντας εἰς τὸ ἐφεστάναι αὑτοῖς). Christian readers have interfered arbitrarily, even drunkenly, in their Gospel texts. Just prior to this passage, Celsus asserts that the Christian Gospel is full of implausible and embarrassing fictions (πλάσματα, Cels. 2.26). Now he suggests that readers have meddled with the texts in a futile attempt to avoid well-deserved critique.

Celsus describes Christians’ alteration or modification (μεταχαράττειν, μεταπλάττειν) of a single text (τὸ εὐαγγέλιον) from an original written form (ἐκ τῆς πρώτης γραφῆς) into “three or four or many” (τριχῇ καὶ τετραχῇ καὶ πολλαχῇ) varying versions. But Celsus does not describe these versions as different works or distinct Gospels. Rather, a single Gospel (εὐαγγέλιον) has been shattered into disagreeing textual forms.38 We might compare the figures described by Galen, readers of medical texts who purportedly meddle with books so they can offer interpretations that they find more satisfying. Celsus’ objection is not that Christians have contradictory Gospels, but that they cannot keep straight the text of a single work known as “the Gospel.”

Unfortunately, Celsus does not offer examples. What kinds of disagreement does he have in mind? This odd lack of detail warrants our attention. We might conclude that Origen does not cite Celsus’ examples of textual disagreement. Yet, both in this treatise and throughout his corpus, Origen avidly discusses Gospel difference. Why would he forego an opportunity here? Origen even objects elsewhere in the Contra Celsum that Celsus must have been an inattentive reader because Celsus fails to bring up such difficulties as the divergent genealogies.39 As Origen urges, “Let [Celsus] read the Gospels”—implying that Celsus has failed to read carefully or at all (Cels. 2.36). Origen portrays the critic Celsus as the one who lacks discernment, attention, and philosophical acumen.40 It is Celsus who is a bad reader. Origen even introduces (and, from his perspective, resolves) the problem of conflicting genealogies in the Gospels according to Matthew and Luke. It is unlikely, then, that Origen would avoid a textual problem if Celsus had offered one. It is equally unlikely, if comparing different Gospels for contradiction was Celsus’ aim, that he would neglect such a striking example as the genealogies.

Origen’s response is telling. He is aware of divergent Gospel texts, and differences within the emerging fourfold Gospel are a constant object of scrutiny throughout his career.41 But here he does not defend a fourfold Gospel or resolve contradictions between texts. Instead, he offers a “false flag” defense. As Origen writes, only those associated with Marcion, Valentinus, and Lucan “have altered the Gospel” (Cels. 2.27). Real Christians, Origen maintains, do not engage in the blameworthy textual meddling that Celsus describes. No, only “heretics” do anything of the sort. They are the ones who dare to tamper with (ῥᾳδιουργεῖν) Gospel narratives.42 This is a convenient way to avoid any implication that Christians engage in the disreputable behavior that Celsus describes as drunken meddling. Origen’s response suggests that he reads Celsus’ objection as I do, and that he turns the tables in two ways: first, by critiquing Celsus as the incompetent and ill-formed reader; and second, by redirecting Celsus’ invective at its appropriate target: the textual practices of heretics.43

4 Christian Accusations of Textual Meddling

As I have argued, Celsus deploys a widespread polemical discourse in his in‑ vective against Christian meddling with the Gospels (Cels. 2.27). Yet this was a game that two could play. Second- and third-century Christian thinkers deploy this discourse for their own ends. As we have seen, Origen turns this polemic against the “heretical” factions of Marcion, Valentinus, and Lucan, who are said to meddle recklessly with Gospel texts.44

Other Christian thinkers in the Antonine and Severan periods use similar invective.45 Consider the complaints that Irenaeus and Tertullian bring against Marcion. As Tertullian asks, “What Pontic mouse is more corrosive than the man who has gnawed away at the Gospels?” (Marc. 1.1.5).46 Repeated accusations focus on Marcion’s incompetent whittling away of existing texts. Tertullian speaks of Marcion as wielding the craftsman’s knife rather than the authorial pen (Praescr. 38.9). According to Irenaeus, Marcion’s labors produce a falsified text, a “fragment” of the whole that the Gospel should be (Haer. 1.27.2). Marcion’s reckless meddling is tied to questions of ethnicity and cultural formation. Tertullian informs us that, although Marcion is wealthy, he remains a mere merchant, lacking proper education and coming from barbarian Pontus in the hinterlands of the empire.47 As a result, Marcion’s attempts at authoring or editing texts end, inevitably, in failure.48

Heresiological descriptions of the late second-century cobbler (σκυτεύς) Theodotus of Byzantium depict the textual practices of Theodotus’ circle by using the same language for textual meddling and irresponsible intervention that we have seen used by Galen, Celsus, and Origen. According to a third-century treatise known as the Little Labyrinth, Theodotus and his followers had scholarly aspirations despite their modest social standing and their lack of appropriate education.49 They engage in incompetent attempts at the textual revision of Christian scriptures. The Theodotians have “tampered recklessly with the divine writings” (γραφὰς […] θείας ἀφόβως ῥερᾳδιουργήκασιν, 5.28.13) and they “applied their hands recklessly to the divine writings, claiming to correct them” (ταῖς θείαις γραφαῖς ἀφόβως ἐπέβαλον τὰς χεῖρας, λέγοντες αὐτὰς διωρθωκέναι, 5.28.15).50 Equipped with artisanal skills (ταῖς τῶν ἀπίστων τέχναις, 5.28.15) rather than with knowledge, the Theodotians are said to copy out these defective copies with their own hands, making changes without good reason (5.28.18). By contrast, the author of the polemical treatise deploys critical judgment—συγκρῖναι, 5.28.17—something that Theodotus’ circle lacks. Theodotus and his circle have the manual alphabetic skills of bookwork, but not the παιδεία required for appropriate intervention.51 They meddle repeatedly, producing incompatible versions. Because they lack discernment, Theodotus and his circle “recklessly” (ἀφόβως, 5.28.13, 15) attempt to “correct” (διωρθωκέναι, 5.28.15; κατωρθωμένα, 5.28.17) these authoritative texts but instead “obscure” (ἠφανισμένα, 5.28.17) and “debase” (παραχαράσσειν, 5.28.18) them.52

We thus find Christian thinkers—including Irenaeus, Tertullian, Origen, and the author(s) of the Little Labyrinth—deploying the same set of polemical strategies as Celsus to delegitimize texts and readers whom they find unacceptable. In disputes, these second- and third-century intellectuals draw on the authorizing power of education and on prejudices grounded in social status. Like Celsus, these Christian thinkers use the rhetoric of παιδεία to position themselves as authoritative arbiters of text and truth.

5 Conclusion

Celsus’ polemic about Gospel books was about class and about the rhetorical construction of “proper” educated reading, not about contradiction. He objects to (what he sees as) unstable manuscript copies and not to collections of distinct and contradictory Gospels. The re-reading that I offer in this article destabilizes the commonplace assertion that early Christians engaged in theological reflection about the multiplicity of Gospel books only in response to external critiques. This, in turn, invites us to rethink why early Christians began to theorize a pluriform Gospel and how they grappled with the complexities of their Gospel books. Rejecting the familiar but flawed narrative about Celsus enables us to observe the ancient Mediterranean politics of reading in which early Christian textuality emerged. Attending to this politics of reading illuminates both Celsus’ critique and a wider discourse about good and bad readers.

For Celsus, Christians’ irresponsible textual practices demonstrate their cultural inferiority and their inadequate παιδεία. This critique reflects elite disdain for “uneducated” and low-status readers (perhaps readers like Julia Felix), who ostensibly lack discernment and arbitrarily alter manuscripts.53 Celsus identifies Christians as “bad readers.”54 As women, barbarian immigrants, artisans, and enslaved workers, they cannot read properly. This is, again, a fiction; it ignores the fact that lower-status workers were expected to “meddle,” edit, and produce texts regularly. Celsus’ critique reflects what feminist philosopher Sara Ahmed has called a “paperless philosophy,” a view of the world that works only in abstraction from the pragmatic, embodied, and frequently exploitative nature of textuality and knowledge.55

Celsus’ invective about the textual instability of Gospel books advances his broader polemic against Christians. As he would have it, their pretensions to philosophy are belied by their irrational and immoral habits and beliefs. Christians are uneducated, οἱ πολλοί, lacking the intellectual capacity or pedagogical formation to read or think well. Christians make ridiculous claims about their philosophical standing but cannot muster the merest shred of education or culture. The textual instability of Gospel books contributes to Celsus’ larger project of denigrating Christians as low-status rabble. In this way, Celsus’ complaint about Gospel books fits a recognizable second- and third-century pattern—one in which early Christian thinkers also participate—of denigrating the authorial and textual activities of those deemed to lack the appropriate social status for “real” παιδεία. His critique of Christians and their books reflects an ancient politics of reading: It’s about gender, class, παιδεία, and power.


This project received funding from the European Union’s Horizon 2020 research and innovation program under Marie Skłodowska-Curie grant agreement no. 891569, “Expanding the Gospel according to Matthew: Continuity and Change in Early Gospel Literature.”


I am also grateful for the support of the Oxford Centre for Hebrew and Jewish Studies and Keble College. Earlier forms of this material were presented at the British New Testament Conference in August 2021, the Jesuit School of Theology in December 2021, the LMU München in December 2021, and the University of Nottingham in May 2022. I am grateful to these audiences for their generous engagement. Many other interlocutors have sharpened my thinking and shared their expertise. I express my gratitude to Markus Bockmuehl, Paul Foster, Tom Geue, Darrell Hannah, Jan Heilmann, Joseph Howley, Samuel B. Johnson, Julia Lindenlaub, Jacob Rodriguez, Kelsie Rodenbiker, Isaac Soon, Francis Watson, and especially Candida Moss for their insights, as well as to a peer reviewer for their incisive comments.

The fresco is from the praedia of Julia Felix (II.4.10) in Pompeii and is preserved in the Museo Archeologico Nazionale di Napoli (MANN) as inv. nr. 8598d. It was buried by the eruption of Vesuvius, which destroyed the city in 79 CE.


On women’s literacy in the Roman Mediterranean and in early Christianity, see K. Haines- Eitzen, “‘Girls Trained in Beautiful Writing’: Female Scribes in Roman Antiquity and Early Christianity,” Journal of Early Christian Studies 6 (1998) 629–646; E.A. Hemelrijk, Matrona Docta: Educated Women in the Roman Elite from Cornelia to Julia Domna (London: Routledge, 1999); K. Haines-Eitzen, The Gendered Palimpsest: Women, Writing, and Representation in Early Christianity (Oxford: Oxford University Press, 2012) esp. 23–38.


On literacy and book culture as status markers among aspirational elites in the Roman Mediterranean, see D.S. Wallace-Hadrill, Rome’s Cultural Revolution (Cambridge: Cambridge University Press, 2008) 315–355. On the circulation of books among artisans, merchants, and others of modest social and economic standing in the Roman Mediterranean, see J. Heilmann, Lesen in Antike und frühem Christentum: Kulturgeschichtliche, philologische sowie kognitionswissenschaftliche Perspektiven und deren Bedeutung für die neutestamentliche Exegese (Tübingen: Francke, 2021) 279–289.


I nuance R.F. Walsh, The Origins of Early Christian Literature: Contextualizing the New Testament within Greco-Roman Literary Culture (Cambridge: Cambridge University Press, 2021), which describes early Christian Gospel literature as part of “elite” cultural production. By locating these early Christian books and readers in non-elite strata of Roman society, my objective is not to reinforce a division between Klein- and Hochliteratur, categories that often concede the polemical constructions of ancient elites; rather, I demonstrate how such ancient polemics dismissed and marginalized early Christian textuality. Cf. J. Coogan, “Gospel as Recipe Book: Nonlinear Reading and Practical Texts in Late Antiquity,” Early Christianity 12 (2021) 40–60.


W.A. Johnson, Readers and Reading Culture in the High Roman Empire: A Study of Elite Communities (Oxford: Oxford University Press, 2010). Johnson focuses on figures from between the end of the first century CE and the end of the second century CE—the precise range for Celsus, although earlier than Origen’s response. On shifts in Roman intellectual culture across the Antonine and Severan periods, see J. Secord, Christian Intellectuals and the Roman Empire: From Justin Martyr to Origen (College Station: Pennsylvania State University Press, 2020) and A. König, R. Langlands, and J. Uden, eds., Literature and Culture in the Roman Empire, 96–235: Cross-Cultural Interactions (Cambridge: Cambridge University Press, 2020). In scholarship on early Christianity, see further M.D.C. Larsen, Gospels Before the Book (New York: Oxford University Press, 2018); J.M.F. Heath, Clement of Alexandria and the Shaping of Christian Literary Practice: Miscellany and the Transformation of Greco-Roman Writing (Cambridge: Cambridge University Press, 2020); C. Keith, The Gospel as Manuscript: An Early History of the Jesus Tradition as Material Artifact (Oxford: Oxford University Press, 2020).


Celsus’ credentials as a Platonist remain debated, but his aspirations to philosophical prestige are clear. On whether one can regard Celsus as a Platonist, see the essays of Sedley, Reydams-Schils, Goldhill, and Boys-Stones in J.N. Carleton Paget and S. Gathercole, eds., Celsus in His World: Philosophy, Polemic and Religion in the Second Century (Cambridge: Cambridge University Press, 2021).


For other recent projects that describe Christian figures as Antonine or Severan intellectuals, see L. Nasrallah, “Mapping the World: Justin, Tatian, Lucian, and the Second Sophistic,” Harvard Theological Review 98 (2005) 283–314; K. Eshleman, The Social World of Intellectuals in the Roman Empire: Sophists, Philosophers, and Christians (Cambridge: Cambridge University Press, 2012); J. König, Saints and Symposiasts: The Literature of Food and the Symposium in Greco-Roman and Early Christian Culture (Cambridge: Cambridge University Press, 2012); König, Langlands, and Uden, eds., Literature and Culture; Heath, Clement; Secord, Christian Intellectuals; Walsh, Origins.


Lucian of Samosata might precede Celsus slightly. Yet although Lucian’s Peregrinus refers to Christian books and may satirize Jesus’ biography, it does not use the language of “Gospel.” For Lucian on Christians, see J.A. North, “Pagan Attitudes,” in Christianity in the Second Century: Themes and Developments (ed. J.N. Carleton Paget and J. Lieu; Cambridge: Cambridge University Press, 2017) 265–280. There is another potential connection: Lucian’s Alexander is addressed to a person named Celsus. Peter Thonemann has recently argued that this Celsus is the same as the critic of Christianity to whom Origen responds (Lucian: Alexander or the False Prophet [Oxford: Oxford University Press, 2021] 62–64). On potential links between Lucian and Celsus, see M.M. Mitchell, “Origen, Celsus and Lucian on the ‘Dénouement of the Drama’ of the Gospels,” in Reading Religions in the Ancient World: Essays Presented to Robert McQueen Grant on His 90th Birthday (ed. D. Aune and R.D. Young; Leiden: Brill, 2007) 215–236. On non-Christian reading of Gospel books, see I.N. Mills, “Pagan Readers of Christian Scripture: The Role of Books in Early Autobiographical Conversion Narratives,” Vigiliae Christianae 73 (2019) 481–506. Alexander sees Celsus as an exception to the rule, citing Tertullian that “no-one turns to our literature who is not already a Christian” (Test. 1.4; cf. L. Alexander, “The Four Among Pagans,” in The Written Gospel [ed. M. Bockmuehl and D.A. Hagner; Cambridge: Cambridge University Press, 2005] 222–237 at 223).


Carleton Paget and Gathercole date Celsus’ treatise to the 160s or 170s CE and Origen’s response to 248–249 CE (Celsus, 1, 8–9). Eusebius’ discussion in Hist. eccl. 6.36.2 suggests that Contra Celsum was written during the reign of Philip the Arab (244–249 CE). The work must have been written before Origen’s death, ca. 254 CE. It is difficult to situate Celsus geographically, although Maren Niehoff argues for an Alexandrian context: “A Jewish Critique of Christianity from Second-Century Alexandria: Revisiting the Jew Mentioned in Contra Celsum,” Journal of Early Christian Studies 21 (2013) 151–175 at 154–155. On the date and context, see further M. Frede, “Origen’s Treatise Against Celsus,” in Apologetics in the Roman Empire: Pagans, Jews, and Christians (ed. M. Edwards, M. Goodman, S. Price, and C. Rowland; Oxford: Oxford University Press, 1999) 131–155.


Translations of Cels. adapt H. Chadwick, trans., Origen of Alexandria: Contra Celsum (Cambridge: Cambridge University Press, 1965). For Greek texts, see M. Borret, ed., Origène: Contre Celse (5 vols.; SC 132, 136, 147, 150, 227; Paris: Cerf, 1967).


E.g., Alexander, “Four,” 223; see further discussion below. I disagree with Alexander about Celsus’ knowledge of Gospel books and about the logic of his critique of Gospel difference. Nonetheless, Alexander’s discussion of Celsus’ objections to Gospel literature as morally defective and historically implausible remains compelling; cf. R.L. Wilken, The Christians as the Romans Saw Them (New Haven: Yale University Press, 2003) 108–112; Mitchell, “Origen, Celsus and Lucian”; F. Watson, The Fourfold Gospel: A Theological Reading of the New Testament Portraits of Jesus (Grand Rapids: Baker Academic, 2016) 175–179.


T. Baarda, “Διαφωνία—Συμφωνία: Factors in the Harmonization of the Gospels, Especially in the Diatessaron of Tatian,” in Gospel Traditions in the Second Century: Origins, Recensions, and Transmission (ed. W.L. Petersen and B. Aland; Notre Dame: University of Notre Dame Press, 1989) 133–154; D.C. Parker, An Introduction to the New Testament Manuscripts and Their Texts (Cambridge: Cambridge University Press, 2008) 334. A similar model is advocated by F. Watson in “Harmony or Gospel? On the Genre of the (So-Called) Diatessaron,” in The Gospel of Tatian: Exploring the Nature and Text of the Diatessaron (ed. M.R. Crawford and N.J. Zola; London: T&T Clark, 2019) 69–92 at 77–78; M.R. Crawford, The Eusebian Canon Tables: Ordering Textual Knowledge in Late Antiquity (Oxford: Oxford University Press, 2019) 59. Cf. W.L. Petersen, Tatian’s Diatessaron: Its Creation, Dissemination, Significance, and History in Scholarship (Leiden: Brill, 1994) 26, 73, 75.


On Irenaeus’ approach to fourfold Gospel and quadriform cosmos, see recent discussion in J. Coogan, “Reading (in) a Quadriform Cosmos: Gospel Books and the Early Christian Bibliographic Imagination,” Journal of Early Christian Studies 31 (2023) 85–103.


K. ní Mheallaigh, Reading Fiction with Lucian: Fakes, Freaks, and Hyperreality (Cambridge: Cambridge University Press, 2014). The focus on the “documentary” and the “pseudo-documentary” is a theme throughout ní Mheallaigh’s chs. 4–6. On the documentary sensibilities of the Antonine and Severan period, see T. Geue, “Keeping/Losing Records, Keeping/Losing Faith: Suetonius and Justin Do the Document,” in Literature and Culture in the Roman Empire, 96–235: Cross-Cultural Interactions (ed. A. König, R. Langlands, and J. Uden; Cambridge: Cambridge University Press, 2020) 203–222, as well as the recent essays in J. Arthur-Montagne, S.J. DiGiulio, and I.N.I. Kuin (eds.) Documentality: New Approaches to Written Documents in Imperial Life and Literature (Berlin: De Gruyter, 2022).


Readers used the concept of ὑπόθεσις to describe the shared textual architecture uniting different works. As defined by ancient thinkers (e.g., Hero, Def. 138.8), this is a more specific form of category-making than “genre.” From Dio Chrysostom comparing different plays titled Philoctetes (Or. 52.2; cf. Quintilian, Inst. 10.1.66) to Galen discussing varied treatises On Surgery (Hipp. off. med. 1.1; cf. HVA 1.4), Roman intellectuals used the concept of “textual architecture” (ὑπόθεσις) or “form” (εἶδος)—neither is the same as “genre” (γένος)—to categorize works with a shared pedagogical, narrative, or argumentative structure. Irenaeus (e.g., Haer. 1.8.1; 1.9.3–4) deploys the concept of ὑπόθεσις for Christian writings, including Gospels.


On second-century critiques of Christians’ inadequate education, see L. Alexander, “Gospel according to Celsus,” in Celsus in His World: Philosophy, Polemic and Religion in the Second Century (ed. J.N. Carleton Paget and S. Gathercole; Cambridge: Cambridge University Press, 2021) 254–294; Wilken, Christians, 95–98. My argument in this article resonates with Mitchell’s analysis of the phrase τὴν καταστροφὴν τοῦ δράματος (“the dénoument of the drama,” Cels. 2.55) in the writings of Lucian and Celsus (“Origen, Celsus and Lucian”). While Mitchell focuses on literary form rather than on textual meddling, she traces another way in which second-century critics (including Celsus) depicted Christians as incompetent readers who lack discernment. Likewise focused on literary structure, note Piscini’s recent discussion of Gospel ἀκολουθία in Celsus and Origen; Piscini traces how both figures accuse their opponents of reading without discernment (G. Piscini, “De l’exégèse à la polémique: la notion d’ἀκολουθία dans les tomes 1–2 du Contre Celse d’Origène,” Vigiliae Christianae 74 [2020] 199–221).


In using the language of “subaltern,” I draw upon postcolonial literary critic and political philosopher Gayatri Chakravorty Spivak, especially “Subaltern Talk: Interview with the Editors (1993–94),” in The Spivak Reader: Selected Works of Gayatri Chakravorty Spivak (ed. D. Landry and G. MacLean; New York: Routledge, 1996) 287–308. Spivak, in turn, draws on Ranajit Guha and other members of the Subaltern Studies group, and ultimately on Antonio Gramsci. For Spivak, the “subaltern” is the oppositional counterpart to the “intellectual” (whether elites from the imperial metropole or local functionaries); this makes her framework apt for thinking about the confluence of empire, education, gender, and class that we find in polemics from the second- and third-century CE Mediterranean.


On “bad reading,” I draw upon M. Emre, Paraliterary: The Making of Bad Readers in Postwar America (Chicago: University of Chicago Press, 2017). Emre describes “the historically contingent production of specific kinds of bad and good readers, whose matter-of-fact opposition to one other—and, more implicitly, their definition through one another—was negotiated in more materially and imaginatively complex ways than the terms good and bad could ever convey” (4).


I use gendered language intentionally. While women, children, and lower-status (including enslaved) men sometimes had access to the intellectual resources of παιδεία, rhetorical constructions of παιδεία denied this possibility.


On polemics against lower-status readers and writers, see further J. Coogan, C.R. Moss, and J.A. Howley, “The Socioeconomics of Fabrication: Textuality, Authenticity, and Class in the Roman Mediterranean,” forthcoming.


Libr. prop. 1–2 (= 19.8–9 Kühn). Text: V. Boudon-Millot, ed., Galien, t. I: Introduction générale, Sur l’ordre de ses propres livres, Sur ses propres livres, Que l’excellent médecin est aussi philosophe, texte établi, traduit et annoté (Paris: Les belles lettres, 2007) 134–135.


Galen presents himself as this learned reader, able to recognize errors in titles and texts. This has been discussed—in relation to this and other passages from Galen’s corpus—by Johnson, Readers, 74–96; C. Bubb, “Medical Literature and Medicine: Going Beyond the Practical,” in Medicine and the Law under the Roman Empire (ed. C. Bubb, and M. Peachin; Oxford: Oxford University Press, 2023), forthcoming; J. Coogan, “The Ways that Parted in the Library: The Gospels according to Matthew and according to the Hebrews in Late Ancient Heresiology,” Journal of Ecclesiastical History (2022) ( T. Raiola similarly traces Galen’s self-presentation as the consummate philologist, physician, and philosopher (Nel tempo di una vita: Studi sull’autobiografia in Galeno [Rome: Fabrizio Serra, 2015] 49–73); cf. V. Boudon-Millot, “Galien par lui-même: Les traits biobibliographiques (De ordine librorum suorum et De libris propriis),” in Studi su Galeno: Scienza, filosofia, retorica e filologia (ed. D. Manetti; Florence: Samerl, 2000) 119–133.


Galen complains in Libr. prop. 1 (19.9 Kühn) that many people “cannot even read well,” but attend lectures in the advanced subjects (θεωρήματα) of philosophy and medicine. Cf. Gellius, NA 1.71 on the hasty desire to correct a manuscript that is prepared with care.


C. Lambert, “The Ancient Entomological Bookworm,” Arethusa 53 (2019) 1–24 at 15–21. While Lambert investigates the metaphor of the bookworm for reading presented as “fragmented” or destructive, as well as the entomological realities that generate this metaphor, Lambert does not discuss how humans meddle with texts. See further C. Lambert, “Bad Readers in Ancient Rome” (PhD diss., Columbia University, 2022).


As Galen asserts, describing such meddlesome readers, “my books have been subject to all sorts of mutilations […] with omissions, additions, and changes” (Libr. prop. 1). Raffaella Cribiore reads “cuts, additions, and alterations” as referring to the public release of Galen’s works before he had revised them for that purpose (“The Dissemination of Texts in the High Empire,” American Journal of Philology 140 [2019] 255–290 at 263). While that is a concern of Galen’s, that is not what he is discussing in this context: His complaint is that others have meddled with his writings.


We see the language of ἐπανόρθωσις in Galen, Hipp. epid. VI (below); cf. Galen’s discussion of his editorial activities in Ind. 14. This is what the Theodotians are said to attempt (unsuccessfully) according to the Little Labyrinth (διωρθωκέναι, Hist. eccl. 5.28.15; κατωρθωμένα, 5.28.17); see further below. On διόρθωσις, see B. Neuschäfer, Origenes als Philologe (Basel: Friedrich Reinhardt, 1987) esp. 85–138; F. Young, Biblical Exegesis and the Formation of Christian Culture (Cambridge: Cambridge University Press, 1997) 76–86; F. Montanari, “Un papiro di Eschine con correzioni (P. Oxy. 2404): Considerazioni sull’ekdosis alessandrina,” Archiv für Papyrusforschung 55 (2009) 401–411; P.W. Martens, Origen and Scripture: The Contours of the Exegetical Life (Oxford: Oxford University Press, 2012) 42–49; F. Montana, “Alexander of Cotiaeum: Teacher, Exegete, Diorthotes,” AION 40 (2018) 7–22 at 15–20.


Galen, Hipp. epid. III (ed. Wenkebach, CMG 87 = Kühn 17a.618–619); trans. modified from S. Johnstone, “A New History of Libraries and Books in the Hellenistic Period,” Classical Antiquity 33 (2014) 347–393 at 363.


Galen, Hipp. epid. VI (ed. Wenkebach, CMG 3–4 = Kühn 17a.793–794; trans. modified from Bubb, “Medical Literature”). Galen on παλαιὰ ἀντίγραφα: Comp. med. loc. 12.814 Kühn; Hipp. prorrh. 16.491; 16.628; 16.738; 16.746; 16.751; Hipp. epid. VI 17a.793; 17a.809; 17a.824; 17a.914; 17a.973; 17a.975; 17a.1005; 17b.11; 17b.15; 17b.242; Hipp. prog. 18b.62; 18b.267; Hipp. off. med. 18b.730; 18b.853. On Galen’s manuscript-based scholarship, see H. von Staden, “Staging the Past, Staging Oneself: Galen on Hellenistic Exegetical Traditions,” in Galen and the World of Knowledge (ed. C. Gill, T. Whitmarsh, and J. Wilkins; Cambridge: Cambridge University Press, 2009) 144–150; Bubb, “Medical Literature.” Information about the χαρακτῆρες appears in Gal., Hipp. epid. III 17a.524–528, 558–561, 600–613, 617–625, 631–634, 731, 751, 765–766 Kühn (= ed. Wenkebach, CMG


This passage is discussed in Bubb, “Medical Literature”; cf. S.A. Gurd, “Galen on ἔκδοσις,” in Perceptions of the Second Sophistic and Its Times—Regards sur la Seconde Sophistique et son époque (ed. P. Fleury and T. Schmidt; Toronto: University of Toronto Press, 2010) 169–184. Gurd discusses Galen’s editorial labors on the Hippocratic corpus, but not this passage.


We could multiply examples. Lucian of Samosata’s “ignorant book collector” has an expensive library but is swindled by booksellers because he lacks the discernment to recognize ancient manuscripts and accurate texts (Lucian, Ind. 1–2, 4–5). Apellicon of Teos is accused of meddling with the manuscripts of Aristotle and Theophrastus in his misguided attempts to repair them (Strabo, Geogr. 13.1.54; Plutarch, Sulla 26.1–2; Aristocles, frag. 2, apud Eusebius, Praep. ev. 15.2.13; Athenaeus, Deipn. 5.214d–e). The oft-repeated account about Apellicon is historically unreliable, but the story offers a window into elite Roman discourses about appropriate and inappropriate textual intervention.


N. Horsfall, “Rome Without Spectacles,” Greece & Rome 42 (1995) 49–56; S. Blake, “In manus: Pliny’s Letters and the Arts of Mastery,” in Roman Literary Cultures: Domestic Politics, Revolutionary Poetics, Civic Spectacle (ed. A. Keith and J. Edmondson; Toronto: University of Toronto Press, 2016) 89–107; C.R. Moss, “Fashioning Mark: Early Christian Discussions about the Scribe and Status of the Second Gospel,” New Testament Studies 67 (2021) 181–204; C.R. Moss, “The Secretary: Enslaved Workers, Stenography, and the Production of Early Christian Literature,” Journal of Theological Studies 74 (2023) (; J.A. Howley, “Despotics: The Theory and Practice of Domination in the Ancient Agronomists,” forthcoming.


To this extent, I build on Alexander, who writes, “Celsus clearly has a detailed knowledge of at least one Gospel text, probably (as Origen surmises) Matthew” (“Four,” 223, citing Cels. 1.34). Other scholars similarly conclude that Celsus’ primary engagement with written Gospel material is with Matthew. As M.Y. MacDonald writes, Celsus’ “account surpasses mere rumor, general impression, or stereotype. He clearly had a detailed knowledge of Christianity and had read such important texts as the Gospel of Matthew and probably various gnostic sources” (“Was Celsus Right? The Role of Women in the Expansion of Early Christianity,” in Early Christian Families in Context: An Interdisciplinary Dialogue [ed. D.L. Balch and C. Osiek; Grand Rapids: Eerdmans, 2003] 157–184 at 159). Watson (Fourfold Gospel, 176) states that Celsus “knows that a number of versions of the Gospel are in circulation, but he is familiar primarily with the Matthean one.” On Celsus’ knowledge of the Gospels, compare Origen’s comments at Cels. 1.40: “[Celsus] takes from the Gospel of Matthew, and perhaps also from the other Gospels, the account of the dove alighting upon our Savior at his baptism by John, and desires to throw discredit upon the statement, alleging that the narrative is a fiction.”


Ed. Borret, SC 147:146–147.


Matt 28:5 (ἄγγελος); Mark 16:5 (νεάνισκος); Luke 24:4 (ἄνδρες δύο); John 20:12 (δύο ἄγγελοι).


Ed. Borret, SC 147:154–157.


The extant material from Origen’s Commentary on Matthew does not preserve his discussion of the number of angels at the tomb. The text is edited in E. Klostermann and E. Benz, eds., Origenes Werke: Band 10. Origenes Matthäuserklärung. Teil I. Die griechisch erhaltenen Tomoi (Leipzig: Teubner, 1935); E. Klostermann, E. Benz, and U. Treu, eds., Origenes Werke: Band 11. Origenes Matthäuserklärung. Teil II. Die lateinische Übersetzung der Commentariorum Series (Leipzig: Teubner, 1976).


Ed. Borret, SC 132:356–357.


Just as Galen demonstrates expertise by collecting and comparing various manuscripts, Celsus implies that he has consulted multiple Gospel books. Whether Celsus has done so is unknown; the sting of the accusation does not require him to have done this work.


On Celsus’ failure to compare the genealogies, see further Frede, “Origen’s Treatise Against Celsus,” 151.


For other critiques of Celsus’ “selective” (so Loveday Alexander) or inattentive reading, see, e.g., Cels. 2.11, 24, 34 (cf. Alexander, “Four,” 223; Piscini, “De l’exégèse à la polémique”).


Origen’s extant Greek corpus includes several dozen discussions of divergent Gospel ἀντίγραφα and extensively compares different Gospels (e.g., Comm. Matt. 15.14; Comm. Jn. 10).


Galen describes this ill-formed approach to the world as ῥᾳδιουργία, which we might translate as “amateurishness” (Galen, Libr. prop. 4 = 19.9 Kühn). The verb (ῥᾳδιουργέω) is used to describe textual tampering in several early Christian texts, including Origen in Cels. 2.27 (ῥᾳδιουργῆσαι τὰ εὐαγγέλια, here Origen’s words and not those of Celsus) and the Little Labyrinth (ῥερᾳδιουργήκασιν, apud Eusebius, Hist. eccl. 5.28.13). Irenaeus mentions those who “tamper with the λόγια of the Lord” (ῥᾳδιουργοῦντες τὰ λόγια κυρίου, Haer.; this is placed in parallel with “evil interpreters” (ἐξηγηταὶ κακοί). Textual meddling and flawed exegesis go hand in hand. Irenaeus uses the term again to describe irresponsible interpretation at Haer. 1.1.6. Compare Galen’s comments about Zeno’s textual corruptions and flawed interpretations of Hippocrates (Hipp. epid. III = 17a.619 Kühn, discussed above). Galen complains about ῥᾳδιουργία elsewhere in his corpus as well (Diff. resp. 7.893; HNH 15.109; Hipp. prorrh. 16.629; Hipp. epid. III 17b.98 Kühn). The language of “daring” (τῶν τολμησάντων ῥᾳδιουργῆσαι τὰ εὐαγγέλια, Cels. 2.27) invites comparison with the accusations of acting “recklessly” (ἀφόβως, Hist. eccl. 5.28.13, 15) in the Little Labyrinth. In these contexts, it refers to those who act without appropriate qualifications. Supporting my argument that Origen was familiar with such polemics, we might note the arguments of Grant that Origen demonstrates knowledge of Galen’s works and even used Galen in the Contra Celsum: R.M. Grant, “Paul, Galen, and Origen,” Journal of Theological Studies 34 (1983) 533–536.


In his article focused on ἀκολουθία, Piscini similarly observes how Origen deploys both polemical strategies (“De l’exégèse à la polémique”).


Origen deploys similar polemics for corpora beyond the Gospels. In his Epistle to Africanus, Origen accuses Jewish readers of meddling with Hebrew scriptures. This is part of how Origen explains differences between Greek and Hebrew texts; it advances Origen’s claim that his preferred text (the Greek version of the Seventy) is superior. Much like Galen, Origen describes his philological intervention into the Greek scriptures at length, advertising his expertise in mending texts that others had damaged. Origen, Ep. Afr. 8–9, 13 (ed. M. Harl and N.R.M. de Lange, eds., Philocalie, 1–20: Sur les écritures et la lettre à Africanus sur l’histoire de Suzanne [Paris: Cerf, 1983] 532–535, 544–545).


Discussing these early Christian polemics, see Haines-Eitzen, Guardians, 37–38, 111–113 (Marcion and Theodotians); Martens, Origen, 107–132 (Valentinus, Basilides, and Marcion); and especially J. Knust and T. Wasserman, To Cast the First Stone: The Transmission of a Gospel Story (Princeton: Princeton University Press, 2018) 114–115 (Celsus and Marcion). None of these recent studies situates Christian invective in the context of the politics of reading that I have here discussed. While Knust and Wasserman briefly engage Celsus, their discussion of Cels. 2.27 is focused on the problem of a pluriform Gospel rather than on the polemical construction of textual intervention.


Quis tam comesor mus Ponticus quam qui evangelia corrosit? (ed. E. Evans, Tertullian: Adversus Marcionem [Oxford: Clarendon Press, 1972] 4–5). Polemic about “nibbling at” texts is, as Lambert demonstrates, directed against grammarians and other lower-status literate workers (Lambert, “Bookworm,” 15–21).


Marcion’s barbarian lack of culture is emphasized in the opening passage of Tertullian’s Marc., which draws on Herodotus to sketch a portrait of Pontic savagery and environmental and cultural desolation.


On Marcion’s textual meddling, see further Irenaeus, Haer. 3.11.7; Tertullian, Marc. 4.4; 4.6.2–3; 4.7; Epiphanius, Pan. 42.9.6; 42.10.3; 42.11.6.


This treatise is often attributed to Hippolytus of Rome. Excerpts are preserved by Eusebius (Hist. eccl. 5.28.3–6, 8–12, 13–19). I cite Eusebius from the GCS edition of Schwartz and Mommsen. The title is mentioned by the fifth-century Theodoret of Cyr in his Haereticarum fabularum compendium 1.5.2 (ed. B. Gleede, GCS N.F. 26:112 = PG 83: 392): ὁ σμικρὸς […] Λαβύρινθος. Theodoret paraphrases material cited by Eusebius. On the authorship and date of this treatise, the best discussion is J.T. Fitzgerald, “Eusebius and the Little Labyrinth,” in The Early Church in Its Context: Essays in Honor of Everett Ferguson (ed. A.J. Malherbe, F.W. Norris, and J.W. Thompson; Leiden: Brill, 1998) 120–146. The Little Labyrinth polemicizes against Theodotus’ circle—themselves devotees of Galen’s thought—in terms thoroughly suited to Galen’s own politics of reading. This may suggest that the author(s) of the Little Labyrinth are familiar with the polemical strategies that I have traced in this article. On the medical preoccupations of Theodotus’ circle, see J. Secord, “Galen and the Theodotians: Embryology and Adoptionism in the Christian Schools of Rome,” Studia Patristica 81 (2017) 51–63.


For these critics (as for Galen), the problem is not διόρθωσις itself but rather the incompetent attempt at διόρθωσις. The Little Labyrinth’s ἐπέβαλον (Hist. eccl. 5.28.15) may allude to technical terminology for manuscript collation (ἀντιβάλλειν); this phrasing also recalls the “casting out” (ἀποβάλλειν) of Theodotus (5.28.6) by Victor, bishop of Rome.


On Theodotus’ social context in second-century Rome, see H.G. Snyder, “Shoemakers and Syllogisms: Theodotus ‘the Cobbler’ and His School,” in Christian Teachers in Second-Century Rome: Schools and Students in the Ancient City (ed. H.G. Snyder; Leiden: Brill, 2020) 183–204. Snyder imagines the response of Roman elites to the intellectual enterprises of these craftspersons as surprise and even, perhaps, “appreciation” (183), but—as I have suggested—the evidence instead points to a response of disdain. Cf. Coogan, Moss, and Howley, “Socioeconomics of Fabrication.”


Compare the use of παραχαράσσειν with the use of μεταχαράττειν in Cels. 2.27. In Haer. 3.5, Irenaeus uses the language of παραχαράσσειν to describe Marcion as “one of those who debases the truth” (τινι τῶν παραχαρασσόντων τὴν ἀλήθειαν), but textual meddling is not directly implied. For παραχαράσσειν in the sense of debasing teaching or thought (but not necessarily text), see Justin, Dial. 82.3; Tatian, Or. 40.1.


On Christians as lacking education, see Cels. 1.27; 3.50, 55, 75; 6.12, 14. Celsus polemicizes against Christians in terms of gender, ethnicity, and social status. For Celsus’ attempts to contrast his own superlative παιδεία with the deficient formation of Christians, see, e.g., Cels. 1.8; 3.44, 49, 73, 75; 6.1; 7.45.


Here I recall the work of Emre, cited above in n. 18.


S. Ahmed, Queer Phenomenology (Durham: Duke University Press, 2006) 33. As Ahmed writes, this “fantasy […] can be understood as crucial not only to the gendered nature of the occupation of philosophy but also to the disappearance of the political economy, of the ‘materials’ of philosophy as well as its dependence on forms of labor, both domestic and otherwise” (33). Crucial to this account is the way that “paperless philosophy” seeks to erase labor and expertise. Cf. S. Ahmed, What’s the Use? On the Uses of Use (Durham: Duke University Press, 2019) 206.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 826 826 53
PDF Views & Downloads 1368 1367 45