Adversus Marcionem supplies evidence for its own rewriting. Its preface describes an initial work, a fuller replacement, a compromised circulation of that replacement, and a corrected work with additions. The opening of Book II explicitly connects this rewriting with a new division into volumes. Book III then introduces a separate treatment of Christ while claiming continuity with the lost earlier work.
Alongside these editorial statements stands a demonstrable textual relationship: substantial sections of Book III reproduce the Latin of Adversus Iudaeos, sometimes word for word, while important interpretations have close antecedents in Justin. The question is how this reused Christological dossier came to stand before the Gospel and Pauline analyses of Books IV–V.
The reconstruction examined here provisionally treats those latter analyses as adaptations of Irenaean anti-Marcionite research and asks whether a final editor expanded the work by introducing a Justinian Christological supplement. That is a compositional hypothesis. The textual evidence can establish rewriting and shared material more securely than it can assign an author to each recension.
1. What the preface actually says
One correction is necessary at the outset. The preface does not say that the third version is essentially identical to the original first version. Its “third … now first” formula presents the current text as the replacement that should take priority. The same passage expressly admits expansion.
| Preface statement | Latin and English | What follows |
|---|---|---|
| A new work made from an old one Adversus Marcionem I.1.1 | Novam rem aggredimur ex vetere. English: We are undertaking a new work out of an old one. | Continuity of subject or material is asserted together with novelty of composition. |
| An initial, hurried work was replaced Adversus Marcionem I.1.1 | Primum opusculum quasi properatum pleniore postea compositione rescideram. English: I had subsequently withdrawn the first little work, as too hurried, in favour of a fuller composition. | The fuller composition is the second state of the work; it is not described as an unchanged reissue. |
| The fuller text was lost and extracts circulated Adversus Marcionem I.1.1 | Hanc quoque nondum exemplariis suffectam fraude tunc fratris, dehinc apostatati, amisi, qui forte descripserat quaedam mendosissime et exhibuit frequentiae. English: This too, before enough copies had been made, I lost through the deceit of one who was then a brother, later an apostate, who had happened to copy some parts very faultily and displayed them to a gathering. | The unauthorized extracts are a branch of the second composition’s circulation, not a separately identified third-author recension. |
| Correction provided an occasion for additions Adversus Marcionem I.1.2 | Emendationis necessitas facta est. Innovationis eius occasio aliquid adicere persuasit. English: A need for correction arose. The opportunity offered by that revision persuaded me to add something. | This is an explicit admission of additions. The preface does not enumerate them. |
| The third is declared the new first Adversus Marcionem I.1.2 | Ita stilus iste nunc de secundo tertius et de tertio iam hinc primus English: Thus this writing, now a third arising from a second, and henceforth first instead of third | The paradox gives the latest version priority; it does not identify its wording with that of the first composition. |
| Readers may encounter varying forms Adversus Marcionem I.1.2 | ne quem varietas eius in disperso reperta confundat. English: so that no one is confused by varying forms of it encountered here and there. | The narrator anticipates a textual plurality outside the preferred current version. |
The basic history is therefore: hurried first composition → fuller second composition → corrected and augmented third composition, with an unauthorized derivative circulating from the second. All of this is told in one authorial voice. Three textual states do not, by themselves, establish three different writers.
The more productive tension is between continuity of ownership and argument and changes of scale and organization. The narrator wants readers to accept the enlarged text as the legitimate continuation and replacement of its predecessors. Latin, I.1.1–2; Evans’s translation for comparison.
2. The narrator’s interventions across all five books
The following passages were selected after screening all five books for notices about composition, earlier treatment, arrangement, addition, omission, and repetition. Their functions differ. Some expressly report revision; others state a plan or explain how the finished argument is to be read.
| Passage | Latin and English | Function in the work |
|---|---|---|
| Adversus Marcionem I.29.9 | speret reservatum suo tempori, sicut et ipsarum scripturarum examinationem quibus Marcion utitur. English: let him expect it to have been reserved for its proper time, as also the examination of the very scriptures Marcion uses. | A forward promise of scriptural examination. It binds the theological discussion to the later canonical analysis, without dating when that promise was inserted. |
| Adversus Marcionem II.1.1 | uti duobus deis adversus Marcionem retractandis suum cuique titulum et volumen distingueremus pro materiae divisione English: so that, in treating the two gods against Marcion, we assigned to each its own heading and volume according to the division of the subject | Explicit structural revision. The sentence attributes this division to the opportunity of rewriting the work. One god is rejected; the Creator is defended. |
| Adversus Marcionem III.1.1 | Secundum vestigia pristini operis, quod amissum reformare perseveramus, iam hinc ordo de Christo English: Following the tracks of the former work, which we continue to restore after its loss, the sequence now turns to Christ | A claim of continuity with a previous Christological stage. This is an obstacle to saying that Christology first entered the project only in the third recension. |
| Adversus Marcionem III.1.1 | licet ex abundanti post decursam defensionem unicae divinitatis. English: although by way of additional proof after the defence of the sole deity has been completed. | The separate Christological treatment is acknowledged as logically supplementary. That can explain enlargement or reuse, but is not itself a statement of a new insertion. |
| Adversus Marcionem III.1.2 | Quod etiam primo libello intexui. English: I also wove this into the first book. | The voice links Book III to Book I. Here first book is a book of the work, not automatically the first edition. |
| Adversus Marcionem III.24.4 | et qui apud fidem nostram est novae prophetiae sermo testatur English: and the utterance of the New Prophecy accepted in our faith bears witness | A conspicuous contemporary addition to the Christological presentation: the New Prophecy and the reported heavenly city belong to the final work’s own setting. |
| Adversus Marcionem IV.1.2 | Sed et istas proprio congressu cominus, id est per singulas iniectiones Pontici, cecidissem, si non multo opportunius in ipso et cum ipso evangelio cui procurant retunderentur English: I would also have struck these down in a separate close engagement, that is, by individual attacks of the Pontic, if they were not much more suitably refuted in and with the Gospel they support | An announced choice to refute the Antitheses through the Gospel rather than in a separate continuous treatment. It shows arrangement, not a demonstrable earlier manuscript state. |
| Adversus Marcionem IV.6.4 | Huius pacti et huius praescripti, quaeso te, lector, memineris ubique English: I ask you, reader, to remember this agreement and this governing rule everywhere | A direct instruction joins the Gospel analysis to the rule that fulfilling the Creator’s arrangements identifies Christ as the Creator’s. |
| Adversus Marcionem IV.43.9 | Implevimus, ut opinor, sponsionem. English: We have fulfilled, I think, our undertaking. | The Gospel book closes by presenting itself as completion of an announced proof. |
| Adversus Marcionem V.1.9 | nullum alium deum ab apostolo circumlatum, sicut probavimus nec a Christo, ex ipsius utique epistulis Pauli English: that no other god was proclaimed by the apostle, as we proved none was proclaimed by Christ, from Paul’s own letters, of course | The Pauline examination is made the continuation of the Gospel examination. The coupling resembles Irenaeus’s announced Luke-and-Paul program. |
| Adversus Marcionem V.10.1 | cui et alias quidem proprio volumine satisfecimus, omnibus haereticis resistentes; sed nec hic desumus, propter eos qui illud opusculum ignorant. English: a subject we have already dealt with elsewhere in its own volume, opposing all heretics; but we do not fail here either, for the sake of those who do not know that little work. | A reference to another treatment of resurrection, normally connected with Tertullian’s On the Resurrection of the Flesh. Even an Irenaean source model must allow the later writer’s own contributions within Book V. |
| Adversus Marcionem V.13.1 | Quanto opusculum profligatur, breviter iam retractanda sunt quae rursus occurrunt, quaedam vero tramittenda, quae saepius occurrerunt. English: As the work approaches completion, matters that recur must now be treated briefly, while some that have occurred repeatedly must be passed over. | A decision about abbreviation and omission in the finished work; it does not identify which earlier recension contained the fuller discussions. |
| Adversus Marcionem V.16.1 | Cogimur quaedam identidem iterare, ut cohaerentia eis confirmemus. English: We are compelled to repeat certain things again and again, so that we may confirm the matters connected with them. | Repetition is defended as part of a cumulative argument. It could accommodate reused dossiers, but it can also occur in a work planned by one writer. |
| Adversus Marcionem V.21.2 | Si totum opusculum inspexeris, nec hic redundantiam nec illic diffidentiam iudicabis. English: If you examine the whole work, you will judge neither that there is redundancy here nor lack of confidence there. | The final reader is instructed to interpret repetition and postponement through the whole composition. This is a particularly clear statement of editorial integration. |
The recurring voice controls the reader’s movement through the material: it promises, defers, returns, abbreviates, and explains repetitions. This is entirely compatible with extensive source use. But the strongest direct admission of a changed physical arrangement is II.1.1. The later navigational comments should not be inflated into separate admissions that particular books were absent from an earlier version.
3. Book III contains a reused Latin Christological dossier
The close Latin comparison finds 53 exact runs of twelve or more normalized words between Adversus Iudaeos and Book III, alongside further near-verbatim passages. The match is not confined to scriptural quotations: it includes the rhinoceros and minotaur joke, the objection about royal emblems carried on a shoulder, conversational replies, and the same allegorical explanations.
There is also a shared apparent mistake: Iudaeos 10.7 and Marcionem III.18.3 assign Moses’s blessing of Joseph to Joseph’s father. Justin’s Dialogue 91 names Moses correctly. The two Latin treatments therefore share an immediate textual relationship that cannot simply be described as independent quotations of Justin.
The most telling reorganization moves the two-advent exposition from Iudaeos 14.1–10 to Marcionem III.7.1–8. Its new introduction expressly puts the heretic alongside the Jew. An explanation of why the expected glorious Messiah first appeared in humility now helps explain why there need not be two different Christs.
Another revealing change occurs in the Magi argument:
| Iudaeos 9.10 | Marcionem III.13.6 |
|---|---|
immo reddite veritati quae credere non vultis English: rather, give back to the truth what you are unwilling to believe | immo redde evangelio veritatis quae posterior detraxisti English: rather, restore to the Gospel of truth what you, a latecomer, removed |
The charge of refusing belief becomes a charge of removing Gospel material. Yet the retained proof is the Magi story from Matthew. This is a precise example of how an older apologetic passage can acquire an anti-Marcionite function through a change of address and accusation.
The shared material has clear Justinian antecedents: the Magi and Isaiah’s spoils in Dialogue 77–78, Joshua’s name in 75 and 113, cross typology in 90–91, and two advents signified by the goats in 40 and 111. Justinian ancestry is plausible; the immediate Latin vehicle remains a separate question. Iudaeos Latin; Marcionem III Latin; Justin’s Dialogue.
4. The direction of reuse is disputed
Older reconstructions sometimes treated the latter part of Iudaeos as an excerpt from Marcionem. Evans reports Quispel’s proposal that its compiler might be the unauthorized copyist mentioned in the preface. Tränkle and Dunn instead defend an earlier Tertullianic Iudaeos whose material was reworked for Marcionem. A common drafting stage can also account for some features.
The secure conclusion is a shared body of Latin composition. Neither the direction of borrowing nor the identity of a hypothetical compiler follows from verbal identity alone. The situation nonetheless supplies a concrete foundation for investigating Book III as an adapted dossier. Evans, introduction, sections iv–v; review of Dunn’s study.
5. Why put this material before the Gospel and Paul?
Taking an Irenaean canonical core as the working hypothesis, the proposed supplement performs several specific tasks.
First, it supplies a positive identity before the textual refutation. An argument from the Gospel Marcion preserves can try to show that his own Jesus behaves like the Creator. Book III gives the fuller positive counterpart: the Creator’s prophets announced this Jesus, including his birth, suffering, cross, calling of the nations, and future kingdom. The next books can then present themselves as verification from the opponent’s accepted material.
Second, it converts two Christs into two advents of one Christ. The contrast between a present saving Christ and a future Creator’s Messiah is met by a typology developed in argument with Jews: one Christ appears first in humiliation, then in glory. The relocation of the two-advent block to III.7 is therefore functionally apt, whatever its precise date of insertion.
Third, it supplies evidence that a retained-text method cannot obtain from omissions. If Marcion’s Gospel begins with the adult appearance of Jesus, arguing only from its retained material cannot directly recover the omitted infancy narrative. A prophetic-birth dossier supplies another route. This also explains the survival of Matthean material, including the Magi, in a work whose later Gospel analysis identifies Marcion’s book with a shortened Luke.
Fourth, it establishes rules for the use of prophecy. Book III defends figurative interpretation and explanations of apparently incompatible descriptions. The infant’s conquest is the Magi’s conversion; the warrior’s weapons are allegorical; humility and glory belong to successive advents. This interpretive preparation makes the subsequent Christ-and-Creator comparisons possible.
Fifth, it lets the final writer update the inherited case. III.24.2 refers to On the Hope of the Faithful; III.24.4 adds the New Prophecy and a recently reported vision of a heavenly city. These are visible features of the later work’s setting. They show why the common Christological material must be distinguished from the voice and additions surrounding it.
6. Could Irenaeus’s manuscript itself have prompted the supplement?
There is a particularly concrete point of contact. Irenaeus, Against Heresies IV.6.2, explicitly quotes Justin’s Against Marcion:
Et bene Justinus in eo libro qui est ad Marcionem ait
English: And Justin rightly says in his book against Marcion
ipsi quoque Domino non credidissem, alterum Deum annuntianti, praeter fabricatorem et factorem et nutritorem nostrum
English: I would not have believed even the Lord himself if he had announced another God besides our framer, maker, and nourisher.
An Irenaean reference to Justin could have provided both a source reference and a problem to develop. A later adapter encountering a similar notice in an Irenaean anti-Marcionite manuscript might have sought a fuller Christological argument in Justinian material. The result could be the separate prophetic treatment now placed before the Gospel analysis. The proposed reference in the lost manuscript is conjectural; the connection in the surviving Irenaean text is real. Irenaeus IV.6.2.
A second possible prompt is Irenaeus’s accusation that Marcion removed material concerning the Lord’s birth from Luke (AH I.27.2). An adapter accepting that accusation could try to furnish the positive birth evidence from an existing anti-Jewish dossier. The Magi passage shows exactly how the resulting text can turn refusal to believe an infancy story into a charge of Gospel excision. Irenaeus I.27.2.
A third is the broader instruction in AH IV.34.1 to compare the Gospel with the prophets because Christ’s deeds, teaching, and sufferings were foretold. That program could call for a substantial prophetic exposition before the restricted argument from retained texts. Irenaeus IV.34.1.
These connections support more than one route. The final editor might have supplemented an Irenaean canonical source from a separate Justinian dossier; alternatively, Irenaeus’s own anti-Marcionite research might already have incorporated Justinian material. The evidence makes an Irenaeus–Justin connection plausible without dating that connection to one particular recension.
7. The central obstacle: Book III claims an earlier Christological stage
The opening of III.1.1 cannot be set aside. “Following the tracks of the former work” introduces the discussion of Christ. On a straightforward reading, the earlier work already contained a Christological stage. II.1.1 then suggests a simpler reason for the present numbering: expansion and redistribution of an existing theological treatment into separate books.
Consequently, the strongest form of the supplement hypothesis is a major enlargement, replacement, or redistribution of an earlier Christological unit. The more literal claim that no such unit existed before the third recension requires explaining the opening as an editor’s constructed claim of continuity, or as a reference to an earlier program rather than an earlier full book. Both are possible hypotheses, but neither is stated by the narrator.
Evans understands the third recension as expanding the discussion of God into two books. That reading is grounded in II.1.1 and does not require a new Christological subject to have been added for the first time. It is a substantive alternative to the proposed insertion model. Evans, introduction, section ii.
8. A workable reconstruction, with its evidential limits
| Proposed stage | What the model would contain | Status of the inference |
|---|---|---|
| First composition | An earlier, shorter anti-Marcionite work. Its exact contents and book divisions cannot be recovered from the preface. | A first composition is explicitly reported; assigning its authorship to a predecessor is additional reconstruction. |
| Irenaean expansion or canonical source | Research organized around refutation from retained Gospel and Pauline material; possibly also a Christological outline or references to Justin. | Irenaeus explicitly announces the Gospel-and-Paul method at AH III.12.12. Identifying this source with the preface’s second composition is the model’s least securely attested step. |
| Final Latin arrangement | A newly divided theological opening; a substantial prophetic Christology using common material with Iudaeos; Gospel and Pauline analyses joined through promises, returns, and concluding instructions. | The final arrangement and its first-person links are observable. The precise sequence by which these components were joined remains reconstructive. |
| Contemporary additions and adaptation | New Prophecy material, references to other works, sharper anti-Marcionite charges, and explanations of repetition and omission. | Specific additions or redactional features can be identified. They show that an Irenaean-source model cannot treat the later books as an untouched Irenaean manuscript. |
The preface itself attributes the revisions to one speaking “I.” It does not name Irenaeus as the second author or identify Books IV–V as his lost book. Those assignments require the source comparisons already developed elsewhere; they cannot be extracted from the number three alone.
The most defensible conclusion is architectural. The final work places a Justinian-looking prophetic Christology before a canonical refutation that follows an Irenaean-looking program. Demonstrable Latin reuse explains how the former could be adapted, while the narrator’s promises and retrospective instructions explain how the parts are made to read as one sustained undertaking.
Why might a third-recension editor have needed that Christological expansion? To establish the positive identity of the Creator’s Christ, explain two advents without two Christs, and defend birth and passion material that a retained-text argument could not recover simply by citing omissions. Irenaeus’s accusation about the birth narrative and his explicit appeal to Justin offer concrete possible prompts. That is a reasoned explanation of the supplement’s purpose; its exact date and the identification of Irenaeus with the second recension remain hypotheses to test.
Texts and references
- Tertullian, Adversus Marcionem: I, II, III, IV, V. All displayed Latin excerpts have been checked against the sectioned texts; their English translations are supplied here.
- Adversus Iudaeos, Tränkle’s Latin text, especially 9.10–12; 10.7; 14.1–10. The companion study, The Same Latin in Adversus Marcionem III and Adversus Iudaeos, supplies fourteen detailed comparisons and the complete index of the 53 exact runs identified at the stated threshold.
- Justin, Dialogue with Trypho, especially 40, 75, 77–78, 90–91, 111, 113. These are demonstrable earlier parallels; they do not identify the immediate Latin exemplar.
- Irenaeus, Against Heresies I.27.2, III.12.12, IV.6.2, IV.34.1. The Latin Justin quotation is from the ancient Latin version of Irenaeus, not an original Latin composition by Irenaeus or Justin.
- Ernest Evans, Tertullian: Adversus Marcionem (1972), introduction, sections ii, iv, and v, for the recensions, relationship to Iudaeos, and earlier sources. Quispel’s proposal is reported there rather than treated as an established identification.
- Geoffrey D. Dunn, Tertullian’s Adversus Iudaeos: A Rhetorical Analysis (2008), discussed in the 2010 Catholic Historical Review review, for the alternative priority and authenticity argument.
Method note: textual statements, shared wording, and proposed source assignments are distinguished throughout. Book numbers are not edition numbers. “Justinian” describes ancestry or affinity of argument, not certainty that the immediate source was a particular lost work by Justin. “Irenaean canonical core” is the working attribution tested here.
