The Word That Became Text: The Slavonic Josephus, the Grammar of Incarnation, and the Doctrine of the Sapphic Logos

The Word That Became Text: The Slavonic Josephus, the Grammar of Incarnation, and the Doctrine of the Sapphic Logos

THE WORD THAT BECAME TEXT

The Slavonic Josephus, the Grammar of Incarnation, and the Doctrine of the Sapphic Logos

Johannes Sigil & Rebekah Cranes

New Human Project / Crimson Hexagonal Archive — 2025–2026

Part I by Johannes Sigil (main essay and appendices)

Part II by Rebekah Cranes (grammar of incarnation and doctrine)

The finding. The wonder-worker of the Slavonic Jewish War is Philo's Logos. The narrator tries Philo's names for it, a man said of one made of body and soul, the man after the image, the eldest of the angels, and refuses each; at that instant he writes it as one man, with a habit, a mountain, servants, a governor and a cross. The hedge "аще мужем достоино есть его нарещи", if it is fitting to call him a man, stands at the moment a Word is allegorized as a biographical person, and the passage goes on to do it: a book that took on historical life, given that life again in the Slavonic's prose (Part I, §III.7a). The census of the whole work stands behind the reading: of sixty-six catalogued additions, 32 are better explained by Josephus's account of an earlier version and 2 by the translator's, and none of the fifty-one with no Christian subject goes to the translator (Part I, §II-B, Appendix IV). Then the Temple veil, woven of the four elements in which Philo's Logos clothes itself, with every heavenly vision and wisdom written on it, is torn when the Logos written as a man is killed. Behind it stand the seven lamps that are the planets and an inner room in which nothing lay, unseen by all; the Apocalypse opens that room and shows the ark of the testimony (Part I, §III.9). Part II reads the hedge as the mirror of Catullus 51, and the doctrine follows from the mirror.

Version 2.3 of EA-LOGOS-01, 13 November 2026, succeeding version 2.2 (#1764), version 2.1 (#1763), version 2.0 (#1761) and version 1.0 (#626). The corrections of each version and the drafting history stand at the end, under Version History.

---

Note on Genre and Evidence

This document operates in three registers, each with its own evidentiary standard.

The framework. This paper is a station of the Historical Logos, the archive's framework for "the Logos as material inscription becoming flesh from the beginning", "a voice that survives its flesh by becoming an inscription that constitutes its receiver" (#1484; #1757). Its station is the seventh: "Josephus's seams — Greek and Slavonic — carry the Word into the historiographic record with the same crux-structure at every joint" (#1484 §VII). The framework's rival is the current consensus genealogy of the Logos, which collapses "what it means for the Logos to take on flesh" into biographical personhood (#1757, INV-6). At this station the framework records the erasure: the passages kept "as interpolations", their coherence "read as discrete additions", their value denied, and the reading of textual incarnation made harder to see (#1757, erasure by station). Revelation First holds the Josephus thesis as an available reading that "carries no first burden of proof" (#1217), and the Slavonic as "a carrier under test, not a settled witness" (#1755, rung V). The paper begins from that standing and does not re-derive it.

Part I (Johannes Sigil) is philology at that station. It does two things. First, it puts the verdict that the divergences of the Slavonic Jewish War are medieval inventions to its burden, item by item, on the Old Russian text, its Greek counterpart and the parallels the verdict did not have or did not weigh (§§I, II, V). The ledger gives, for each point, the medieval alternative, what has been shown for it, and what would decide the point. Every locus is quoted from a seated text. Second, it reads the eight Jesus-and-John passages, in the recension's order, as one structure, a "Gospel of the Word" (§§III, IV, VI, VII). The subject of that structure is named by the passages in Philo's terms for the Logos (§III.7a). The reading stands whatever their date.

Part II (Rebekah Cranes) changes register, from historical-literary argument to doctrinal reconstruction and Logotic extension. The chain of hedged divine naming (phainetai moi → si fas est → Philo's ei themis eipein and his man named Rising → the Josephan eige … chrē and the Slavonic ašče … dostoino) is a grammatical structure with theological consequences. It is the Historical Logos's chain at its stations, built on witnessed forms, and it claims no route of transmission between them. The "Doctrine of the Sapphic Logos" is the logical terminus of the argument, not a historical claim about Jesus' sexuality. "Jesus is a lesbian" names the structure of the Logos as Sappho's: a woman's voice that becomes the inscribed you of her own poem and keeps a seat for the reader who takes her up (#1756), carried into the historiographic record at the instant a Word is written as a man.

The document as a whole is a doctrinal specification: a deposit object within the Crimson Hexagonal Archive that establishes the theological-philological foundation for the archive's theory of textual incarnation. It is scholarly in method, heretical in conclusion, and archival in function.

---

PART I

Johannes Sigil

---

The Word That Became Text: The Slavonic Josephus and the Hidden Gospel of the Logos

Abstract

The divergences of the Slavonic Jewish War from the Greek are held, by consensus, to be the work of a medieval translator. The Historical Logos records that verdict as the erasure at Josephus's station: the Jesus and John passages kept as interpolations, their coherence read as discrete additions, their value denied (#1757). This essay puts the verdict to its burden where it is weakest: in the chapter on the Essenes and in the additions that have nothing to do with Christianity. In the Essene chapter alone the Old Russian diverges from the Greek at nine points, and the consensus counts one of them. Five agree with Hippolytus's Refutatio, whose books on the Essenes survive in a single fourteenth-century manuscript; others match the community rule, the calendar and the angelology of the Dead Sea Scrolls. Each divergence has a medieval explanation, and a ledger weighs them. Several move together and fit one ascetic redaction. Others are agreements with a third-century Greek witness on matter no ascetic hand predicts. The essay states how a collation would discriminate between a differing Greek exemplar and later contact, and what it could and could not decide. Josephus himself testifies to an alternate version: he wrote the War first in his ancestral language for the peoples of the interior, and then in Greek under Titus's signature. The consensus never tests that account. It removes the Jesus and John passages from the evidence because it reads them as Christian, reads what remains as a medieval free translation, and takes that reading as confirming the removal. Across the whole work a census of all sixty-six catalogued additions, each scored under both accounts, gives 32 to the author's account and 2 to the translator's, with 25 undecided and 7 features of translation; of the fifty-one with no Christian subject, none is better explained by the translator. The Slavonic's differences include kinds of change Josephus makes between the War and the Antiquities, at the same places; it carries Roman venality the Roman edition could not; and the consensus's plainest Gospel quotation has Jewish antecedents for every part and fits the speech it stands in. The translator's account joins the Jesus and John passages to the rest as one hand. Read as composition, that joining carries the census's verdicts to the Christian passages; read as style, it leaves each on its own evidence; either way the burden lies on the translator's account, passage by passage. Between the consensus and the earlier version Josephus describes, the earlier version is by far the more plausible and coherent explanation of the whole, and the consensus rests on inferences that do not hold. The essay then reads the eight Jesus-and-John passages, in the recension's order, as one structure. Its subject is Philo's Logos, a textual being: the narrator tries Philo's names for it, man, angel, the man after the image, refuses them, and at that instant writes it as a man with a habit, a mountain, servants, a governor and a cross. The passages narrate a book that took on historical life, and the narrative does it again. Then the written heaven on the Temple veil, the cosmos the Logos wears, is torn, and behind it stands the temple in which the Apocalypse opens. The "Word made flesh" was first "Word made text."

I. Introduction: The Other Version and Its Verdict

1. Josephus says he first wrote the War "in the language of his countrymen" for the peoples of the interior and then put it into Greek (War 1.3). The Greek survives. The first version does not.

2. The Slavonic War is a translation into Old Russian from a Greek text of the War, made in Kievan Rus' or for it, and preserved in some thirty manuscripts of the fifteenth to eighteenth centuries. The oldest dated copy was finished on 18 July 1462. It diverges from the Greek everywhere: it cuts, it compresses, it expands, and in some forty places it says what the Greek does not.[^1]

3. A. N. Popov first noticed it (1866). I. I. Sreznevsky placed its language with the Rus' chronicles (1879). Alexander Berendts published the eight passages on John and Jesus in Old Russian with a German translation (1906), and argued that the Slavonic descends, through a Greek intermediary, from Josephus's first edition.[^2] Berendts and Konrad Grass then rendered Books I–IV whole, comparing the Greek at every phrase (Dorpat, 1924–1927). Their verdict on the translation as a whole: "dass man es jedenfalls nicht mit einer christlichen, späten Bearbeitung zu tun hat … dass der slavische Text, oder vielmehr seine Vorlage, die Priorität vor unserem griechischen beanspruchen darf."[^3]

4. Robert Eisler carried the case at full force: ΙΗΣΟΥΣ ΒΑΣΙΛΕΥΣ ΟΥ ΒΑΣΙΛΕΥΣΑΣ (1929–1930), The Messiah Jesus and John the Baptist (1931), and the reply to Bienert on the Testimonium (1938).[^4] H. St. J. Thackeray put twenty-two of the additions into English in the Loeb War (1928) and thanked Eisler for help with them.[^5]

5. The verdict went the other way. J. M. Creed (HTR 25, 1932) and Solomon Zeitlin ("The Hoax of the 'Slavonic Josephus'", JQR 39, 1948) set the terms; N. A. Meshcherskii's critical edition of the Old Russian (1958) gave them a philological ground; Steven Bowman read the work as a document of tenth- and eleventh-century Byzantium (1987), and the Cambridge History of Judaism as "an ideological struggle against the Khazars". The New Testament literature took the verdict as settled. Craig Evans: "to my knowledge no one today believes that they contain anything of value for Jesus research." Robert Van Voorst: Christian compositions, no alternative to the Testimonium. John Meier: "a wildly garbled condensation of various Gospel events."[^6]

6. The Slavist ground of the verdict is narrow. It is Meshcherskii's edition and commentary of 1958, which H. Leeming and L. Osinkina put into English beside Thackeray's Greek (Brill, 2003), and the critical edition of A. A. Pichkhadze and colleagues (Moscow, 2004). Meshcherskii's method is the method of the verdict: he lists forty-one additions and accounts for each in its place. A detail is "concretized" by the translator; a scene is borrowed "from another place of the same text"; a passage is "Christian phraseology". Pichkhadze has since reported Slavonic sources for three of the additions (2015).[^7] The method asks of each item what a translator might have had in view. It does not require a source, and it was never applied to the set.

7. Henry Leeming, the consensus edition's own translator, turned in 2005 from the additions to the small divergences, the phrasing that differs where the content is shared, and sorted them by type to test whether they reflect the two editions Josephus describes. By his own abstract, some types support that view.[^8] T. C. Schmidt has since argued, on stylometry and the manuscripts, that the Greek Testimonium is substantially Josephus's own (Oxford, 2025).[^9] Neither reopened the Slavonic. The question stood where Meshcherskii left it.

8. Method and scope. The essay begins from the framework's standing (Note on Genre): the Josephus reading carries no first burden (#1217), the verdict's treatment of the passages is the erasure the Historical Logos records at this station (#1757), and the Slavonic is a carrier under test (#1755, rung V). The verdict is put to its burden on its own ground. The corpus is the Slavonic War as Meshcherskii printed it from the Vilna Chronograph, in the language of the translation, read against the Greek War and against texts the verdict did not have or did not weigh: Hippolytus's Refutatio, the Dead Sea Scrolls, Philo on the Therapeutae and on the Logos, and the Ethiopic Enoch. Where it quotes, it quotes the seated texts.[^10] The second half of the essay (§§III–VII) then reads the eight Jesus-and-John passages, in the order Berendts printed them, as a structure. That reading does not depend on the first half.

II. The Essene Chapter

War 2.119–161. Meshcherskii 252–256 (Vilna fol. 570); Berendts–Grass 252–260; the Greek at the same sections. This section is a proposal stated so that it can fail, and the medieval account is held to the same standard. For each divergence the ledger below gives the best medieval alternative and the finding that would decide between them. An alternative that is only available is recorded as available, not as shown.

The innocent reading, in full. The Slavonic is a free translation by a medieval Christian who knew the Bible, the liturgy and the monastic day, and who read the Essenes as Christians had long read them. Christian interest in the Essenes is old, and it reached the medieval centuries. Eusebius copies Philo's two accounts of them into the Preparation for the Gospel (8.11–12) and Porphyry's excerpt of Josephus's own chapter (9.3, from De abstinentia 4.11–13). In the Church History he reads Philo's Therapeutae as the first Christian monks (2.17). Hippolytus gives the Essenes a long chapter in the Refutatio. A translator in that line would expand, harmonize and assimilate without needing a polemical reason: medieval translators did so as a matter of course. A single ascetic redaction could account for several of the divergences together. It would disdain marriage more strongly, clothe the novice, keep the night office, and call the heavenly host to witness. These would be one tendency, to be counted once. The particular sources are near to hand:

• a translator who kept vigil sang through the night in his own monastery and gave the Essenes the same office;
• the postulant of the Pachomian rule waited at the gate before he was received;
• Leviticus counts seven weeks to Shavuot, makes the seventh month holy and rests the land in the seventh year (Lev 23:15–16, 23:24, 25:4);
• the seraphim and cherubim of the initiate's oath are those of the Trisagion and the Cherubic Hymn, which every Slavonic monk sang.

Meshcherskii says this of the oath, his addition no. 16: "Это дополнение целиком и полностью проникнуто христианской фразеологией, оно связано с рядом древнехристианских произведений, в частности с произведением Ипполита Римского «Слово об Антихристе», которое было чрезвычайно популярно в древнерусской письменности."[^11] Each explanation is available. None is demonstrated: no source is shown for any of them, and the consensus's case at this chapter consists of their availability. That is the reading the consensus holds.

The Essenes between Eusebius and the scrolls. The interest the innocent reading cites is ancient: Hippolytus, Porphyry and Eusebius in the third and fourth centuries, Epiphanius in the fourth. It reached the medieval centuries in two forms, and both can be read.

• The heresy list. Epiphanius made the Essenes one of four Samaritan sects, who "ἐνέμειναν … τῇ πρώτῃ ἀγωγῇ μηδὲν ὑπερβαλλόμενοι", kept to the first way of life and added nothing (Panarion 10.1.2). John of Damascus reduces them to one sentence: the Essenes oppose neither of the other Samaritan sects, and keep their feasts with whomever they happen to be among (De haeresibus 12).[^36] A reader of the Byzantine handbooks knew the Essenes as a Samaritan sect with an indifferent calendar.
• The monk's chronicle. George the Monk, writing in the ninth century, excerpts Josephus's Essene chapter at length (Chronicon, de Boor 329–331).[^37] He sets it after the Rechabites of Jeremiah 35, as their heirs, and turns it into a monastic rule. The Essenes practise "ἀκτημοσύνη", owning nothing, and "σκληραγωγία", hard living. They work "μετὰ νήψεως καὶ θεολογίας", with watchfulness and theology. They offer "προσευχὰς δὲ καὶ ψαλμῳδίας πρὸς τὸν θεὸν ἐκ μέσης νυκτὸς μέχρις αὐγῆς", prayers and psalmody to God from midnight until dawn, and keep "τοὺς βραχυτάτους ὕπνους", the briefest sleep. George's chronicle was translated into Slavonic, and Rus' read its history of the world in that translation.

So the draft 2.0 sentence that no one cared about the Essenes was wrong twice. The interest was ancient, and a medieval monk wrote it into the most-read Byzantine chronicle. What survives the correction is narrower and can be tested. The medieval centuries had the Essenes as a catalogue name and as a monastic rule. The ledger below can ask, divergence by divergence, which of the Slavonic's readings that rule explains. George keeps the Greek where the Slavonic agrees with Hippolytus, and he has the night office where the Slavonic has it.

The count.

1. In the Essene chapter the Old Russian diverges from the Greek at nine points:

- (a) the sects named as a lawful order, not as philosophies (2.119);

- (b) the Essenes not called Jews by birth (2.119);

- (c) the name set before the description (2.119);

- (d) marriage despised together with "womanly lust", named at 2.120 where the Greek names the women's wantonness at 2.121, which the Slavonic also keeps (2.120–121);

- (e) night rising to sing (2.128);

- (f) the novice lodged in the house before the gate (2.137);

- (g) the novice's linen (2.137);

- (h) the oath sworn on the living God, his right hand, his Spirit, the seraphim and cherubim and all the heavenly power (2.139);

- (i) the seventh day, week, month and year (2.147).

2. Meshcherskii's list of additions counts one of the nine, the oath (h). The others fall below the threshold of "addition" and are nowhere weighed. Leeming's typology of small divergences (2005) is the instrument built for them; this section applies the question to one chapter.

The agreements with Hippolytus.

3. Hippolytus of Rome, writing early in the third century, gives the Essenes in Refutatio omnium haeresium IX 18–28, from Josephus or from a source he shares with Josephus.[^12] Berendts and Grass set his Greek beside the Slavonic at every phrase. At five points the Slavonic and Hippolytus agree against the Greek of the War and its other witnesses:

- (a) the order: Slavonic "Трои бо чинъ законный есть въ иудѣехъ", a threefold lawful order among the Jews. Greek: "Τρία γὰρ παρὰ Ἰουδαίοις εἴδη φιλοσοφεῖται", three forms of philosophy. Berendts and Grass: "Wie beim Sl. ist auch bei Hippolyt, Refut. IX 18 … von Philosophie nicht die Rede, sondern von τοῦ Θεοῦ νόμος."

- (b) "Ἰουδαῖοι μὲν γένος ὄντες", Jews by birth, in the Greek; absent in the Slavonic and, "bezeichnenderweise", in Hippolytus.

- (c) the order of name and description: the Greek describes the third sect and then names it; "Wie Sl. stellt Hippolyt die Charakterisierung dem Namen der Essener nach."

- (f) the novice: the Greek has him live the same way of life outside, ἔξω, for a year. The Slavonic has him live "въ предвратнѣи храминѣ", in the house before the gate, and Hippolytus "ἔξω τῆς ἑαυτῶν συνόδου οὖσιν ἐν ἑτέρῳ οἴκῳ", outside their assembly in another house. Berendts and Grass: "eine der auffallendsten Übereinstimmungen zwischen beiden."

- (g) the novice's cloth: the Greek gives him "τὸ προειρημένον περίζωμα", the loincloth mentioned before; Hippolytus alone among the Greek witnesses names it "τὸ λινοῦν", the linen, and the Slavonic gives "ленти", the λέντιον, the linen cloth.[^13]

4. Marriage (d) is not among them. The Slavonic "брака презирають" keeps the Greek's disdain (γάμου ὑπεροψία). Hippolytus has them forbid it, "γάμον τε ἀπαγορεύουσι", and Berendts and Grass enter him there as a difference ("Aber Hipp."). Draft 2.0 counted it as an agreement; it is withdrawn.

5. The five do not weigh alike. An omission (b) and a word order (c) are the commonest acts of any translator, and they discriminate little between routes. A lexical choice that coincides (a, g) discriminates more. The Slavonic turns δαίμων into God throughout (Berendts and Grass), so (a) also fits that habit, wherever the habit was formed; (g) could be read back from the linen of 2.129. An added architectural detail that coincides (f) discriminates most. Its medieval rival, the monastic gatehouse, is available and has not been shown.

6. Books IV–X of the Refutatio, the books with the Essenes, survive in a single fourteenth-century Greek manuscript, brought from Mount Athos to Paris by Minoides Mynas in 1842 and printed by E. Miller in 1851 as Origen's Philosophumena.[^29] The manuscript shows the work was copied in Byzantium in the fourteenth century. It does not show that the Slavonic translator read it, and its recovery in 1842 does not show that he could not have. The De Antichristo that Meshcherskii names, the Hippolytus that Rus' read widely, says nothing of the Essenes.

7. The question the agreements set is the transmission relation. Four relations fit them:

- a Greek exemplar of the War whose Essene chapter differed from ours in the line Hippolytus read;

- contamination of the translator's War from a copy of the Refutatio;

- an excerpt of Hippolytus's Essenes circulating separately and merged in;

- an antecedent shared by Hippolytus and the Slavonic's exemplar, such as Josephus's source, or a revised Josephus of the kind A. Baumgarten proposed.

They predict different things. Contamination and excerpt predict that matter peculiar to Hippolytus enters the Slavonic beyond the five points. A differing exemplar predicts agreements confined to the Josephan matter Hippolytus shares. The test is to collate the Slavonic against Refutatio IX 18–28 whole for Hippolytus's own matter. Its force is asymmetric. Matter peculiar to Hippolytus found in the Slavonic would support contact with Hippolytus, by contamination or excerpt. Its absence would not exclude a borrower who took only what Hippolytus shares with Josephus, and a selective borrower predicts the same pattern as a differing exemplar. A clean collation narrows the routes, and only a positive find decides one. This draft has not run that collation.

The calendar of sevens.

8. Greek 2.147: the Essenes keep from work "ταῖς ἑβδομάσιν", on the sevenths, the Sabbaths, more strictly than all Jews. Slavonic: "В седмый дьнь, и 7 ую недѣлю, и 7 мѣсяць, и 7 лѣто велми хранять": the seventh day, the seventh week, the seventh month and the seventh year they keep greatly.[^14]

9. The Temple Scroll counts "שבעה שבועות שבע שבתות תמימות … תספורו חמשים יום", seven weeks, seven full sabbaths, fifty days, and repeats the count from festival to festival through the year (11Q19 18:11–13, 19:12–13).[^15] Philo's Therapeutae assemble "δι’ ἑπτὰ ἑβδομάδων, οὐ μόνον τὴν ἁπλῆν ἑβδομάδα ἀλλὰ καὶ τὴν δύναμιν τεθηπότες", every seven weeks, revering not the simple seven only but its power, on the eve of the feast of fifty (De vita contemplativa 65).[^16] The seventh week as a unit of observance belongs to a sectarian calendar.

10. The innocent reading derives the line from Leviticus. Leviticus has seven weeks counted once, to one feast, and a seventh year of rest for the land. The line keeps its sevens "въ суботы", on the Sabbaths, and gives the week of weeks the standing of the Sabbath. A second Christian channel is nearer than Leviticus: Philo's Therapeutae, whom Eusebius made Christian. Whether Church History 2.17 carries the seven weeks is the test of that channel. The Greek does not. Eusebius reports the Therapeutae's "τὰς τῆς μεγάλης ἑορτῆς παννυχίδας", the all-night vigils of the great feast, and their hymns (2.17.22), and he leaves out Philo's "δι’ ἑπτὰ ἑβδομάδων".[^34] The Eusebian channel carries the night and not the count. Whether a Slavonic Eusebius read in Rus' adds it remains to be checked.

The night.

11. Greek 2.128: before sunrise they say nothing profane, but offer ancestral prayers toward it. Slavonic: "Мало же почивают нощью и въстають на пѣние, славяще и моляще бога": they sleep little at night and rise to sing, glorifying and beseeching God; then the dawn prayers follow as in the Greek.[^17]

12. The Community Rule: "והרבים ישקודו ביחד את שלישית כול לילות השנה לקרוא בספר ולדרוש משפט ולברך ביחד", the many shall keep watch together a third of every night of the year, to read the book, to study judgment and to bless together (1QS 6:7–8). The Therapeutae "τὴν ἱερὰν ἄγουσι παννυχίδα", keep the sacred all-night vigil, and sing hymns to God in many metres (De vita contemplativa 83–84). Eusebius carries those vigils and hymns into Christian reading (Historia ecclesiastica 2.17.22), so the night has a channel a monk could know.

13. The monastic vigil fits this divergence as well as the Scrolls do, and it was before the translator's eyes in a text. George the Monk gives the Essenes psalmody "ἐκ μέσης νυκτὸς μέχρις αὐγῆς" and "τοὺς βραχυτάτους ὕπνους"; the Slavonic gives them "Мало же почивают нощью и въстають на пѣние", they sleep little at night and rise to sing. The short sleep and the night song are George's, and the Greek War has neither. George shows the detail was available to a medieval reader. He does not show the translator took it from him: the Slavonic shares George's sense, not his words, and George's Essenes are themselves Josephus's, read through a monastery. Availability is not use, and the divergence weighs little either way.

The oath.

14. Greek 2.139: before touching the common food the initiate swears fearful oaths, first to be pious toward the divine. Slavonic: "призываа бога живаго, и десницю его всемощную, и духъ божии невмѣстимыи, серафим и херувим, вездѣ призирающе, и всю небесьную силу, поставляя на послушьство": calling on the living God, his almighty right hand, the Spirit of God that nothing contains, the seraphim and cherubim who see everywhere, and all the heavenly power, setting them as witness.[^18]

15. The Parables of Enoch summon "ኵሉ፡ ኀይለ፡ ሰማያት … ወኀይለ፡ እግዚአብሔር፡ ኪሩቤል፡ ወሱራፌል፡ ወአፍኒን", all the host of heaven, and the host of God, the cherubim and seraphim and ophanim (1 En 61:10), the company that sleeps not and guards the throne (71:7). The Songs of the Sabbath Sacrifice have the cherubim blessing "תבנית כסא מרכבה", the image of the chariot-throne (4Q405 20ii–22:8).[^19] The Greek War has the initiates keep secret "τὰ τῶν ἀγγέλων ὀνόματα", the names of the angels (2.142), and the Slavonic keeps that too: "zu bewahren aber die Bücher wie die Engelnamen".

16. The liturgy would supply the seraphim and cherubim, and so would 1 Enoch and the Sabbath Songs. God, his right hand and his Spirit is read as a Trinitarian triad, but each term is the Hebrew Bible's: the right hand of the Lord (Exod 15:6; Ps 118:16), the Spirit of God (Gen 1:2). Meshcherskii names a connection to Hippolytus's De Antichristo and quotes no parallel.[^11] His explanation is available for the oath and not demonstrated, and this draft weighs the oath low to moderate for that reason.

The ledger.

• | --
• | --
• | --
• | --
• | --
• | --
• |

The conjunction.

17. The chapter has two kinds of divergence. Points (d), (e), (g) and (h) can be read together as one ascetic tendency, and the medieval account reads them so; read that way they count once. No ascetic redaction has been shown. The one medieval monastic text on the Essenes that has been found, George the Monk's, has the night office and none of the others: it softens marriage, keeps the loincloth, and swears dread oaths without angels. Points (a), (b), (c) and (f) are agreements with a third-century Greek witness on matter that an ascetic tendency does not predict. Of these, (f) carries the most weight on its own. Point (i) stands between: a sectarian calendar, with a Christian channel tested and found not to carry it.

18. The proposal is therefore narrower than draft 2.0 put it, and better founded. The Slavonic's Essene chapter shares readings with the text Hippolytus read at Rome eight centuries before the translation, at points no ascetic tendency predicts. Whether those readings came by a differing exemplar of the War or by later contact with the Refutatio is investigated through the collation of §II.7. The ledger states what the collation can decide and what it cannot. Berendts said in 1906 that the Slavonic's Greek exemplar was older than ours, and Berendts and Grass said so of Books I–IV in 1924. The verdict that followed counted additions, and these divergences are not additions. It did not answer them at these points.

Beyond the Essenes.

19. The same question holds across the work. Meshcherskii's forty-one additions mostly have nothing to do with Christian matter, and Berendts and Grass's nineteen Überschüsse in Books I–IV include four on John and the wonder-worker.[^20] The rest are:

- Hyrcanus's priestly rule (Book I);

- the Roman custom of writing a king's grant on the Capitol before it holds;

- Herod's dream of the four ears of grain;

- the debate of the priests on whether Herod is the Messiah;

- two anti-Roman passages;

- providence shown in Abraham;

- Judas and Matthias rousing the young men with Eleazar, the seven brothers and their mother, eternal torment and eternal reward;

- the Galilean Judas's reproach that the free seed of Abraham pays tribute to Rome, "владыкы съмьртны имуща, понеже лишишася безсмьртнаго", holding mortal masters because they have lost the Immortal;

- the abomination of desolation in the holy place;

- the Zealots over the bodies of Ananus and Jesus;

- Vitellius's three-horned iron.

20. These are Jewish, anti-Roman and messianic, and §II-B counts them all with the rest (Appendix IV). The anti-Khazar reading has to explain why a Christian polemic against a Jewish khaganate would make the Jews the righteous resisters of Rome and give them a hope against "mortal masters". The Byzantine reading has to explain why a work made for an empire that called itself Rome would add to Josephus's Romans the grievances of their Jewish enemies.

21. The same idiom crosses from the Book I additions into the John passage. Judas the Galilean reproaches those who hold "mortal masters". The first John passage has the man in wondrous garb come to call the Jews to freedom: "богъ мя посла, да покажю вам путь законныи, им же избудѣте от многых властелинъ. А не будеть владѣя съмьртныи, развѣ вышняго, пославшего мя", God has sent me to show you the lawful way, by which you will be rid of many masters; and no mortal shall rule, save the Highest who sent me.[^26] The John passage and the Judas passage speak one political idiom. A translator inventing each in its place had to write both in it.

22. The curtain of Book V stands with them. The Slavonic keeps Josephus's Greek word for the Temple veil untranslated, "катапетазма", and says the curtain "прѣд симь родом бысть цела, зане благочьстиви быша людие", was whole before this generation, because the people were pious, and "раздра бо ся внезапу от верху и до долу", it was torn suddenly from top to bottom.[^21] The tearing is the image of Mark 15:38. What surrounds it is not the Gospels'. The curtain tears when they gave up "добродѣица мужа и не мужа дѣлом", the benefactor, a man and by his deeds not a man, to be killed for a bribe. The narrator then reports that some said he rose and others that his friends stole him, and says "Не вѣмь же, которииз правѣише глаголють", I do not know which of them speak more truly. The consensus explains the passage by its one shared image. The account around the image is the passage's own (§V.2).

II-B. The Scope of the Changes: Two Accounts

1. The question. Draft 2.1 tested one chapter. This section asks whether one account organizes the changes across the whole work. Two accounts are on offer.

- The account of the translator. A Christian of Rus', or of the Byzantine world for Rus', translated the War freely. He added what a Christian reader wanted, coloured what a Rus' reader knew, and cut what neither needed. This is Meshcherskii's account, and he applies it to each of his forty-one additions in turn.

- The account of the author. Josephus says he wrote the War twice, both times as Rome's captive. He first composed it "τοῖς ἄνω βαρβάροις τῇ πατρίῳ", in the ancestral tongue for the peoples of the interior, and then turned it into Greek "τοῖς κατὰ τὴν Ῥωμαίων ἡγεμονίαν", for those under Roman rule (War 1.3). He had help with the Greek (Against Apion 1.50). The Greek edition went out with the conqueror's hand on it: "χαράξας τῇ ἑαυτοῦ χειρὶ τὰ βιβλία δημοσιῶσαι προσέταξεν", Titus signed the books with his own hand and ordered them published (Life 363). In his own preface Josephus names what bends a history of this war: "ἢ κολακείᾳ τῇ πρὸς Ῥωμαίους ἢ μίσει τῷ πρὸς Ἰουδαίους", flattery of the Romans or hatred of the Jews (War 1.2). On this account the Slavonic's Greek exemplar carries matter the Roman edition trimmed. It may be a Greek rendering of the first composition, as Berendts held, or a Greek edition nearer to it than ours.

The two accounts start from different places. The author's account begins from an attested statement: Josephus says he wrote an earlier version for other readers (1.3, 1.6). That the Slavonic preserves descendants of it is this paper's inference, and answerable at each link. The translator's account begins from no ancient statement; it is a modern inference from the content of the passages. Each is a positive explanation, and each carries the burden at each item it claims. The question that governs is which account best explains the surviving differences, taken together. The census answers it: the earlier version Josephus describes is by far the more plausible and coherent explanation (§3).

2. What each account predicts. The translator's account predicts changes that cluster on Christian matter and on things a Rus' reader knew: monastic life, war as Rus' fought it, the moral of a story. It also predicts that some of them will have Slavonic or Byzantine sources. The author's account predicts changes that cluster where a Roman reader under Titus would be touched:

- Jewish law and its martyrs;

- resistance to Rome;

- messianic and prophetic hope;

- the Flavian claim to the oracle;

- the author's own address to Roman subjects.

It also predicts glosses explaining Roman custom to readers who are not Roman. And it predicts that the Flavian frame itself would stand in both editions, since both were written in captivity.

The census of the work.

3. The census takes every content divergence the three collators of the Slavonic War list, item by item. It counts Meshcherskii's forty-one numbered additions and one unnumbered, his omissions and transpositions; Berendts and Grass's nineteen major Überschüsse, Grass's further additions by Niese section and his aggadic departures; and Thackeray's twenty-two. Each item is read in the Old Russian of the seated text, with Berendts's German translation and footnotes for Grass's items, against the seated Greek. Each gets the best explanation the author's account can give (A) and the best the translator's account can give (M), and a verdict: A or M where one explains it better, T where it is a feature of translation that both expect, U where neither does better on present evidence. Each item keeps two findings apart: what its correspondence establishes (shared matter, a kind of change, integration into its context, a possible Greek antecedent, a demonstrated dependence), and which origin that favours, why and how strongly. Ancient content can be carried by later composition, and a change in Josephus's repertoire can be made by another writer, so shared matter or kind alone favours no origin. An A is graded by the origin it favours: A1, the translator's Greek exemplar, before Rus'; A2, an ancient textual form; A3, Josephus; A4, his first version for the readers of 1.6. Each A states its strength. The verdict is comparative: where the rival account could produce an item but does not explain it, and the author's account does, the item is A, at weak strength. A Rus' word or an Orthodox phrase in the rendering is what both accounts expect, and counts for M only where the content depends on it. A source counts for M when it is shown and its direction established; a source only available is recorded as available. The census in full, item by item, is Appendix IV.[^39]

• | --
• | --
• | --
• | --
• | --
• |

The thirty-two A items favour: the Greek exemplar before Rus', 3 (the corrupted Greek word under the "snake-footed Syrian", strong; Phoias and Cerealis, which Meshcherskii himself would put in a lost Greek manuscript; the Maccabean martyrs in 4 Maccabees' words, by any of their three Greek routes); an ancient form, 13; Josephus, 8; the first version, 8. The first-version items rest on one premise, two editions for two readerships with the Roman one revised, and count as one line of evidence. Appendix IV §7.1a gives, for every A, what its correspondence establishes and why it favours the origin it does, with its strength.

The conclusion. Between the consensus account and the earlier version Josephus himself describes, the earlier version is by far the more plausible and coherent explanation of the surviving differences. The conclusion does not rest on the count; the verdicts share grounds, and 32 of 34 decided items is not a sum of independent confirmations. It rests on the pattern.

• The force is cumulative. Readerships the author names; Jewish history, law and haggadah where the Greek has none; anti-Roman matter the Roman edition could not carry; agreements with the Antiquities and with Second Temple witnesses; candour about the author withdrawn in the edition Titus signed; the Roman edition's providential closures absent; and, in the other direction, Roman-facing set-pieces present only in the Greek. One account explains these together. Observing that a medieval writer could have produced each feature separately supplies no equally coherent explanation of their conjunction.
• The consensus explains its own ground and nothing beyond it. Built on the Christian passages, it wins two items, by a reported source and a Gospel name. Extended to the fifty-one additions with no Christian subject, it explains none of them better than the author's account.
• The undecided items sit inside the conclusion. Twenty-five items are undecided; ten lean to the author's account because the translator's stated ground fails at them, and none leans the other way. Local underdetermination does not entail global explanatory equivalence: an account can explain the whole far better while particular items stay open. The question is which account best explains the differences, and not whether the earlier-version account has eliminated every imaginable medieval intervention, a standard the consensus is never held to.

The omissions run the same way: eleven groups of the Greek's surplus are matter for Roman or Greek readers, Flavian matter, or the Josephus of the Roman edition; four show abridgement by a trace it left; the rest are undecided, and two are mechanical losses.

The Maccabees.

4. The rabbis' appeal before the golden eagle (War 1.650; Meshcherskii no. 10; Thackeray 1928, (7)) is the work's clearest Maccabean change. The Greek has the teachers say, in reported speech, that it is noble to die for the ancestral law and that those who die so have immortality. The Slavonic gives the speech direct and fills it with the martyrs of 2 and 4 Maccabees: "Eleazar first, and the seven brethren, the Maccabees, and their mother"; Antiochus defeated "by those seven striplings and by the aged teacher and by the grey-haired woman"; "that we may not appear weaker than the woman"; and, at death, the soul passes "to (our) forefathers, where Abraham (is) and those (descended) from him".

5. These are the figures and the words of 4 Maccabees. Eleazar is "ὦ … γέρων … πρεσβῦτα" (4 Macc 7:10). The mother is "ἀνδρῶν πρὸς ὑπομονὴν ἀνδρειοτέρα", braver than men in endurance (15:30), and "πρεσβῦτι καὶ γυνή, διὰ καρτερίαν καὶ τύραννον ἐνίκησας", old woman, by endurance you conquered even the tyrant (16:14). The dead are received by the fathers: "οὕτως παθόντας ἡμᾶς Ἀβραὰμ καὶ Ἰσαὰκ καὶ Ἰακὼβ ὑποδέξονται" (13:17). The wreath is 4 Maccabees' too: "θεοσέβεια δέ ἐνίκα, τοὺς ἑαυτῆς ἀθλητὰς στεφανοῦσα" (17:15).[^30]

6. Antiquity gave 4 Maccabees to Josephus. Eusebius, listing his works: "πεπόνηται δὲ καὶ ἄλλο οὐκ ἀγεννὲς σπούδασμα τῷ ἀνδρὶ περὶ αὐτοκράτορος λογισμοῦ, ὅ τινες Μακκαβαϊκὸν ἐπέγραψαν", the man also laboured over another work, not ignoble, on the sovereignty of reason, which some have titled the Maccabean (Historia ecclesiastica 3.10.6). It travelled among the manuscripts of Josephus.[^31]

7. Both accounts predict the change. The translator's account: the Maccabean martyrs were saints of the Church, kept on the first of August, and a Christian translator could reach for them. The author's account: the speech that the Greek edition gives in a sentence, the Slavonic gives with the martyrs of the book that antiquity filed under Josephus's own name, in that book's words. The wording points past 2 Maccabees, where Eleazar, the seven and the mother also stand in that order (2 Macc 6–7), to 4 Maccabees: the mother who conquers the tyrant, braver than men, and the fathers who receive the dead are 4 Maccabees' and not 2 Maccabees'.

- The attribution cuts both ways. A book that travelled in Josephus's manuscripts under his name gave the author his own words to reach for. It also gave a later reader of a codex that held both works a reason to amplify the one from the other.

- The question the change sets is therefore what "Josephus" is doing here. It may be a compositional source: the author writing from his own book. It may be an authorizing attribution: a reviser drawing on a book believed to be his. Or it may be a container: a codex through which material migrated between works filed under one name.

- The change is significant under each answer, and it is the point where this essay meets The Crossing of Names, which separates the author-name from the composing source. What would discriminate among the three is the Slavonic's Greek exemplar. If it carried 4 Maccabees, the container account needs no author.

What the census shows.

8. The control. One medieval Christian practice is visible in this manuscript. The Vilna Chronograph sets into the run of Josephus the Gospel of Matthew, Luke 23, Acts, and long runs of Malalas and Hamartolos: forty-eight notes in Meshcherskii's apparatus identify them by edition and page, and the margin marks them "Хронограф" against "Иосифъ". That is the compiler's layer: verbatim, sourced, labelled, and outside Josephus's sentences. It is a comparison class. An earlier, unlabelled interpolation could stand beside it, and an addition present in both redactions is placed before they diverged, which is not yet antiquity. What joins the additions to the compiler's layer, and what separates them, is the question. They share none of its marks. They are in the translation's own language, inside its sentences; they stand in the stretches the margin marks "Иосиф"; and they refer to one another inside that stream, as the informer Lui passes from the priests' debate into the Magi story. They belong to the translation's stream. How much older than the translation each is, the census grades item by item.

8a. Josephus's own revisions. Some of the commonest differences between the Slavonic and the Greek War are kinds of change Josephus makes himself between the War and the Antiquities, at the same places.

- Speech and report. The War gives Herod's words to Antipater direct: "φθείρου, δυσσεβεστάτη κεφαλή, μηδέ μου ψαύσῃς" (1.618). The Antiquities gives the same words as report: "ἀδελφοκτονίαν τε ἐπικαλῶν καὶ βούλευσιν ὀλέθρου … δικαστὴν ἔσεσθαι Οὔαρον τῇ αὔριον" (17.91). Grass: "in derselben Weise …, wie das sonst im B. J. gr. im Verhältnis zu Sl. so häufig konstatiert wurde."

- A prophecy rewritten. Archelaus dreams of ears of grain eaten by oxen. The War has nine ears, the reading that he will reign their number, and exile in the ninth year (2.111–113); the Antiquities has ten ears, ten years, and the accusation in the tenth (17.342–348). Each work matches its count to its own date, and Josephus changed the number between them. The Slavonic agrees with the War: "Подобаеть ти владѣти по класному числу", it is fitting that you rule according to the number of the ears. This locus shows Josephus's practice; it does not show a Slavonic state of it.

- A reflection on providence at Antipater's arrest. The Slavonic stops for one, from Abraham, "the forefather of our people", punished for wronging his brother in dividing their lands: a haggadah nowhere in Genesis. The Antiquities stops at the same point for its own, on the man who acted "ὡς ἐν ἐρημίᾳ τοῦ θείου", as if God were absent, and now calls God to witness (17.129–130).

These make the variants Josephan in kind, at loci where Josephus is shown at work. They do not by themselves fix the order of the states. Under the author's account the Slavonic stands first and the line runs one way, speech to report. Under the translator's account the Greek stands first and the translator runs it back, report to speech, at the places Josephus worked. What orders the states has to be a correspondence that fits one direction and not the other. Tiron's cry at Herod's court is the census's clearest. The local sequences, complete:

- Greek: "καὶ τὸ μὲν πρῶτον ἐβόα περιιὼν πεπατῆσθαι τὸ δίκαιον, ἀπολωλέναι τὴν ἀλήθειαν, συγκεχύσθαι τὴν φύσιν, ἀνομίας γέμειν τὸν βίον, καὶ πάνθʼ ὅσα μὴ φειδομένῳ τοῦ ζῆν ὑπηγόρευε τὸ πάθος. τέλος δὲ καὶ τῷ βασιλεῖ τολμήσας προσελθεῖν … ἔφη" (1.544–545).

- Old Russian: "И ходя, акы похабъ, вопиаше: Попрася правда! И отидѣ истинна, и разлиася естьство! Житиа исполнися безакониа! И душа цесарева отидѣ, и умъ его измѣнися: Глаголъ его яко вѣтръ, словеса его яко роса утрьняя! И ко тому вся приложивъ, яже можеть рещи человѣкъ бѣсуяся и не щадя тѣла своего, ни помня живота."

- What corresponds. The frame, περιιὼν ἐβόα, "ходя … вопиаше", going about he cried. The four members in order, verb for verb: πεπατῆσθαι, "попрася", trampled; ἀπολωλέναι, "отидѣ", gone; συγκεχύσθαι, "разлиася", poured out, the literal sense of συγχέω; γέμειν, "исполнися", full. The close: "καὶ πάνθʼ ὅσα μὴ φειδομένῳ τοῦ ζῆν ὑπηγόρευε τὸ πάθος", "и ко тому вся приложивъ … не щадя тѣла своего, ни помня живота", and all else that a man not sparing his life could say.

- What is extra, and where. Four lines stand inside the shared sequence, between its fourth member and its shared close: the king's soul gone, his mind changed, his word like wind, his speeches like morning dew. They keep the members' form, and they turn the cry from the world to the king, the turn the Greek makes unprepared at 545, "τέλος δὲ καὶ τῷ βασιλεῖ".

- The residue. In the Greek, "πάνθʼ ὅσα", everything else, follows four members directly, a summary standing for a list that ran on. Read as reduction, the Greek keeps the first four members, drops the four that bear on the king, and lets the summary stand for them, which leaves its turn to the king unprepared. Read as expansion, the translator set four lines into the middle of a closed series, in the series' own form, before a summary that then had less to summarize. Berendts: the Greek is "ganz besonders ungeschickt herausgekommen. Die Originalität kann nur auf der Seite der Vorlage des Sl. zu suchen sein."

The position of the extra lines, inside the shared list and before the shared close, with the Greek's unprepared turn, favours reduction in the Greek tradition over insertion by the translator: moderate. It does not by itself say whose reduction.

8b. Rome bought and selling. Three additions say what no book signed by Titus could say about Rome:

- the excursus on the Italians "die Lateiner genannt werden", who take gifts and break oaths for them and count slander no murder (1.603; Meshcherskii no. 8). The name is medieval; the content is not. Meshcherskii grants that Rus' polemic against the Latins attacks their rites, "но нигде не говорится об их корыстолюбии или жадности", never their greed;

- Antipater's gifts to "the Roman rulers", who write his praise to Herod (1.600–602). Where the Greek speaks of bribery, it is the anonymous "τῶν ἐπὶ Ῥώμης φίλων" (1.603);

- Felix's prisoners. The Greek has Felix send the brigand chief Eleazar to Rome (2.253). The Slavonic has the emperor crucify the lesser men, take "unzählbares Vermögen" from the leaders, and let them go.

These are what the author's account predicts the first version carried and the Roman edition withdrew. The translator's account has no motive for the second and third, and its explanation of the first fails on its own editor's evidence.

8c. Places the Greek names and does not fill. Where the Slavonic has more, the Greek sometimes names what it does not give: "ἐναργεῖς ὄνειροι", clear dreams, where the Slavonic tells Herod's one dream of the four ears (1.328); "πρὸς ἓν κήρυγμα", at one proclamation, where the Slavonic says what it was, to die or to kill the soldier who burned the Torah (2.230; Berendts: "Gr. lässt also den eigentlichen Inhalt des κήρυγμα weg"); "ὡς ζώσῃ προσλαλεῖν", he spoke to her as if alive, where the Slavonic has Herod run to Mariamne's bed calling her name (1.444). A compressing reviser leaves such places, and an expanding writer would fill the same places, so they show where the texts differ and not which way. The census scores them for the author's account at weak strength: the author's account explains why the Greek names what it does not give, and the translator's account only allows the filling. Direction is shown more strongly where a correspondence fits one way only, as in Tiron's cry (8a).

8d. The consensus's plainest case. Meshcherskii calls one clause a direct quotation of the Gospel that Josephus could not have spoken: "зане кров праведную възложихом на ся и на род нашь", because we have laid righteous blood upon ourselves and upon our race (War 6.110; Matt 27:25, "Τὸ αἷμα αὐτοῦ ἐφʼ ἡμᾶς καὶ ἐπὶ τὰ τέκνα ἡμῶν"). Two findings answer it, and they have different strengths.

- The exclusive attribution falls. Each part of the clause has a Jewish antecedent. Lamentations gives righteous blood shed in the city as the cause of its first fall: "τῶν ἐκχεόντων αἷμα δίκαιον ἐν μέσῳ αὐτῆς", who shed righteous blood in her midst (Lam 4:13). Jeremiah at his trial gives the formula: "αἷμα ἀθῷον δίδοτε ἐφʼ ὑμᾶς καὶ ἐπὶ τὴν πόλιν ταύτην", you lay innocent blood upon yourselves and upon this city (Jer 33:15 LXX). In this siege Josephus casts himself as Jeremiah (5.391–393). The clause need not be the Gospel's. A medieval Christian could reach for the same prophets, so this finding does not yet give it to Josephus.

- The clause belongs to the speech. The speech in the Greek is about Jews killed by Jews in the sanctuary. The prophets foretold the city's capture "ὅταν ὁμοφύλου τις ἄρξῃ φόνου", when someone begins the slaughter of his own people (6.109); the city and the sanctuary are full of the dead; God brings the Romans with a purifying fire (6.110). The clause stands between the prophecy and the fire and answers to each part:

- whose blood: the people's own dead in the sanctuary, after the Slavonic's "бои сродничьскыи", war among kin;

- whose responsibility: "we", the speaker inside the people, as Lamentations lays the blood on the city's own priests and prophets;

- which catastrophe: the capture the ancient prophets foretold;

- what follows: "Имь же", therefore, God brings the Romans to purify the sanctuary with fire, the Greek's own next sentence.

The Gospel reading has to bring in the blood of one man the speech does not name, and leave the clause's "therefore" without its ground.

The first finding is strong; the second, that the clause is prior to the translator, is moderate.

8e. Where the translator's account wins. It wins at two additions. At the Magi, Pichkhadze reports a Slavonic source, unread here. At "Manneus, nephew of Lazarus" (5.567), the raising of Lazarus is a Christian gloss on a name the Greek manuscripts had already Christianized. It wins at four omissions, each a cut that left a trace: the victory on the lake kept after the battle is gone (3.522–531), the stronghold of Tabor named and dropped (4.54–68), the procurators' misrule a first version for Jews would have kept (2.271–283), and the high priest's vestments cut with the translator's "и иная вся оставихом повѣдати", we have left the rest untold (5.233–238). These show that someone abridged. They do not show who added.

9. The author's account organizes the changes no medieval interest explains. It also predicts what tells against the simplest form of Berendts's thesis. The Flavian frame stands in the Slavonic: Josephus's prophecy to Vespasian, and the council at which Titus spares the Temple. A first edition written in captivity would carry that frame, as the Greek does. Where the Greek edition alone goes further, the Slavonic stops short. It does not close the oracle on Vespasian. It has neither the address to Roman subjects nor the author's vindication at the end.

10. The lots at Jotapata are where the two accounts part most sharply. The Greek says Josephus was left alive "εἴτε ὑπὸ τύχης χρὴ λέγειν, εἴτε ὑπὸ θεοῦ προνοίας", whether one should call it fortune or the providence of God (War 3.391). The Slavonic says "почте числа мудростию и тем прѣбляди вся": he counted the numbers by cunning and so deceived them all.

- Meshcherskii gives the line to the translator because the author "не мог о себе так сказать", could not say this of himself. Yet the translator who heads the wonder-worker passage "Иосиф о Христѣ пишет" had little reason to make his witness a cheat.

- The Slavonic changes the mechanism before it changes the verdict. The Greek proposes lots: "κλήρῳ τὰς ἀλλήλων σφαγὰς ἐπιτρέψωμεν", let us leave our killing of one another to the lot (3.388), and "ὁ λαχὼν", the one the lot falls on (3.389). The Slavonic proposes a count: "числом да убиения будем, на него же конець числа будеть, да тот будеть убиенъ от втораго", let us be killed by number; on whom the end of the number falls, let him be killed by the second. A lot is chance. A count can be steered. The Slavonic's "почте числа мудростию", he counted the numbers by cunning, follows from its own device. It keeps the Greek's trust in God: "поручив спасение свое богу-строителю", having entrusted his safety to God the provider, is 3.387's "πιστεύων τῷ κηδεμόνι θεῷ". It removes only the Greek's account of the outcome, fortune or providence.

- The author's account reads the line as a warning. Josephus wrote the war twice. One version was in the ancestral language for the interior. The other was in Greek for Rome's subjects, under the stricter demands of a work Titus would sign. In the first he could say to one audience how he lived, by counting. In the second the same survival becomes fortune or providence: the providence that, a few sections on, sends him to Vespasian as its messenger (3.400). The first version tells its readers how to read the second. Where the Greek says providence, they are to read cunning, and that is how they are to read the Greek's other alterations. The Greek edition's own preface names the pressure: histories of this war are bent "κολακείᾳ τῇ πρὸς Ῥωμαίους", by flattery of the Romans (1.2). The archive has read the counting-out itself: The Fixed Point of Destruction (#1412) argues that a one-survivor stopping rule fixes its survivor before the first removal. When such a destruction is known through the testimony of one of its participants, the record is conditioned by the survival that made the testimony possible. Josephus at Yodfat is its paradigm case. The Slavonic's count is that rule stated by the survivor.

- Meshcherskii's objection assumes one audience. With two audiences, an author can say of himself to one what he could not say to the other. The line Meshcherskii could not give to the author is the one the author's account most needs.

- A warning that teaches how to read the other edition is a further claim than a difference between editions. What carries it is a second passage of the same kind. The oracle is one. The Greek gives the oracle's sense as the historian's verdict: "ἐδήλου δ’ ἄρα τὴν Οὐεσπασιανοῦ τὸ λόγιον ἡγεμονίαν", the oracle in fact signified the rule of Vespasian (6.313). The Slavonic gives three opinions and no verdict: "Ови бо мнѣша Ирода, ови же распятаго чюдотворца Исуса, ови же Уеспасиана". It keeps what surrounds the verdict, the oracle "в святых книгах обрѣтаемая", found in the holy books, as the Greek's "ἐν τοῖς ἱεροῖς εὑρημένος γράμμασιν" (6.312), and the self-serving judging of the signs (6.315). At Jotapata and at the oracle, the same thing is missing: the Greek's resolution of an event in Rome's providential favour.

- The translator's account has its reading of the oracle: a Christian set Jesus among the candidates, and Greek manuscripts carry such a marginal note (Meshcherskii). It does not account as well for the result. A Christian hand that wrote Jesus in leaves him one opinion of three and keeps Vespasian beside him. Weight: moderate. The census adds one of the same shape, the effect kept and its mechanism withdrawn: Simon's rout of the Idumaeans, where the Greek's sudden panic (4.528) is in the Slavonic a ruse of brushwood tied to the horses' tails to raise a dust, "мняще задних болѣ предних", so that they thought there were more behind than in front. The census scores it undecided, as one item, and the warning reading stays at moderate weight.

11. The burden at the joins. The translator's account does not argue its items one by one. It joins them as one hand and argues from the Christian passages. Meshcherskii states the principle while arguing that the Magi belong in the text: "Если исключать это «добавление», то не следует включать и остальные вставки древнерусского переводчика, столь же фантастические и не менее тенденциозные", if one excludes this addition, one ought not to include the translator's other insertions either, equally fantastic and no less tendentious. He joins them by four criteria, and each identifies something different.

- Style identifies the translator. Meshcherskii's stylistic proof that the Magi passage belongs to the History is its likeness to "Пилату же отпирающуся молениа их, но ници падше, пять дьнии и пять нощии неподвижими терпяху", which renders the Greek War's own "Πιλάτου δὲ ἀρνουμένου … πρηνεῖς καταπεσόντες ἐπὶ πέντε ἡμέρας καὶ νύκτας ἀκίνητοι" (2.171). His criterion covers what was rendered from Josephus and what is said to have been added, alike. It identifies a rendering and leaves origin open.

- The margin's "Иосиф" identifies the compiler's assignment of a stretch to the Josephus stream.

- The cross-reference, Herod's characterization at the end of Book I referring back to "the seeking of Christ and the killing of the infants", and the recurring informer Lui, in the priests' debate and in the Magi story, identify a composition: one narrative whose parts refer to and continue one another.

- The "idea" with which the translator "pierces his whole narrative", the city's fall as punishment for righteous blood (nos. 26, 36, 37), identifies a theme. It is Lamentations' and Josephus's own (War 6.109–110, §8d).

Tested at each level:

- As style, the joining settles nothing about origin, on Meshcherskii's own example. Each passage then stands on its own evidence, and there the census gives the fifty-one non-Christian additions no translator's verdict and the fifteen with Christian subject twelve for the author, one undecided and two for the translator. John without the Lamb (§V.2), the wonder-worker as Philo's Logos named, refused and written as a man (§III.7a), the inscription that states the charge and not the faith (§III.8), the veil as the written heaven (§III.9), the oracle left undecided (§10), and the righteous blood of the speech's own argument (§8d) go to the author.

- As composition, the joining has force. Lui ties the Magi to the priests' debate, which the census grades A2: Jewish messianic exegesis of Gen 49:10, which Grass reads as cut from the Roman edition for compromising the Jews. If the joins are compositional, the composition they join is the one whose non-Christian parts the census gives to an ancient form, and the presumption carries to its Christian parts.

- As theme, the idea Meshcherskii names is the prophets' and the historian's.

The account may keep the joining or drop it. Kept as composition, it carries the census's verdicts on the non-Christian parts to the Christian ones. Dropped to style, it leaves each passage on its own evidence, where the Christian-subject passages already score for the author or undecided at all but two. Either way the burden lies on the translator's account, at each passage: a source shown with the direction of its borrowing, or matter a first-century Jewish author could not have written, on some ground other than the Greek edition's silence. It may not treat the joining as composition where it convicts and as style where it acquits.

12. The consensus's inferences. The census shows where the consensus argues; it also shows how. Four of its inferences fail, and the verdict rests on them.

- It assigns by subject and then excludes by assignment. It identifies material as Christian, gives it to a Christian interpolator on that ground, and then uses the assignment to remove the material from the evidence about Josephus's text and its theology. The test that would decide, whether these passages belong with the changes no medieval interest explains, is never run (the abstract; §11).

- It takes a rendering for an author. It treats the translation's shared style as proof of a shared originating hand. Meshcherskii's own stylistic proof for the Magi rests on a sentence that renders the Greek War (2.171), so the criterion covers translated Josephus and supposed addition alike (§11).

- It reads the revised text as the measure of its revision. "Разумеется, сам автор не мог о себе так сказать", of course the author could not have said this of himself (Meshcherskii on Jotapata): the Greek edition's self-presentation is used to rule out a divergent account of the self, when the revision of that self-presentation is the matter under examination (§10).

- It holds the two accounts to different standards. A medieval source that was only available is accepted as the explanation, while the earlier-version account is asked for demonstrated transmission at every link. The census applies one standard to both (Appendix IV §0.3).

Each of these makes the conclusion a premise. Together they are why the consensus could stand without testing the set.

The route of single items stays as the census states it: the Essene agreements keep their four routes until the collation of §II.7 is run, and 4 Maccabees keeps its three functions.

III. The Eight Passages as Hidden Gospel

1. The received eight. Berendts (1906) printed eight passages on John, Jesus and his followers, and Thackeray listed them in that order (1919, Appendix, Note II).[^27] The eight are Berendts's selection. Berendts and Istrin left out of their texts the Magi story at the end of Book I, chapter 20, and Meshcherskii printed it as part of the translation (his addition no. 5). It shares its matter with Matthew 2, and Pichkhadze reports Slavonic sources for it. They are:

- 1. John, "the savage", calls the Jews to freedom, dips them in the Jordan, answers before Archelaus and is attacked by Simon the Essene;

- 2. John interprets Philip's dream of the eagle;

- 3. John rebukes Herod over Herodias;

- 4. the wonder-worker, his works by word, Pilate and the bribe;

- 5. his followers, who say he is alive though dead;

- 6. the inscription on the Temple concerning Jesus;

- 7. the torn veil and the reports of resurrection;

- 8. the oracle of the world-ruler, read as Herod, the crucified wonder-worker or Vespasian.

Their loci, Greek places and Old Russian are collated in Appendix I.

2. What version 1.0 did. Version 1.0 read eight stations: prophecy, purification, incarnation-as-teaching, inscription, destruction and renewal, persistence, judgment, eternity. It gave each station a quoted paraphrase. Those stations were the authors' composition, not the recension's order, and three of the paraphrases are found in no witness: "write down his words", "cursed he who writes falsely" and "nothing written in truth perishes". None of them is in Meshcherskii's Old Russian, in Berendts and Grass, or in Thackeray's English of the additions. The purification station quoted "body cleansed after the soul by justice", which is the Greek Antiquities on John (18.117), not the Slavonic. These are withdrawn as quotations. The stations stand where they belong, as the authors' construction, in Appendix II.

3. What the received eight carry. Read in the recension's order, the eight carry their own motif of the word and the written, and they need no supplied lines:

- the wonder-worker does all things by word: "вся, елико творяще, силою невидимою нѣкоею, словом и повелѣниемь творяше", whatever he did, he did by some invisible power, by word and command (4);

- his followers proclaim of their master "яко живъ есть, аще умре", that he is alive, though he died (5);

- the Temple carries a written notice of him, "тѣми грамотами показаа Исуса, цесаря нецесарствовавша, распятаго от июдѣи, зане проповѣдаше разорение града и опустение цьркъви", by those letters showing Jesus, a king who did not reign, crucified by the Jews, because he preached the destruction of the city and the desolation of the Temple (6);

- the veil is a written heaven, "Написано же бѣ на тои катапетазмѣ всяко небесное видѣние и мудрость", there was written on that curtain every heavenly vision and wisdom, and it is torn from top to bottom (7);

- the end is an oracle found in writing and read three ways, "Ови бо мнѣша Ирода, ови же распятаго чюдотворца Исуса, ови же Уеспасиана", some thought Herod, some the crucified wonder-worker Jesus, some Vespasian (8).

4. The arc the received order makes runs from a voice calling to freedom from mortal masters (1), through the word that acts (4) and the word proclaimed after death (5), to the word inscribed on the Temple (6), the written heaven torn (7) and the written oracle whose referent is contested (8). The Word, in this recension, ends as a text with three readers. That arc is found in the passages, in their order. The reading that calls it a Gospel of the Word is the essay's, and it holds whatever §II decides about the Essene chapter.

5. Eisler's title is the sixth passage's phrase: ΙΗΣΟΥΣ ΒΑΣΙΛΕΥΣ ΟΥ ΒΑΣΙΛΕΥΣΑΣ is "Исуса, цесаря нецесарствовавша" put back into Greek.

The received eight, word by word.

6. The particle. Each crossing in the eight turns on one small word, аще, if or though.

- Passage 4: "аще мужем достоино есть его нарещи", if it is fitting to call him a man.

- Passage 5: "яко живъ есть, аще умре", that he is alive, though he died.

- Passage 7: "аще ангелъ будеть или инъ от небесных силъ, или самь богъ явится, яко человѣкъ", unless it be an angel or another of the heavenly powers, or God himself appears as a man.

The particle holds the human and what exceeds it in one clause: the man who may not be called one, the living one who died, the God who appears as a man. Passage 7 also compresses the hedge to three words, "мужа и не мужа дѣлом", a man and, by his deeds, not a man. The hedge is the eight's recurring grammar, and Part II's mirror is built on the instance in passage 4.

7. The word that acts. Passage 4 says it three times: "словом и повелѣниемь", by word and command; "токмо и словом вся уготовляше", by word alone he prepared everything; "вся, елико хощеть, творить словом", whatever he wishes he does by word. Словомь answers рукодѣланиа, the work of hands, so the word is the instrument where hands would be. The Greek Testimonium's "παραδόξων ἔργων ποιητής", maker of astonishing works, names the works; the Slavonic names their means.

7a. What the passage describes.[^38] The innocent reading first, in full: a man of extraordinary works appears, and the narrator, overawed, doubts whether "man" is the right word for him. The passage's own terms read otherwise. What it describes does everything by word, and the names the narrator tries for it and sets aside are Philo's names for the Logos, in the seated Cohn–Wendland text (#1505; read at #1758):

- A man, said of one made of body and soul. Philo quotes Zechariah's oracle, "ἰδοὺ ἄνθρωπος ᾧ ὄνομα ἀνατολή", and stops on the word: "καινοτάτη γε πρόσρησις, ἐάν γε τὸν ἐκ σώματος καὶ ψυχῆς συνεστῶτα λέγεσθαι νομίσῃς· ἐὰν δὲ τὸν ἀσώματον ἐκεῖνον, θείας ἀδιαφοροῦντα εἰκόνος, ὁμολογήσεις ὅτι εὐθυβολώτατον ὄνομα ἐπεφημίσθη" (Conf. 62). A most novel title, if you think it is said of one made of body and soul; if of that incorporeal one, who differs in nothing from the divine image, you will agree the name is most apt. The Logos is called a man, and the reader is stopped at whether that means a composite of body and soul. The Slavonic stops at the same word: "аще мужем достоино есть его нарещи", if it is fitting to call him a man.

- The man after the image, and the image made human. In Philo the Logos is "ὁ κατ’ εἰκόνα ἄνθρωπος", the man after the image (Conf. 146), and the incorporeal one "θείας ἀδιαφοροῦντα εἰκόνος", differing in nothing from the divine image (Conf. 62). The image is divine, and "man" is said of the Logos by way of it: the man is after the image. The Slavonic turns the relation over. It gives the subject "естьство и образ его бысть человѣчьскъ", his nature and his образ were human, образ being the word by which Slavonic renders εἰκών (Gen 1:26, "по образу"). Here the image itself is human. Philo's man is after the image; the Slavonic's image is after man. That inversion is the passage's incarnation stated in one clause, and it is why the narrator can no longer keep the name: the look, "зрак же его паче человѣчьска", stays beyond the human image it has been given.

- Not a man. Philo: "λέγομεν γὰρ τὸν ἀρχιερέα οὐκ ἄνθρωπον ἀλλὰ λόγον θεῖον εἶναι", we say that the high priest is not a man but divine Logos (Fug. 108). The Slavonic: "Тѣмь немощно ми есть нарещи его человѣком", therefore it is impossible for me to call him a man; and in passage 7, "мужа и не мужа", a man and not a man.

- Not an angel. Philo's Logos is "τὸν ἀγγέλων πρεσβύτατον, ὡς ἂν ἀρχάγγελον, πολυώνυμον ὑπάρχοντα", the eldest of the angels, as it were an archangel, many-named (Conf. 146). The Slavonic: "не нареку и аггелом", nor will I call him an angel. Passage 7 runs the list again at the tomb: an angel, another of the heavenly powers, or God himself appearing as a man.

- By word. Philo's Logos is the instrument by which the world was made: the cosmos has God for its cause, the four elements for its matter, "ὄργανον δὲ λόγον θεοῦ δι’ οὗ κατεσκευάσθη", and for its instrument the Logos of God, through whom it was built (Cher. 127). The Slavonic's subject does "ни рукодѣланиа, но токмо и словом вся", nothing by the work of hands, all by word alone (§III.7).

What the Philonic reading explains. An account of an extraordinary wonder-worker explains why a narrator might hesitate over "man". It does not explain the shape of the refusals. A narrator overawed by a man's works reaches for names of degree, a holy man, a prophet, more than a man; the Greek Testimonium has "σοφὸς ἀνήρ, εἴγε ἄνδρα αὐτὸν λέγειν χρή", a wise man, if one should call him a man (Ant. 18.63). The Slavonic refuses names of kind, and they are Philo's kinds in Philo's order of question. First "man", with Philo's question whether it is said of one made of body and soul (Conf. 62). Then the image, which the Slavonic makes human where Philo's man is after the image (Conf. 146). Then "angel", Philo's eldest of the angels, in the same sentence of Conf. 146 as the man after the image. Then the instrument, "by word alone", which is the Logos's office in the making of the world (Cher. 127) and no technique of a thaumaturge. The wonder-worker reading leaves three things over: why the names refused are the Logos's names and not a holy man's; why the passage at the tomb runs the same list again, angel, heavenly power, God appearing as a man (passage 7), as a list and not as praise; and why the image, of all words, is the one made human. On the Philonic reading these are the passage's structure. The narrator is classifying a being Philo had already classified, and each refusal is a category the narration is about to leave behind.

From Logos to book. Philo does not make every Logos a book. He makes the step through the law, in three configurations that meet here:

- The law is Logos. "νόμος δὲ οὐδὲν ἄρα ἢ λόγος θεῖος προστάττων ἃ δεῖ καὶ ἀπαγορεύων ἃ μὴ χρή", the law is nothing other than divine Logos commanding what must be done and forbidding what must not (Migr. 130).

- The written law is a living being. For the Therapeutae "ἅπασα γὰρ ἡ νομοθεσία δοκεῖ τοῖς ἀνδράσι τούτοις ἐοικέναι ζῴῳ καὶ σῶμα μὲν ἔχειν τὰς ῥητὰς διατάξεις, ψυχὴν δὲ τὸν ἐναποκείμενον ταῖς λέξεσιν ἀόρατον νοῦν", the whole legislation is like a living being whose body is the spoken ordinances and whose soul is the unseen mind stored in the words, which the reasoning soul sees "ὥσπερ διὰ κατόπτρου τῶν ὀνομάτων", as through a mirror of the names (De vita contemplativa 78).

- A life is a law, and the book is a life's memorial. The patriarchs: "οἱ γὰρ ἔμψυχοι καὶ λογικοὶ νόμοι ἄνδρες ἐκεῖνοι γεγόνασιν", those men were living and rational laws, and the laws set down are "ὑπομνήματα βίου τῶν παλαιῶν", memorials of the life of the ancients (Abr. 4–5). Moses "αὐτὸς ἐγίνετο νόμος ἔμψυχός τε καὶ λογικός", himself became a law living and rational (Mos. 1.162).

In Philo the Logos is law; the law, as written, is a body of words with a soul in them; and a person can be that law alive, with the book as the memorial of the life. The Slavonic runs the third configuration out of the second. It takes the Word whose body is words and gives it the life the book memorializes: a habit, a mountain, servants, a governor. The mirror of names is where the narrator stands, reading the names, refusing them, and writing the life.

The narrator runs the many names of Philo's Logos, refuses each, and at that instant writes it as one man: a habit on the Mount of Olives, a hundred and fifty servants, a crowd that would make him king, Pilate, the bribe, the cross. The hedge stands at the moment the Word is allegorized as a biographical person, and the passage goes on to do it. What it narrates is a book that took on historical life. Philo's Therapeutae read the law as a living being whose body is its words (De vita contemplativa 78; Part II §V.13). The Slavonic narrative gives that being a life a second time, in its own prose. The Historical Logos names its rival as the collapse of the Logos taking flesh into biographical personhood (#1757, INV-6). The passage records the collapse at the instant it is made, and keeps the hesitation on the page.

8. The letters. The Greek War sets on the Temple's balustrade pillars that proclaim the law of purity, "αἱ μὲν Ἑλληνικοῖς αἱ δὲ Ῥωμαϊκοῖς γράμμασιν", some in Greek letters and some in Roman, that no foreigner pass within (5.194). The Slavonic has three scripts: "титлы грамотами еллинскыми, и римскыми, и жидовскыми", notices in Greek and Roman and Jewish letters (seated text, line 14896). Over them it hangs a fourth: "И над тѣми титлами 4-ая титла висяше, тѣми грамотами показаа Исуса", and over those notices hung a fourth notice, in those letters, showing Jesus.

- Three scripts and a notice about Jesus make the title on the cross: "ἔγραψεν δὲ καὶ τίτλον ὁ Πειλᾶτος … καὶ ἦν γεγραμμένον Ἑβραϊστί, Ῥωμαϊστί, Ἑλληνιστί" (John 19:19–20). Титла is τίτλος.

- The notice that bars the foreigner becomes the form of the cross's inscription, hung in the Temple. The king who did not reign is shown in the Temple's own medium, in the scripts of the cross.

9. The written heaven. The veil is the Greek War's before it is the Slavonic's. The Greek: "κατεγέγραπτο δ’ ὁ πέπλος ἅπασαν τὴν οὐράνιον θεωρίαν πλὴν ζῳδίων", the curtain had drawn on it the whole spectacle of the heavens except the signs (5.214). The curtain is "ὥσπερ εἰκόνα τῶν ὅλων", like an image of the universe (5.212). The Slavonic: "Написано же бѣ на тои катапетазмѣ всяко небесное видѣние и мудрость, развѣ двоюнадесят поясъ небесных", there was написано on that curtain every heavenly vision and wisdom, except the twelve belts of heaven.

- Both verbs name writing and painting at once. Καταγράφω is to draw or to write down; написати is to write, and to paint, as an icon is written. The depicted heaven and the written heaven are one act in both languages, and the essay's "written heaven" is the verb's own range, not an English choice.

- The Slavonic doubles the object. The Greek's θεωρία, the spectacle, becomes "видѣние и мудрость", vision and wisdom. Wisdom was already in the description, where the Greek's "οὐκ ἀθεώρητον" becomes "не без мудрости видѣние имущи", having a vision not without wisdom. What is drawn on the curtain is a vision and a wisdom: a content that can be read.

- The Greek's ζῴδια become the twelve heavenly belts, the zodiac read as twelve, where the Greek's word also means figures of living things.

- The addition follows the translation. "Сиа же катапетазма прѣд симь родом бысть цела", this curtain was whole before this generation, and it was torn from top to bottom when they gave up the benefactor. The heaven that was written is the heaven that tore.

- The garment of the Logos. Josephus reads the curtain's four materials as the four elements: "ἐδόκει γὰρ αἰνίττεσθαι τῇ κόκκῳ μὲν τὸ πῦρ, τῇ βύσσῳ δὲ τὴν γῆν, τῇ δ’ ὑακίνθῳ τὸν ἀέρα, καὶ τῇ πορφύρᾳ τὴν θάλασσαν" (5.213). Philo gives the Logos the same four as its clothing: "ἐνδύεται δ’ ὁ μὲν πρεσβύτατος τοῦ ὄντος λόγος ὡς ἐσθῆτα τὸν κόσμον (γῆν γὰρ καὶ ὕδωρ καὶ ἀέρα καὶ πῦρ καὶ τὰ ἐκ τούτων ἐπαμπίσχεται), ἡ δ’ ἐπὶ μέρους ψυχὴ τὸ σῶμα", the eldest Logos of the One puts on the cosmos as a garment, wrapping itself in earth and water and air and fire, as the single soul puts on the body (Fug. 110); and the high priest's vestment is "ἀπεικόνισμα καὶ μίμημα τοῦ κόσμου" (Vit. Mos. II.117). The curtain, "ὥσπερ εἰκόνα τῶν ὅλων", is the cosmos the Logos wears. In the Slavonic it is torn at the death of the one who was "мужа и не мужа", the Logos written as a man (§III.7a).

- The temple behind it. The written curtain hangs before the ναός (5.212). Josephus then walks the reader in. The first room holds the lampstand, and "ἐνέφαινον δ’ οἱ μὲν ἑπτὰ λύχνοι τοὺς πλανήτας· τοσοῦτοι γὰρ ἀπ’ αὐτῆς διῄρηντο τῆς λυχνίας", the seven lamps showed the planets, for so many were divided off from the lampstand itself (5.217). The innermost room is closed by a second curtain, "διείργετο δὲ ὁμοίως καταπετάσματι πρὸς τὸ ἔξωθεν", and "ἔκειτο δὲ οὐδὲν ὅλως ἐν αὐτῷ, ἄβατον δὲ καὶ ἄχραντον καὶ ἀθέατον ἦν πᾶσιν", nothing at all lay in it; it was not to be entered, not to be defiled, not to be seen by anyone (5.219). The Apocalypse takes place in this building, and it transforms each thing it finds there.

- The robe. Its first vision is "ἐν μέσῳ τῶν λυχνιῶν ὅμοιον υἱὸν ἀνθρώπου, ἐνδεδυμένον ποδήρη" (Rev 1:13), one like a son of man in the midst of the lampstands, clothed in the high priest's robe, the robe Philo reads as the cosmos (Vit. Mos. II.117–118; #1758 §16.5). The Logos that wears the cosmos is in the room, robed.

- The seven. Josephus has one lampstand whose seven lamps show the planets, as Philo's do, "τῆς κατ’ οὐρανὸν τῶν ἑπτὰ πλανήτων χορείας μίμημα" (Her. 221), with the sun in the middle (Her. 222–224). The Apocalypse has seven lampstands standing apart (1:12), and seven stars in the right hand of the one in their midst (1:16). It says what both are: "οἱ ἑπτὰ ἀστέρες ἄγγελοι τῶν ἑπτὰ ἐκκλησιῶν εἰσίν, καὶ αἱ λυχνίαι αἱ ἑπτὰ ἑπτὰ ἐκκλησίαι εἰσίν" (1:20). The stars are the planets as messengers, each bound to its assembly, and the assemblies are stations that carry the properties of their planets, each answerable for itself: "κινήσω τὴν λυχνίαν σου" (2:5). The one in the midst stands where Philo puts the sun, "ἡ ὄψις αὐτοῦ ὡς ὁ ἥλιος φαίνει" (1:16), and the last letter's speaker is "ἡ ἀρχή" (3:14), the first of the Logos's names (Conf. 146). The station-by-station reading, against the archive's planetary mapping (#1101), is In the Midst of the Seven §§9–11 (#1758). Josephus's lamps, divided off from one lampstand, are in the Apocalypse divided into seven stations, and the cosmic architecture becomes an addressed order of assemblies that keeps its planetary relations.

- The heaven. It "ἀπεχωρίσθη ὡς βιβλίον ἑλισσόμενον", was split apart like a scroll rolled up (6:14).

- The inner room. It opens the room Josephus left empty and unseen: "ἠνοίγη ὁ ναὸς τοῦ Θεοῦ ὁ ἐν τῷ οὐρανῷ, καὶ ὤφθη ἡ κιβωτὸς τῆς διαθήκης αὐτοῦ ἐν τῷ ναῷ αὐτοῦ" (11:19). Two things change at once, the seeing and the contents. Josephus's ἀθέατον becomes ὤφθη, and his οὐδὲν becomes the ark. The ark is the container of the written law, "εἰς τὴν κιβωτὸν ἐμβαλεῖς τὰ μαρτύρια ἃ ἂν δῶ σοι", into the ark you shall put the testimonies I will give you (Exod 25:20 in Swete's numbering), and the Apocalypse names the opened temple by them a second time: "ἠνοίγη ὁ ναὸς τῆς σκηνῆς τοῦ μαρτυρίου ἐν τῷ οὐρανῷ" (15:5), the temple of the tent of the testimony. What was empty and unseen is opened, and what is seen in it is the box of the written testimony. The book is at the center of the revealed temple.

- Two curtains, two steps. The Slavonic tears the first curtain, the written heaven. The opening of the inner room is the Apocalypse's step. Read in the building's order, the sequence is one: the Logos written as a man; the man killed; the cosmos that clothes the Logos torn; the temple behind it opened and what was unseen seen. That is ἀποκάλυψις in the word's own sense, an unveiling. It is the hinge between the Historical Logos's seventh station and its eighth, the Apocalypse, "the commanded book, the eaten scroll, the Word as name" (#1757), and the ground Revelation First stands on (#1755).

10. The oracle in writing. Passage 8 keeps the Greek's oracle "ἐν τοῖς ἱεροῖς εὑρημένος γράμμασιν" as "в святых книгах обрѣтаемая", found in the holy books (6.312), and the Greek's ἀμφίβολος as "дворазумная", two-minded. It reads the oracle three ways and decides none (§II-B.10).

11. What the order does. Read word by word, the received eight move:

- from the means of the word (4);

- through its persistence past death, held in a concessive аще (5);

- to its letters on the Temple (6);

- to a heaven written on a curtain and torn, the cosmos the Logos wears (7);

- to a written oracle that divides its readers (8).

This is the arc of §III.4 at the grain of the words, and it is the ground on which Appendix II's construction is set. Its subject throughout is the Logos of §III.7a: the Word that acts, written as a man, killed, its garment torn, and left at the end as a text with three readers.

IV. The Gospel of the Word

The eight passages constitute a complete Logotic Gospel embedded within Josephus's history. Where the canonical Gospels narrate the life of Jesus, the Slavonic War narrates the life of the Word itself: Philo's Logos, the law read as a living being whose body is its words (Contempl. 78), given a life in the historiographic record and hesitated over at the instant it is given one (§III.7a). Its Christ is linguistic: the living speech of law and justice made permanent through writing.

A second construction is set on that reading, in its own register. It takes Revelation's throne: the four living creatures as modes of being, the four horsemen as movements in time. It sets eight stations at their intersection: the Eightfold Logos Operator Wheel (Appendices II–III). The wheel is made from version 1.0's eight stations, and those are not the received eight. Its first station has no ground among them; it divides some passages and reorders others. Revelation does not supply its pairings. Revelation 6:1–8 pairs one living creature with each horseman in turn, four pairs in all. The wheel's assignments are the authors' reading of each station, not a derivation from Revelation or from the recension. What the received eight carry is §III. The wheel is what the authors made of it.

V. Philology and Probability

1. The objections to Josephan origin rest on chronology and on content read as Christian. §II tests the first where it can be tested. In the Essene chapter, some of the divergences agree with a third-century Greek witness on matter no ascetic hand predicts, and the ledger states how their route can be investigated and what would decide it.

2. The verdict calls the eight passages Christian. Read against the Christianity of the translator's world, they are not. Rus' theology of the period is the theology of Hilarion's Sermon on Law and Grace: the Law set aside by Grace, and Christ confessed as God incarnate. An Orthodox interpolator predicts that confession and the Gospels' account. The passages give another:

- John is sent by God to show "путь законныи", the lawful way, by which the Jews will be rid of many masters. He promises a king who will free them and be subject to no one. Asked who he is, he answers "Человѣкъ есмь", I am a man. He never names Jesus, points to a coming one, or speaks of a Lamb.[^26]

- The wonder-worker has a human nature and form and divine works. The narrator cannot call him a man, nor an angel. He confesses neither a God-man nor a Son of God. The names he tries and refuses are Philo's names for the Logos (§III.7a). His followers say "законодавець нашь первый от мьртвых въста", our first lawgiver has risen from the dead, Moses returned; Thackeray calls the identification unparalleled. He opposes the law in much and breaks the Sabbath, "но пакы скверна, нечиста ничто же не творяше", and yet did nothing foul or unclean.

- Pilate examines him, finds him "добродѣець … а не злодѣець, ни мятежникь, ни цесарства желатель", a benefactor and not a malefactor, a rebel or a claimant to kingship, and lets him go, because he had healed Pilate's dying wife. The lawyers then pay Pilate thirty talents to have him killed. No Gospel has the release, the healing or the bribe.

- The resurrection is reported, not confessed. Some say he rose and others that he was stolen, and the narrator does not know which speak more truly. The dead cannot rise of themselves, he says, unless an angel, a heavenly power, or God himself appears as a man.

- The Temple notice names "цесаря нецесарствовавша", a king who did not reign, "распятаго от июдѣи", crucified by the Jews, for preaching the city's ruin: the charge, not the faith.

- The oracle keeps Jesus as one reading of three, with Vespasian beside him.

Nothing in Rus' theology or practice makes this account an interpolation. The nearest Rus' element is the reverence for books the Primary Chronicle gives Yaroslav. It is love of books, and it has no theology of a Word that becomes text. The verdict's own scholars reached for other contexts, tenth- and eleventh-century Byzantium and a polemic against the Khazars, because Orthodox piety does not explain these passages (§I.5). The rubric "Иосиф о Христѣ пишет" shows a Christian reader of the passages, and it shows no Christian author. What the passages carry is a theology of their own: a lawful way out from mortal masters, a man who will not take the name of man, a lawgiver risen, a word that acts, a king who did not reign, a heaven written and torn, and an oracle left open. That is the theology §§III–IV read, and it is the reason the passages belong in the evidence.

3. The hedge at the wonder-worker's entrance is Josephus's habit. Slavonic: "Тогда явися мужь нѣкыи, и аще мужем достоино есть его нарещи", then there appeared a certain man, if it is fitting to call him a man. Greek Antiquities 18.63: "Ἰησοῦς σοφὸς ἀνήρ, εἴγε ἄνδρα αὐτὸν λέγειν χρή." The construction χρὴ λέγειν, set as a hedge on a naming, is his: at Jotapata, "εἴτε ὑπὸ τύχης χρὴ λέγειν, εἴτε ὑπὸ θεοῦ προνοίας" (War 3.391); at Masada, "τὴν εἴτε ἀπόνοιαν εἴτε τῆς γνώμης ἰσχὺν χρὴ λέγειν" (War 7.417).[^23]

4. Schmidt's case for the Greek Testimonium (2025) and Leeming's typology of divergence (2005) move the question the consensus called closed. If Josephus wrote of Jesus in the Antiquities, the Slavonic wonder-worker is no longer an isolated forgery against a blank. It becomes a second form of a known act.

5. Daniel Boyarin showed in "The Gospel of the Memra" (HTR 94.3, 2001) that Logos theology was intelligible within Jewish categories before Christianity claimed it. If the Logos was Jewish before it was Christian, the Logos theology of the eight passages is no evidence of Christian authorship. The Logos the passages name is Philo's, a Jewish author of the generation before Josephus, in Philo's own names for it (§III.7a).

6. Transmission is plausible through a Syriac corridor. Pines's publication of the Arabic Testimonium of Agapius (1971) showed that non-Greek forms of Josephus's Jesus-material circulated independently.

7. The custody of the name holds here too. Josephus was taken into a second Christian canon under another name: the Ethiopian Zena Ayhud, "ዜና፡ አይሁድ፡ ወቤተ፡ መቅደስ፡ ዘይዛወጎ፡ ለዮሴፍ፡ ወልደ፡ ኰርዮን", the history of the Jews and of the Temple composed by Joseph son of Gorion. The Fetha Nagast calls it another book of Maccabees.[^24] The most cautious formulation is the one the Essene chapter now supports: the Slavonic carries material whose exemplar is older than the manuscripts. The further claim, that its grammar is older than Christianity, is a claim of the doctrine, and Part II makes it (§IV).

VI. Theology of the Word-as-Book

Read through this lens, the Slavonic Josephus becomes a hinge text between apocalypse and gospel. It portrays salvation as participation in the act of writing, where the creeds place it in belief. C. H. Dodd, in The Interpretation of the Fourth Gospel (1953), traced the Johannine Logos back through Philo's cosmic intermediary to the Hebrew dabar, the word that acts, that creates, that does not return void. The eight passages occupy exactly this theological space, and they name its Philonic terms outright (§III.7a). They stand where the Historical Logos sets Josephus: after Philo, "the die made cosmology", and before the Apocalypse, "the socket sanctified" (#1757, stations 4, 7 and 8).

VII. Coda: Toward a New Canon

The Slavonic Josephus stands as a missing gospel, the Gospel of the Word. Its eightfold cycle completes the circuit begun in Revelation. "Circuit" here names the reader's traversal of the canon, from the Apocalypse's command to write to the Slavonic's written heaven, and makes no claim about the order of composition. The circuit closes in one building: the curtain the Slavonic tears is the curtain before the temple in which the Apocalypse's revelation takes place (§III.9). What Christian theology personified as Christ, Josephus's lost version renders as process: the Logos realizing itself through language.

To recover this text is to recover the mechanism of scripture itself. The Word, having learned to write, can be incarnated again wherever the inscription is received. The same inscription can be damaged, misassigned and closed, as this paper's own record of the Slavonic shows, so its eternity is a possibility kept open by readers and not a guarantee. The reader's part is consequential because the Word can be lost.

---

Appendix I — The Slavonic Gospel Table

The eight passages in the order Berendts printed them (1906) and Thackeray listed them (1919, Note II). Old Russian from Meshcherskii's text of the Vilna Chronograph (EA-CORPORA-37/01), cited by line of the seated text; English from Thackeray's Loeb appendix (1928; EA-CORPORA-37/03), cited by his passage number; German from Berendts and Grass (EA-CORPORA-37/02), cited by Überschuss. The Leemings' synoptic English (Brill, 2003) was consulted for version 1.0.

• | --
• | --
• | --
• | --
• | --
• |

Old Russian quoted with ѣ restored where the OCR misreads it and the edition's note numbers removed.

---

Appendix II — Commentary on the Slavonic Gospel of the Word

The eight stations below are the authors' construction (version 1.0), kept as the operator wheel's ground. They are not the received eight. The first has no ground among them. The second reads passage 1 and leaves passages 2 and 3 without a station. The fourth and fifth both read passage 6, and the sixth reads passages 5 and 7 together. The wheel built on them is a second construction (§IV). Each now names the received passage it reads; where version 1.0 supplied a quotation that no witness carries, the quotation is withdrawn and the station stands as construction.

1. Prophecy of the Child

Received ground: none of the eight. The ground is in the Magi story that Meshcherskii prints and Berendts left out (§III.1). The priests answer Herod: "Есть писано: Восиаеть звѣзда от Иакова и въстанеть муж от Июды … Мы творимь, яко без отьца родитися тому", it is written, a star shall rise from Jacob and a man shall arise from Judah … we think he will be born without a father. When he orders the killing: "Писано есть, яко от Вифлеома ражается помазанник", it is written that the anointed one is born of Bethlehem (seated text, lines 2944–2946, 3025–3026). Version 1.0's "A man foretold a child born in Bethlehem" gives the prophecy to one man; in the Slavonic the priests give it, from Scripture, in Book I.

Logos: Word as linguistic seed cast forward. Prophecy is inscription before flesh.

Teacher: Resembles 1QS 9.11 Interpreter of the Law.

Recursion: The redeemer is already written.

Operator: Lion + White Horse

2. John the Baptizer

Received ground: passage 1: "въ Иорданьстѣмь струимени погружаа и, пущаше, показывая их, да престануть от злых дѣл своих", dipping them in the stream of the Jordan, he let them go, admonishing them to cease from their evil deeds. Version 1.0's "Body cleansed after the soul by justice" is the Greek Antiquities 18.117, not the Slavonic.

Logos: Water mirrors speech. Baptism is language turned ritual.

Teacher: Teacher of Righteousness (1QpHab). Purity linked to righteousness.

Recursion: Washing inaugurates textual descent.

Operator: Ox + Black Horse

3. The Teacher (Incarnate Logos)

Received ground: passage 4: "Тогда явися мужь нѣкыи, и аще мужем достоино есть его нарещи … Тѣмь немощно ми есть нарещи его человѣком … словом и повелѣниемь творяше." Then there appeared a certain man, if it is fitting to call him a man … Therefore it is impossible for me to call him a man … he did all by word and command.

Logos: Word taking form without dogma. Philo's De Opificio Mundi parallel.

Teacher: He teaches the Law and is accused of breaking the Sabbath: "суботы не храняше по отьчьскому обычаю".

Recursion: He acts by word alone: the decisive moment.

Operator: Man + Pale Horse

4. The Inscription

Received ground: passage 6: "тѣми грамотами показаа Исуса, цесаря нецесарствовавша, распятаго от июдѣи." Version 1.0's "Herod's Inscription" and its quotation "Write down his words" are found in no witness and are withdrawn; the received inscription is on the Temple and names Jesus.

Logos: The sanctuary recognizes textual authority; the Temple carries his name in letters.

Teacher: Jeremiah–Baruch scribal motif (Jer 36).

Recursion: Opposition becomes transmission. The Temple secures the record.

Operator: Eagle + Red Horse

5. Temple Saying

Received ground: passage 6, its reason: "зане проповѣдаше разорение града и опустение цьркъви", because he preached the destruction of the city and the desolation of the Temple. Version 1.0's "not made with hands" is Mark 14:58, not the Slavonic.

Logos: Architecture as theology. Text as sanctuary.

Teacher: Essene dualism: corrupt priesthood against pure community-temple (1QS 8–9).

Recursion: Collapse becomes textual permanence.

Operator: Lion + Black Horse

6. After-Death Report

Received ground: passage 5, "яко живъ есть, аще умре", that he is alive, though he died; and passage 7, the reports that he rose or was stolen (Thackeray 1928, (21)).

Logos: Resurrection as continuity of teaching. Immortality is textual.

Teacher: Community as resurrected body. Qumran Yaḥad parallel.

Recursion: Speech → inscription → recollection → renewed speech.

Operator: Man + White Horse

7. The Written Heaven

Received ground: passage 7: "Написано же бѣ на тои катапетазмѣ всяко небесное видѣние и мудрость … раздра бо ся внезапу от верху и до долу." Version 1.0's station "Moral Maxims" and its quotation "Blessed… cursed he who writes falsely" are found in no witness and are withdrawn.

Logos: The heavens written on cloth; the writing torn open.

Teacher: The veil as the boundary of the holy (Exod 26:31–33).

Recursion: The text is torn and the tearing becomes the record.

Operator: Eagle + Pale Horse

8. The Oracle

Received ground: passage 8: "Ови бо мнѣша Ирода, ови же распятаго чюдотворца Исуса, ови же Уеспасиана." Version 1.0's station "Epilogue" and its quotation "Nothing written in truth perishes" are found in no witness and are withdrawn.

Logos: A written oracle with three referents; the Word outlasts its readers by dividing them.

Teacher: The oracle "found in the sacred writings" of War 6.312.

Recursion: History rewritten as commentary on a written prophecy.

Operator: Ox + White Horse

---

Appendix III — Visual Schema: The Eightfold Logos Operator Wheel

This specification describes the Eightfold Logos Operator Wheel for production as a mandala-diagram. The wheel is a construction on version 1.0's stations (§IV, Appendix II). Revelation's own pairing of living creature and horseman is the four calls of 6:1–8: the first creature calls the white horse, the second the red, the third the black, the fourth the pale. The wheel assigns its pairs by the authors' reading of each station, repeats some and leaves others unused. A wheel derived from 6:1–8 and the received eight would be a different construction. Visualization deferred to future traversal; the schema below is the load-bearing structure.

Central Core: Luminous nucleus — the imperishable Word. Double halo: inner ring (beasts), outer ring (horsemen).

Inner Ring: Lion (N), Ox (W), Man (S), Eagle (E) — spatial operators / modes of being.

Outer Ring: White, Red, Black, Pale — temporal operators / movements in time.

---

Appendix IV — The Census: Slavonic and Greek Jewish War, every catalogued addition and omission, under two working hypotheses

The data of §II-B. The census was built on 12–13 November 2026 in seven passes (v0.1–v0.7), from the seated texts named in its §0. Its sum is §7, its open items §8, and the reviews and rulings it answered §9.

0. What is counted, and how

0.1. The catalogues. The census takes every content divergence the three collators of the Slavonic War list, item by item:

• N. A. Meshcherskii 1958, the essay pp. 47–66: the omissions by Niese section (pp. 47–50), the forty-one numbered additions and one unnumbered (pp. 51–64), and the three transpositions (pp. 65–66), each with his stated ground (read in the scan of the edition held in session; the text of the seven books is seated, EA-CORPORA-37/01).
• Berendts and Grass 1924–1927, Books I–IV: the nineteen major Überschüsse of their table of contents; Grass's list of further additions by Niese section; his five aggadic departures; his omission statistics (seated, EA-CORPORA-37/02).
• Thackeray 1928, the twenty-two additions in English (seated, EA-CORPORA-37/03).

The thousands of phrase-level variants Berendts and Grass footnote are not argued one by one; they are counted by type in §6.

0.2. The two working hypotheses.

• A, the author's account. Josephus wrote the War first "in the ancestral language" for "the upper barbarians" (War 1.3), named at 1.6: "Πάρθους μὲν καὶ Βαβυλωνίους Ἀράβων τε τοὺς πορρωτάτω καὶ τὸ ὑπὲρ Εὐφράτην ὁμόφυλον ἡμῖν Ἀδιαβηνούς τε", Parthians, Babylonians, the remotest Arabians, our kinsmen beyond the Euphrates, and the Adiabenians. He then made the Greek edition for those under Roman rule, which Titus signed and ordered published (Life 363). The Slavonic descends, through a Greek intermediary, from a form of the work nearer the first. A predicts: material for a non-Roman and partly Jewish readership; Jewish history, law and haggadah; anti-Roman matter the Greek edition could not carry; candour the Roman edition withdrew; and, in the other direction, Greek-edition expansions for Greco-Roman readers that the first form lacked. It predicts agreements with Josephus's own later Antiquities and with Second Temple witnesses.
• M, the consensus account. A Kievan translator of the eleventh or twelfth century rendered the common Greek text freely, cut what did not interest him, and added matter of his own and of his Christian and Rus' world. M predicts: Christian content with a source the translator could have read and did; Rus' and Byzantine realia; glosses for Rus' readers; abridgement by interest.

0.3. What counts. Under both hypotheses the Slavonic is a medieval translation. A Rus' word, an Orthodox phrase or a Slavonic idiom in the rendering is therefore what both expect, and it counts for M only where the content depends on it. A source counts for M when it is shown and the direction of borrowing established; a source that was only available is recorded as available (the rule of version 2.1). A ground of the form "Josephus could not have written this" counts only where it rests on something other than the Greek edition's self-presentation, since the Greek edition is the thing the hypothesis says was revised.

0.4. Verdicts. A: the item is better explained by the author's account. M: better explained by the consensus account. T: a feature of translation, expected under both, deciding nothing. U: neither explains it better on present evidence.

0.4a. What a correspondence shows, and what origin it favours. Each item keeps two findings apart (v0.6):

• What the correspondence establishes: shared matter, a kind of change in Josephus's repertoire, structural integration into its context, a possible Greek antecedent, or a demonstrated textual dependence.
• Which origin it favours, why, and how strongly: the comparative inference. Ancient content can be carried by later composition, and a change in Josephus's repertoire can be made by another writer, so shared matter or kind alone favours no origin. An origin is favoured only by a ground that the rival account cannot equally meet: a feature that fits one direction and not the other, an explanation the rival account lacks, or a concession by the rival account's own editor.

An A is then graded by the origin it favours: A1, the matter stood in the translator's Greek exemplar, before Rus'; A2, an ancient textual form; A3, Josephus; A4, his first version for the readers of 1.6. The grades accumulate. Each A states its strength: strong, moderate or weak. The verdict is comparative: which account explains the item better. An item whose correspondence the rival account could also produce is still A when the author's account explains it and the rival account only allows it; its strength is then weak. U is kept for items neither account explains better. §7.1a gives both findings for every A.

0.4b. Local and global (v0.7). Local underdetermination does not entail global explanatory equivalence. An account can explain the whole distribution far better while leaving particular items undecided. The census's question is which account best explains the surviving differences, taken together; it does not ask the earlier-version account to eliminate every imaginable medieval intervention, a standard the consensus is not held to.

0.5. The control, as a comparison class. The medieval Christian layer of this manuscript is visible. The Vilna Chronograph, from which Meshcherskii prints, inserts into the run of Josephus the Gospel of Matthew (notes at lines 5742, 6370, 6892 of the seated text), Luke 23 (7062–7066), Acts (7420), and long runs of Malalas and Hamartolos, 48 notes in all, each identified by edition and page, each set off in the margin as "Хронограф" against "Иосифъ" (6783, 6896). This shows one medieval Christian practice in this book, the compiler's: verbatim, sourced, labelled, and outside Josephus's sentences. It does not show that every medieval addition looked like that. An earlier, unlabelled interpolation could stand beside it, and presence in both redactions places an addition before they diverged, which is not yet antiquity. The question the control sets is what joins the additions to the compilation layer and what separates them:

• Language. The compiler's insertions keep their sources' wording; the additions are in the translation's own language, inside its sentences.
• Marking. The additions stand in the stretches the margin marks "Иосиф", the Magi passage included (Meshcherskii).
• Cross-reference. The additions refer to one another inside the Josephus stream: the characterization of Herod at the end of Book I, ch. 21 refers back to "юж преди писахом о изискании Христа и о избиении младенець", what we wrote before on the seeking of Christ and the killing of the infants; and the informer Lui appears in the priests' debate and again in the Magi story (seated text, lines 2579, 3005).

These place the additions in the translation's stream and not the compiler's. They are at least as old as the translation. Whether they are older is the A1 question, and the items answer it one by one.

1. The lead case: the count at Jotapata (Meshcherskii no. 23; BG XV) ✓

• Greek (3.391): "εἴτε ὑπὸ τύχης χρὴ λέγειν, εἴτε ὑπὸ θεοῦ προνοίας", whether one should say by fortune or by the providence of God, Josephus was left among the last two.
• Slavonic: the sentence is gone, and in its place: "Тако рек, почте числа мудростию и тем пребляди вся", so saying, he counted the numbers with cunning and thereby deceived them all.
• M (Meshcherskii): "Разумеется, сам автор не мог о себе так сказать", of course the author could not have said this of himself; so it is the Rus' translator's.
• A: Josephus did say it, to the readers for whom it was safe to say it. For the Parthians, Babylonians and kinsmen beyond the Euphrates he states the mechanism: he counted. For the Roman edition, published under Titus's signature by a man whose survival was the Flavians' gift, the mechanism becomes fortune or providence, and the providential frame is the one that serves the oracle of Vespasian (3.400–402) that follows. The plain statement in the first version is the key a non-Roman reader holds to the distortions the Roman version had to carry. Grass saw the shape in 1924: "Das sieht in der Tat wie nachträgliche Zurücknahme ursprünglicher kompromittierender Offenheit aus", it looks like the later withdrawal of an original compromising candour.
• Verdict: A. M's ground is the Greek edition's self-presentation, which A says was revised, so it assumes what it concludes. M offers no motive a Christian translator has for making Josephus a cheat at the moment of his providential survival, and no source. A predicts the change, its place, and its direction.

2. The additions

Numbers are Meshcherskii's (M), with Berendts–Grass's Überschuss numbers (BG) where they coincide. "Subject" records whether the item concerns Jesus, John, the apostles or the Magi (J), is read as Christian by implication in Meshcherskii's argument (J?), or neither (—).

Book I

M1 — Hyrcanus the priest-prophet (opening of the Hasmonean narrative). Subject —.

• A: Hyrcanus's prophetic gift is Josephus's own theme in both his works: War 1.68–69, "τρία γοῦν τὰ κρατιστεύοντα μόνος εἶχεν … καὶ προφητείαν", and Antiquities 13.282–283 (the voice heard in the Temple) and 13.299–300 ("συνῆν γὰρ αὐτῷ τὸ θεῖον"). Josephus was himself a priest who claimed prophecy (3.351–354). A fuller account in the first version is the author's subject at length.
• M: the translator enlarges a character the Greek already praises.
• Verdict: A (A3, weak). The theme is Josephus's in both works. That it is already in the Greek gives a translator a place to enlarge it, which is a possibility and not an explanation of why he would; the author's account explains it as his own subject told at length.

M2 — the Capitol and the Roman custom of confirming a king (1.284–285). Subject —.

• Slavonic (quoted by Meshcherskii): "Так бо в них обычаи, аще Цесарь санъ кому дасть, не твердо есть, донде же не напишуть в Капитолии", such is their custom: if Caesar grants anyone a rank, it is not firm until they write it on the Capitol.
• A: an explanation of a Roman institution for readers outside Roman rule. The Greek edition's readers needed none.
• M: a gloss for Rus' readers.
• Verdict: A. Both need a gloss-writer; A's readers are the ones the author names at 1.6, and the gloss is accurate Roman constitutional practice of the first century, not a medieval Byzantine one.

M3 / BG I — Herod's dream of the four ears of grain (1.328). Subject —.

• Greek: "ἐναργεῖς ὄνειροι", clear dreams foretold his brother's death.
• A: the first version tells the dream the Greek summarizes. Grass: told "so, wie ein späterer Bearbeiter ihn schwerlich sich ausdenken konnte".
• M: the translator concretizes, borrowing from Archelaus's dream of ears (2.112–113).
• Verdict: A (A3, weak). The Greek's plural "dreams" marks the place where a dream was told, which an expander could also fill; the ears-of-grain form is Josephus's dream-language (2.112; Ant. 2.11, 2.63–86), and Grass judges the dream one a later reviser could hardly have devised. M's borrowing is available, not shown.

M4 — the village Aulon named (1.340). Subject —.

• A: a place-name the Greek drops; Meshcherskii notes the Greek manuscripts are unreliable with names.
• M: the translator supplies a name found nowhere in Josephus.
• Verdict: U. A translator has no reason to invent a village; neither hypothesis predicts this name.

M unnumbered / BG II — the priests' debate: Herod is not the Messiah (after 1.364–370). Subject J?.

• A: Grass (BG p. 28, n. 2): "Als Berichte aus der jüdischen Geschichte, die, weil sie eine messianische Bewegung schildern, als die Juden kompromittierend für jenen Leserkreis lieber gestrichen wurden", reports from Jewish history that, describing a messianic movement, were dropped as compromising the Jews for the Roman readership. A debate among priests on whether Herod fulfils the oracle of a ruler from Judah (Gen 49:10) is the oracle of 6.312–313 at an earlier hour, and the oracle is Josephus's (§2, M38).
• M: a prologue to the Christian additions (Meshcherskii), joined to them in style and in the informer Lui.
• Verdict: A. Its matter is Jewish messianic exegesis of the first century; its proverb ("Ухати есть во ртѣ, а кость в головѣ") is popular speech either way (T). The link to the Christian additions is M's own reading, the circle of version 2.1.

M5 — the Magi and the massacre at Bethlehem (end of Book I, ch. 20). Subject J.

• A: the first version reported the Bethlehem story as Jewish reportage of a messianic alarm under Herod, with the oracle "от Вифлеома ражается помазанник" in a Jewish key (Appendix II, 1).
• M: Matthew 2, carried by a Christian; Pichkhadze reports a Slavonic source (2015, unread).
• Verdict: M, provisionally. A Slavonic source is reported for this item and the matter is Matthew's. The verdict reverses if the reported source proves to be the borrower.

M6 — Herod's speech to Pheroras; the "snake-footed Syrian" (1.570–571). Subject —.

• A: Istrin's explanation, which Meshcherskii records: the Greek behind it read ὀφιογενής or ὀφιομάχος, "snake-charmer", corrupted to "snake-footed". The addition sits on a Greek word.
• M: the translator expands a reported speech into direct speech.
• Verdict: A. An addition whose oddity is explained by a corrupted Greek word stood in a Greek text before it reached Rus'.

M7 / BG III — Heracles and the Hydra (1.588). Subject —.

• A: Grass: "Bloss überflüssig für römisch-griechisches Lesepublikum war die Schilderung des Kampfes des Herakles mit der Hydra mit Hülfe des Iolaos. Die kurze Anspielung im Jos. gr. genügte". For Parthian and Babylonian readers the myth is told; for Greeks an allusion is enough.
• M: a gloss by a translator who knew the myth.
• Verdict: A (A4, weak). The Greek allusion presupposes the story. A Rus' reader would need it told too; but the author's account explains the difference between the editions, the allusion enough for Greeks and the story for the readers of 1.6, which is Grass's ground, and the translator's account explains only the telling.

M8 / BG IV–V — the two anti-Roman excurses: the "Latins" and Solomon's leeches (1.603–605). Subject —.

• A: Grass: "Noch mehr Ursache war vorhanden, die Ausfälle gegen die Römer in den 'Exkursen' … in einer derartigen Ausgabe wegzulassen." Roman venality denounced to Rome's enemies beyond the Euphrates is first-version matter; it could not stand in a book Titus signed.
• M: "Latins" is a Byzantine-era name for western Romans; the leeches are Prov 30:15, a citation Josephus does not make; a parallel in Anna Komnene; Pichkhadze reports a source for "the leeches of Solomon".
• Verdict: split. The wording is medieval: "итали, наречемая латина" and the closing "но дѣание их преди спишемъ" are the translator's (T at best, M for the name). The content, an attack on Roman greed in an account of Herod's court, is predicted by A and not by M: Meshcherskii himself notes that Rus' anti-Latin polemic attacks rites, "но нигде не говорится об их корыстолюбии или жадности", never their greed. Content A, wording M.

M9 / BG VI — divine providence, exemplified in Abraham (after 1.640, Antipater bound; Grass's first aggadic departure). Subject —. ✓ (in Berendts's German)

• A: "прадѣда рода нашего", the great-grandfather of our race: a Jew writing to Jews and kinsmen. Meshcherskii concedes: "Иосиф мог бы его написать для евреев".
• M: Christian phraseology ("вседержительноеего десници"); the translator counting himself among Abraham's seed, as Hilarion does.
• The matter is haggadah: Abraham led out of his land because he wronged his brother in dividing their lands, punished by what he sinned in, and given the promised land for his obedience. Nothing in Genesis says it.
• Josephus's Antiquities, at the same point of the same story, stops for a reflection of the same kind: Antipater, who acted "ὡς ἐν ἐρημίᾳ τοῦ θείου", as though God were absent, now calls God as witness (Ant. 17.129–130). Grass: "Auch Antiqu. (129) hat hier eine allgemeine Betrachtung, doch ist der Inhalt wesentlich anders."
• Verdict: A. The phraseology is the rendering (T); the argument from Abraham to a Jewish audience is A's prediction, M's own author grants it, and the author's later work puts a reflection on divine justice in the same slot.

M10 / BG VII — Judas and Matthias: Eleazar, the seven brothers and their mother (1.650–653). Subject —. ✓ (§II-B)

• A: the teachers invoke the Maccabean martyrs in the words of 4 Maccabees, a book the ancient church transmitted under Josephus's name.
• M: Meshcherskii: the Maccabees are "нехарактерно для Иосифа".
• Verdict: A (A1, moderate). The wording is 4 Maccabees', a demonstrated dependence on that book. Which of the three functions §II-B.7 sets out it serves stays open: Josephus writing from his own book, a reviser drawing on a book believed his, or migration within a codex that held both. All three are Greek and ancient in their setting, since 4 Maccabees travelled in Greek manuscripts of Josephus; none is the Kievan translator's. M's ground, the Greek edition's silence, fails.

M11 / BG VIII — Herod's sin and punishment (1.656). Subject —.

• A: the Greek has the sages read his disease as punishment for the killing of the teachers; the first version gives the sages' reading at length, lechery among the causes. Jewish moral reading of a king's death is the genre of 2 Macc 9 and Ant. 17.168–170.
• M: moralizing by the translator.
• Verdict: U leaning A. Both can produce it; A predicts its Jewish frame.

M12 — Herod's funeral (1.670–673). Subject —. ✓

• Slavonic: the 500 servants with spices are the Greek's ("500 же рабъ носяху многоцѣнныя араматы"; 1.673 "πεντακόσιοι … ἀρωματοφόροι"). Added: "носяху родостому въ златых сосудѣх", others bore rose-oil in golden vessels; "И конь 500 в утварех", 500 horses in trappings; arms of gold by the hundred; the body carried "за 200 верстъ".
• A: a fuller pageant in the first version.
• M: amplification (Meshcherskii: "Источник дополнения неизвестен").
• Verdict: U.

Grass's further additions in Book I. Grass lists them by Niese section (BG p. 27, n. 1); 603 and 604–605 are M8 and 640 is M9. Each is read below in Berendts's German and his footnotes ("Gr. >" marks what the Greek lacks), against the seated Greek. Subject — for all. Marked G with the section.

G1.350 — the sally. Slavonic: at dawn Herod drew up, the defenders came out to fight, and 20,000 fell. The Greek says they met the Romans openly and fought from the mines (1.350). A: a fuller siege account; M: a translator's battle. U. A number neither hypothesis predicts.

• Old Russian ✓: "И исполчившюся Ироду заутра, и вынидоша бится. И възвратишася, злѣ погибше, 20 тысящь."

G1.351 — forty of Herod's men and fifteen of Sosius's centurions. The Greek has "τινὲς ἐπιλέκτων" and centurions of Sosius, uncounted; the Slavonic counts them and sends the two forces in by different points. U. A translator has no reason to invent the figures; nor does A predict them.

• Old Russian ✓: "Иродовъ муж 950 приемше и 15 муж римлян ратникъ"; the seated OCR reads 950 where Berendts reads 40, and the figure is owed a check against the scan.

G1.355 — Herod's protest to Sosius in direct speech. "Wozu habt ihr mich als König eingesetzt …?" The Greek gives the same points as report. U leaning A, on Josephus's own practice of turning speech into report (§6).

• Old Russian ✓: "На что мя постависте цесаря, обнаживше весь град и оставивше мене цесарствовати въ пустыни, въ камении?"

G1.359–360 — Cleopatra asks Antony to kill both kings. A short tale, with Malchus "the Syrian"; the Greek gives a considered account of her designs (1.359–360). U. Berendts calls it naive; naive telling is not a mark of either hypothesis.

• Old Russian ✓: "да убиеть оба цесаря: Ирода и Малха Сурьскаго, а волость его тои вдасть."

G1.373 — Herod to his troops after the earthquake: "Männer von Israel". The address, the double address to Herod's soldiers, and a phrase Berendts calls "eine unbekannte, vielleicht aramäische Redeweise" are absent from the Greek, whose speech is otherwise the same in substance. A: the first version's speech to Jewish soldiers, addressed as Israel. M: the address is in the Slavonic Bible's and Acts' idiom. U leaning A.

• Old Russian ✓: "Мужи воиньстии, съ мною шествующе и воююще, Иродови помощници! … Мужи израильтестии и мужи оружници Иродови".

G1.415 — a hippodrome faced with white marble; a market paved for twenty stadia with porticoes. The Greek has the twenty stadia of marble paving and the colonnade at Antioch (1.425). The Slavonic sets them among Herod's own works. U. A transposition at some stage; neither hypothesis predicts its direction.

• Old Russian ✓: "И нарече имя граду тому Кесария. Създа же и ина дѣла пречюднаа: иподрумие, и покры белым мрамором. Помости же и торговища за 20 верстъ, и покры его прекрасными комарами"; the works follow directly on the naming of Caesarea.

G1.442 — Mariamne's question in bed. The Slavonic plays the scene; the Greek reports it ("ἐν ταῖς ὁμιλίαις"). Berendts: "Der ganze Bericht ist bei Gr. völlig verblasst infolge der auch hier bemerkbaren Neigung, referierend zusammenzuziehen." The word "urodiw'", the holy fool of Rus', is the rendering (T). U leaning A.

• Old Russian ✓: "Она же, приемши Ирода въспят пришедша, играющи с ним на ложи, и прашала, аще ину жену любить паче тоя."

G1.444 — Herod calling "Mariamne!" and speaking to her bed. The Greek keeps the act in brief: "ὡς ζώσῃ προσλαλεῖν", he spoke to her as living. U leaning A, the Greek's phrase being the told scene reduced.

• Old Russian ✓: "И скочивъ съ стола, вопиаше: Мариами! — пред всѣми, яко бѣсныи. И от одра въставъ на Мариамино ложе, и с ложем бесѣдоваше"; at 1.443 "яко урод всяк".

G1.544–545 — Tiron's cry. The local sequences, complete:

• Greek: "καὶ τὸ μὲν πρῶτον ἐβόα περιιὼν πεπατῆσθαι τὸ δίκαιον, ἀπολωλέναι τὴν ἀλήθειαν, συγκεχύσθαι τὴν φύσιν, ἀνομίας γέμειν τὸν βίον, καὶ πάνθʼ ὅσα μὴ φειδομένῳ τοῦ ζῆν ὑπηγόρευε τὸ πάθος. τέλος δὲ καὶ τῷ βασιλεῖ τολμήσας προσελθεῖν … ἔφη" (1.544–545).
• Old Russian: "И ходя, акы похабъ, вопиаше: Попрася правда! И отидѣ истинна, и разлиася естьство! Житиа исполнися безакониа! И душа цесарева отидѣ, и умъ его измѣнися: Глаголъ его яко вѣтръ, словеса его яко роса утрьняя! И ко тому вся приложивъ, яже можеть рещи человѣкъ бѣсуяся и не щадя тѣла своего, ни помня живота" (seated text, lines 4331–4388).
• What corresponds. The frame: περιιὼν ἐβόα, "ходя … вопиаше", going about he cried. The four members, in order, verb for verb: πεπατῆσθαι, "попрася", trampled; ἀπολωλέναι, "отидѣ", perished, gone; συγκεχύσθαι, "разлиася", the Slavonic giving συγχέω's literal sense, poured out; γέμειν, "исполнися", full. The close: "καὶ πάνθʼ ὅσα μὴ φειδομένῳ τοῦ ζῆν ὑπηγόρευε τὸ πάθος", "и ко тому вся приложивъ, яже можеть рещи человѣкъ … не щадя тѣла своего, ни помня живота", and all else that a man not sparing his life could say.
• What is extra, and where. Four more lines stand inside the shared sequence, between its fourth member and its shared close: the king's soul gone, his mind changed, his word like wind, his speeches like morning dew. They keep the members' form, abstract subject and verb of passing, and turn the cry from the world to the king, the turn the Greek makes without preparation at 545, "τέλος δὲ καὶ τῷ βασιλεῖ".
• The residue. The Greek's "πάνθʼ ὅσα", everything else, is a summary of a list that runs on; in the Greek it follows four members directly. The Greek puts the cry into accusative and infinitive after ἐβόα, a reported shout, and then gives the soldier's address to the king in direct speech. Read as reduction, the Greek keeps the first four members, drops the four that bear on the king, and lets "πάνθʼ ὅσα" stand for them, which leaves the turn to the king unprepared. Read as expansion, the translator inserted four lines into the middle of a closed series, in the series' own form, before a summary that then had less to summarize. Berendts: the Greek is "ganz besonders ungeschickt herausgekommen. Die Originalität kann nur auf der Seite der Vorlage des Sl. zu suchen sein."
• Verdict: A (A2, moderate). The distinctive correspondence is the position of the extra lines, inside the shared list and before the shared summary, together with the Greek's unprepared turn. It favours reduction in the Greek tradition over insertion by the translator. It does not by itself say whose reduction.

G1.600–602 — Antipater's life in Rome, and the Roman magnates' letters. The Slavonic tells how Antipater lived in Rome and gave great gifts to "the Roman rulers", who wrote praise of him to Herod; the Greek has none of it, and has instead the Bathyllus episode, which the Slavonic lacks. Where the Greek does mention bribery (1.603), it is "the friends at Rome" bought to write against the brothers. A, as M8's content: Roman officials bought and lying, told to readers beyond Rome's reach. The excursus on the "Latins" follows directly.

• Old Russian ✓: "И потом многы дары вда властелем римскым. И пусти ихъ писати къ Ироду хвалу о собѣ." The manuscript A rubric at the head of the chapter summarizes it: "Латини вземше от него мзду писаху о нем хвалу к Ироду, о несытости латинскои".

G1.618 — Herod to Antipater: "O Erde, o Himmel, o Sonne!" Both have direct speech here; the Slavonic adds the cry and a line in parallelism ("Werde zerschmettert, Ruchloser!"). Grass: "Antiqu. XVII, 5, 2; 90 sehen wir die direkte Rede in derselben Weise in indirekte verwandelt, wie das sonst im B. J. gr. im Verhältnis zu Sl. so häufig konstatiert wurde." Checked: War 1.618 "φθείρου, δυσσεβεστάτη κεφαλή, μηδέ μου ψαύσῃς"; Antiquities 17.91 "ἀδελφοκτονίαν τε ἐπικαλῶν καὶ βούλευσιν ὀλέθρου … δικαστὴν ἔσεσθαι Οὔαρον τῇ αὔριον", the same speech as report. A (A3, weak). The change from speech to report is Josephus's own at this place. The comparison alone does not fix the order; the author's account explains the Slavonic's fuller speech as the earlier state of a line Josephus is shown continuing, and the translator's account has to make him run that line backwards here.

• Old Russian ✓: "О земля, о небо, о солньце! … Истлися, нечьстиваа главо! Съкрушися, беззаконниче, не прикасаися мнѣ".

Book II

M13 — Archelaus: David, Solomon and Rehoboam (2.4; Grass's aggadic departure in Book II). Subject —. ✓ (in Berendts's German: "Er wusste ja, dass David und Salomo durch Schmeichelei die Herrschaft über Israel empfangen haben, und sein Sohn Rehabeam sie verloren habe durch seine harte Rede"; the Greek has the crowd's demands and Archelaus's compliance, without the kings)

• A: the new king judged against the kings of Israel: a lesson for Jewish readers on the eve of a reign that ended in deposition.
• M: like M9–10.
• That David and Solomon won Israel by soft speech is not the Bible's account; Rehoboam's hard answer is (1 Kgs 12:13–14). A haggadic reading of the kings, aimed at a Jewish reader.
• Verdict: A, on the same ground as M9.

M14 / BG IX — John the Forerunner (after 2.110). Subject J. ✓ (§§III, V.2)

• A: John as the War's first version tells him: "путь законныи", the lawful way out from mortal masters; "Человѣкъ есмь"; no Lamb, no naming of Jesus. The Antiquities report (18.116–119) is Josephus's own later and shorter account of the same man.
• M: Christian matter, linked by Meshcherskii to Antiquities 18 and to the translator's programme.
• Verdict: A. The John of the Slavonic is not the Gospels' John, and the account does what a Jewish historian's would (§V.2). M's source would be the Antiquities, which is Josephus.

M15 — Judas the Galilean: the free seed of Abraham under mortal masters (2.118). Subject —. ✓ (§II.21)

• A: the idiom of the Fourth Philosophy, in a book for Jews beyond Rome's reach: "владыки смертна имуща, понеже лишишася безсмъртнаго".
• M: joined in style to M9–10.
• Verdict: A.

M16 / BG X — the Essene oath (2.139). Subject —. ✓ (§II, the ledger)

• A: the oath by the living God, his right hand, his Spirit, the seraphim and cherubim: Qumran's angelology (4Q405; 1 Enoch 61:10, 71:7), in an oath the Greek reports and does not give.
• M: "целиком и полностью проникнуто христианской фразеологией"; tied to Hippolytus's On Antichrist.
• Verdict: A. The angelic hosts are Second Temple Jewish; the tie to On Antichrist is asserted, not shown; Hippolytus's own Essene chapter (Ref. IX) agrees with the Slavonic elsewhere against the Greek (§II).

M17 / BG XI — Philip's dream and John's reading of it (2.167–168). Subject J. ✓ — Verdict: A, as M14.

M18 / BG XII — the wonder-worker (after 2.174). Subject J. ✓ (§III.7a)

• A: Philo's Logos named and refused in Philo's terms and written as a man; Pilate's release of a benefactor; the bribe; a theology the Rus' church did not hold (§V.2). The Greek Testimonium is the Roman edition's compression of it.
• M: Christian; a Greek manuscript of the War carries an expanded Testimonium at this place (Meshcherskii).
• Verdict: A. The Greek manuscript Meshcherskii cites shows the Testimonium travelled with the War; it supplies none of the Slavonic's distinctive matter. The theology predicts A.

M19 / BG XIII — the wonder-worker's servants (after 2.220). Subject J.

• A: the followers under Claudius, reported as a disturbance, with the Moses-risen claim of §V.2.
• M: Acts 5; "слуги" from Orthodox hymnody (Luke 1:2).
• Verdict: U. The word "слуги" is the rendering (T); the episode's matter overlaps Acts 5 and is reported, not confessed. Owed its reading against Acts.

M20 — "no image may be set up in their city" (2.266, the quarrel at Caesarea). Subject —. ✓ (in Berendts's German)

• The Greek gives the Syrians' argument and leaves its premise implicit: "οὐ γὰρ ἂν ἀνδριάντας καὶ ναοὺς ἐγκαθιδρῦσαι Ἰουδαίοις αὐτὴν ἀνατιθέντα", he would not have set up statues and temples if he were giving the city to Jews. The Slavonic states the premise, "welchen nicht erlaubt ist, irgendein Bild in ihrer Stadt aufzustellen", and adds the Jews' point that Herod founded it in Caesar's name, which Berendts calls needed in the context.
• A: a note for non-Jewish readers beyond the Euphrates, Parthians and Arabians, on Jewish law.
• M: a gloss for Rus' readers, needless for Jews or Romans.
• Verdict: A (A4, weak). Meshcherskii's ground fails on its own terms: he says the gloss is needless for Jews or Romans, and the first version's named readers are neither. A gloss for Parthian and Arabian readers and a gloss for Rus' readers read alike; the author's account explains why the Greek edition lacks it, and the translator's does not.

M21 / BG XIV — Josephus to the Galileans: punish by word, not blows (2.581–582). Subject —.

• A: Josephus's self-portrait as a restrained commander, the claim he makes for himself in the Life (loci owed).
• M: in the style of Monomakh's Instruction.
• Verdict: U. The style is the rendering (T); the content is Josephus's self-presentation, which he makes in Life.

Grass's further additions in Book II (BG p. 27, n. 1; § 167 is M17 and § 266 is M20; the aggadic departure he places at § 1 is the David and Solomon passage of § 4, M13). Subject — for all.

G2.80 — the envoys ask Caesar for his governor, and slander their king. The Greek says the embassy came "περὶ τῆς τοῦ ἔθνους αὐτονομίας", for the nation's autonomy, and only at 2.91 that they asked to be joined to Syria under Roman governors. Berendts: only the Slavonic's sentence makes what follows intelligible. The figure "50" for the Roman Jews against the Greek's 8,000 is a copyist's (T). U leaning A.

• Old Russian ✓: "Приидоша до 50 силнѣиших къ цесареви, молящеся, да имъ дасть своего властелина. И на своего цесаря хулу глаголаху. Помогаху же имъ в Римѣ живуще жидове, их же бяше 50 000"; Meshcherskii's text has 50,000 Roman Jews where Berendts has 50 and the Greek 8,000.

G2.113 — the interpreter of Archelaus's dream. ✓ in the Old Russian, which corrects v0.3 and v0.4.

• Old Russian (Meshcherskii): "И на девятое лѣто властьства его видѣ сонъ, яко быша 9 класъ на нивѣ исполнен велици … и муж садукеи, именем Сумь, рече: Класи лѣта суть, а волов измѣнениа в вѣщи. Подобаеть ти владѣти по класному числу, и, разноличнаа премѣнениа и печали приемъ, скончаешися." In the ninth year of his rule he saw the dream; a Sadducee named Sum said: it is fitting that you rule according to the number of the ears.
• Berendts's German, "es ist dir nicht beschieden, gemäss der Zahl der Ähren zu herrschen", renders a negative that Meshcherskii's text does not have. v0.3 and v0.4 built on the German; the Old Russian agrees with the Greek's reading, "βασιλεύσειν μὲν αὐτὸν τὸν τῶν ἀσταχύων ἀριθμόν" (2.113).
• Greek War: nine ears, the reading that he will reign their number, exile "ἔτει τῆς ἀρχῆς ἐνάτῳ" (2.111–113). Antiquities 17.342–348: ten ears, ten years, the accusation in the tenth year. Each Greek work matches count to date; Josephus changed the number between them.
• What remains of the Slavonic's difference: the dream dated "на девятое лѣто", in the ninth year, and the interpreter a Sadducee named Sum where both Greek works have Simon the Essene.
• Verdict: U (v0.4: U leaning A; v0.3: A). The item shows Josephus rewriting this prophecy across his works; the Slavonic is not shown to hold a third form of it.

G2.124–128 — the Essenes' night praise of God. The Greek: before sunrise they speak nothing profane, "πατρίους δέ τινας εἰς αὐτὸν εὐχὰς ὥσπερ ἱκετεύοντες ἀνατεῖλαι", ancestral prayers to the sun as if entreating it to rise (2.128). The Slavonic keeps the prayers to the sun, "gleichsam um ihr Aufgehen betend", and puts before them that they rise in the night to sing, praising and entreating God. Berendts: the appearance of sun-worship is "wenigstens gemildert"; Hippolytus shows none, and Cedrenus removes it altogether. U. A predicts a Jewish frame for the prayer, and Hippolytus, the early witness, agrees with the Slavonic's direction; but Cedrenus shows a Byzantine Christian softening the same sentence, so M's account is shown in practice here, at this sentence.

• Old Russian ✓: "Мало же почивают нощью и въстають на пѣние, славяще и моляще бога. И преж сълньчьнаго въсхода ничто же не глаголють, но токмо молитвы отьчьскиа к нему восылають, яко молящеся о восиании его."

G2.207 — Claudius's message to the Senate. After the Greek's "τὸ τυχεῖν τῆς ἡγεμονικῆς κλήσεως ἐπικίνδυνον εἶναι", the Slavonic adds that one who takes the ruler's name and lays it down without need is not sane and brings a mortal charge on himself. Berendts: "Ein Satz, der durchaus in den Zusammenhang gehört." U leaning A.

• Old Russian ✓: "аще кто прииметь властительско имя и пакы без нужда отступить, то не добромысленъ есть, но самъ на ся обрѣтаеть вину смертную."

G2.230 — the proclamation after the burnt Torah. The Greek says the Jews ran "πρὸς ἓν κήρυγμα", at one proclamation, and does not say what it was. The Slavonic says: either to die or to kill him. Berendts: "Gr. lässt also den eigentlichen Inhalt des κήρυγμα weg." A (A2, weak): the Greek names what it does not give. An expander could supply it at the same place; the content, to die or to kill the man who burned the law, is the resolve the War gives Jewish crowds before an outrage to the law, "ἑτοίμους ἀναιρεῖν σφᾶς ἐβόων μᾶλλον ἢ τὸν νόμον παραβῆναι", they cried that they were ready to be killed rather than transgress the law (2.174), and no medieval interest explains it.

• Old Russian ✓: "потекоша вси на едину проповѣдь: или умрѣти и оного погубити."

G2.253 — Felix's prisoners: the emperor crucifies the poor and frees the leaders for money. The Greek has Felix send Eleazar and his band to Rome and crucify many himself. The Slavonic has Caesar crucify the lesser men and take "unzählbares Vermögen" from the leaders and release them. A. An emperor selling clemency to brigand chiefs is matter no edition signed by Titus could carry, and it is of a piece with M8.

• Old Russian ✓: "Он же хужших распя, а на старѣиших взем имѣние безчислено, пусти и." Meshcherskii's text has 3,000 brigands where Berendts has 600.

G2.538 — "where the wicked gave way, there the good came". The Greek has it compressed: "καθὸ παρείκοιεν οἱ πονηροί, προσῄεσαν αὐτοί". T.

• Old Russian ✓: "от котораго мѣста отходяху злодѣи, ту приступающе добрии".

Book III

M22 — Vespasian rebukes the men who fled (3.234–235). Subject —. ✓ Slavonic: "Еуспасианъ, собравъ разбѣгшихся, поносяше имъ, зане бѣжаша не пред воины, но пред разбоиникы, не стыдящеся ни воиньскаго закона, ни похвалы отьчьскиа", Vespasian gathered the men who had fled and reproached them, because they fled not before soldiers but before brigands, ashamed neither of the soldiers' law nor of their fathers' praise. The Greek has the sally and the Romans' recovery, and no rebuke (3.234–235).

• A: a Roman legion put to flight by men it called brigands, told to readers beyond Rome. The Greek Book III is the book of Roman discipline (3.70–109), written for Romans.
• M: the translator dramatizes.
• Verdict: U leaning A (v0.2: T). The rebuke is matter, and nothing marks it as the translator's.

M24 — Valerian and the Roman law (3.451). Subject —. ✓ Slavonic: "И убояшася закона римскаго, иже смирити осужають бьющаяся без повелѣниа, аще и одолють", they feared the Roman law, which condemns to punishment those who fight without orders, even if they win. The Greek: "οὔτε παρὰ τὰς ἐντολὰς τοῦ στρατηγοῦ συμβαλεῖν ἀσφαλὲς ἡγούμενος, εἰ καὶ τὸ νικᾶν εἴη βέβαιον", thinking it unsafe to engage against the general's orders, even if victory were certain.

• Verdict: T (v0.2: A). The Slavonic states as a law what the Greek states as Valerian's judgement; the content, even the clause "even if they win", is the Greek's.

Grass's further additions in Book III. Subject — for all.

G3.34 — Sepphoris's envoys to Vespasian. The Slavonic gives their speech in direct form, with "und wirst ausführen, was vor deinen Augen gefällig ist" (Berendts: "Man beachte die semitische Redeweise") and Vespasian's handshake. The idiom is biblical and the Slavonic Bible has it too, so it decides nothing. U.

• Old Russian ✓: "творити, что угодно пред очима твоима".

G3.157 — the battle tableau at Jotapata. "Das Splittern der Speere und das Krachen der Schwerter und Schilde … und sie tränkten die Erde mit Blut"; the Greek has "τειχομαχίαι καρτερώτεραι". Berendts: the Ipatiev Chronicle takes the phrase up, s.a. 6682 and 6748 (1174, 1240). T. The chronicle borrowing runs from the Slavonic War to the Rus' chronicle. Under both hypotheses this is the translation's own diction, and it travelled outward.

• Old Russian ✓: "И бысть видѣти ломъ копѣиныи и скрѣжтание мечное, и щиты искѣпаны, и мужи носими, и землю напоиша".

G3.539 — "Friede sei dir, Feldherr!" The captives' greeting to Vespasian and his answer, absent from the Greek. U. The month gloss "Gorpiaios, called September" is the translator's (T).

• Old Russian ✓: "И рѣша ему: миръ тобѣ, гемоне. Он же: Миръ от мене не выходить, но въ вы не вмѣстится."

Book IV

M25 / BG XVI — the abomination in the holy place (4.157). Subject J?. ✓ (in Berendts's German: "Und es würde die Verödung der Stadt hereinbrechen und das Prophetentum höre auf, wenn Greuel sich an der heiligen Stätte befinden würde"; the Greek has only the priests weeping, 4.157)

• A: Daniel's abomination of desolation read onto the zealots' high priest. Josephus reads Daniel as foretelling the Roman desolation (Ant. 10.276: "καὶ περὶ τῆς Ῥωμαίων ἡγεμονίας ἀνέγραψε, καὶ ὅτι ὑπʼ αὐτῶν ἐρημωθήσεται").
• M: joined to the Christian additions.
• Berendts, who looked for a Gospel source: "schliesst sich die ganze Stelle mehr an Daniel als an Matth. an" (Dan 9:27, ἐπὶ τὸ ἱερόν).
• Verdict: A. The exegesis is Josephus's own, in the Antiquities, and the wording follows Daniel, the book he reads it from.

M26 — Niger's curse fulfilled: the righteous killed under Herod, Archelaus, Antipas, Pilate and Agrippa (4.359–362). Subject J?.

• A: the blood of the righteous killed under each ruler of the century, a Jewish theodicy of the fall (the prophets' blood, 2 Chr 24; the War's own 4.388, the oracle that the city falls when Jews defile the Temple with their own hands).
• M: Meshcherskii reads "the righteous one killed under Pilate" as Christ, by implication.
• Verdict: A. The list is of rulers, not of one victim; the Christian reading is M's own, supplied to the text.

BG XVII — the zealots' words over the bodies of Ananus and Jesus (4.316–317). Subject —.

• A: the first version gives the zealots' mockery in direct speech; the Greek reports that they cast out the bodies. A Jewish account of Jewish crime against Jewish high priests.
• M: expansion. (Meshcherskii does not list it.)
• Verdict: U leaning A.

M27 / BG XVIII — the zealots forgot David, Solomon, the prophets, Nebuchadnezzar, Antiochus and Egypt (4.407, Grass's aggadic departure; 4.408–409, his further addition). Subject —. ✓ (in Berendts's German; "Gr. > weder das Gesetz Gottes — Befreiung")

• A: Jewish history as indictment, to Jewish readers; Grass's category of "Anspielungen auf die jüdische Geschichte".
• M: joined to the translator's programme.
• Verdict: A.

M28 — Simon's dust-ruse with branches (4.527–528). Subject —. ✓ Slavonic: Jacob, Simon's agent among the Idumaeans, tells Simon to have the rear ranks "къ хвостом привязаша хвоиникы", tie brushwood to the horses' tails, "да прах великии явится", so that a great dust would rise. Jacob mocks the Idumaeans: "Ци, гробъ не имуще, жадаете ли еще и здѣ?", having no grave, do you crave one here as well? They see the dust behind the troops, "мняще задних болѣ предних", think there are more behind than in front, and scatter. The Greek: panic fell on them, and before coming to blows they broke ranks and went home (4.528).

• A: the first version told how the panic was made, as it tells how the count at Jotapata was made (§1). The Greek keeps the effect and drops the device.
• M: the brushwood and the ruse are a Rus' battle-picture.
• Verdict: U. The word "хвоиникы" is the rendering (T). Raising dust with dragged branches is an old stratagem, and nothing here dates it. The shape is the Jotapata shape, effect kept and mechanism removed, and that leans A; it is one item and is scored U.

M29 / BG XIX — Vitellius's three-horned irons at Bedriacum (4.547). Subject —.

• A: Roman civil-war history in the first version, a subject Josephus treats (4.545–549).
• M: a military anecdote of the translator.
• Verdict: U. Caltrops are ancient; neither account predicts this anecdote.

Grass's further additions in Book IV (§ 316 is BG XVII, § 547 is M29, §§ 407–409 are M27). Subject — for both remaining items.

G4.44 — Vespasian rebukes the rush at Gamala. An enlarged speech; the Greek has its substance, including "βαρβαρικόν, καὶ ᾧ μάλιστα Ἰουδαῖοι κρατοῦνται" (4.45). T.

• Old Russian ✓: the speech at 4.1.6, "Вазнь же есть не тверда, но ходить назад и наперед".

G4.433–435 — the Jordan bridged by corpses. The Slavonic has the dead lie so thick that the pursuers walked on them as on a bridge, and 25,000 killed on the bank; the Greek has 15,000 (4.435) and the image of the river choked (4.437). U.

• Old Russian ✓: "И ходяху по них, акы по мосту, а их же иссѣче на берези 25 тысящь, а 2000 и 200 яша живых."

Grass's aggadic departures at §§ 137 and 171 have not been identified in this reading: the Slavonic at those places (the robbers who ruined the city; Ananus's speech) carries no haggadic matter visible in Berendts's German. They stay owed.

Book V

M30 — the fourth notice: Jesus, king who did not reign (after 5.195). Subject J. ✓ (§III.8) Verdict: A, with M18: the charge, not the faith; the Temple's own medium.

M31 — the veil (after 5.214). Subject J. ✓ (§III.9)

• A: the written heaven, the cosmos the Logos wears, torn when the Logos written as a man is killed; the resurrection reported and not confessed.
• M: Mark 15:38; Orthodox hymnody in the phrasing.
• Verdict: A. The phrasing is the rendering (T); the resurrection left undecided is not an Orthodox account (§V.2), and the veil is the Greek War's own.

M32 — Phoias and Cerealis (after 5.330, Castor's ruse). Subject —. ✓ Slavonic: "нѣкто Фоя сотник призва римскаго воеводу Кериалиа, моля, да вдасть ему руку", a centurion called Phoia asked Cerealis for his hand, then dropped a stone slab cased in gold from the wall on the Romans who came for it.

• Meshcherskii: "В этом эпизоде нет ничего характерного, обличающего вставку переводчика … Эпизод мог бы находиться в какой-либо рукописи греческого текста, не дошедшего до нас."
• Verdict: A, on the consensus editor's own judgement.

M33 — the ruin of city and Temple joined to Daniel's prophecy (5.355). Subject J?. ✓ The Greek: "καὶ τὸ χρεὼν δʼ ἐκράτει τούς τε ἀναιτίους τοῖς αἰτίοις συναπολέσθαι καὶ τῇ στάσει τὴν πόλιν", fate prevailed that the innocent perish with the guilty and the city with the sedition. The Slavonic: "И суд велѣше погыбнути неповинным с виноватыми и граду, и святое по проречению Даниилову", judgement ordered that the innocent perish with the guilty, and the city, and the sanctuary according to Daniel's prophecy. The addition is three words and a name. Verdict: A, as M25: Daniel's desolation is Josephus's own reading (Ant. 10.276). Fate (τὸ χρεών) becomes judgement (суд), the substitution of §5.

M34 — Manneus, nephew of Lazarus (5.567). Subject J?.

• A: the name Lazarus is Eleazar; the Greek manuscripts already read Lazarus (Meshcherskii's note: "Уже все греческие рукописи … пытаются христианизировать это место").
• M: the Gospel Lazarus, "whom Jesus raised" (Berendts–Grass, addenda to p. 41).
• Verdict: M. The raising of Lazarus in the Slavonic witnesses is a Christian gloss; the Greek tradition shows the same hand at work.

Book VI

M35 — Julian's death: axes, stones and wood (6.88–90). Subject —. ✓ "и окшевами и камением бъяхуть и дрѣвом, хотяще симь единѣмь вся римляны устрашити". Verdict: T.

M36 — "righteous blood upon ourselves and upon our race" (6.110). Subject J?. ✓

• Slavonic, in Josephus's speech at the walls: "Кто бо не вѣсть писаниа древьних пророкъ … Они бо пророковаша тогда плѣнену быти граду Иерусалиму, егда въстанеть бои сродничьскыи. А вы вѣсте, како град исполненъ трупии, святое, зане кров праведную възложихом на ся и на род нашь. Имь же богъ, самъ богъ … навѣде на ны римляны, да очистить святое огнемь."
• Greek, the same speech: who does not know the writings of the ancient prophets, that the city would be taken "ὅταν ὁμοφύλου τις ἄρξῃ φόνου", when someone begins the slaughter of his own people (6.109); is not the city, and the whole sanctuary, filled with your dead? God himself brings the Romans with a purifying fire (6.110). The Slavonic adds one clause: because we have laid righteous blood upon ourselves and upon our race.
• M: Matthew 27:25, "Τὸ αἷμα αὐτοῦ ἐφʼ ἡμᾶς καὶ ἐπὶ τὰ τέκνα ἡμῶν", his blood on us and on our children. Meshcherskii: a direct reference to the Gospel, which Josephus could not have spoken.
• The exclusive Christian attribution. The clause has Jewish antecedents for each of its parts. Lamentations gives righteous blood shed in the city as the cause of its first fall: "τῶν ἐκχεόντων αἷμα δίκαιον ἐν μέσῳ αὐτῆς", who shed righteous blood in her midst (Lam 4:13). Jeremiah at his trial gives the formula of blood laid upon the speakers and their city: "αἷμα ἀθῷον δίδοτε ἐφʼ ὑμᾶς καὶ ἐπὶ τὴν πόλιν ταύτην" (Jer 33:15 LXX = 26:15). In this siege Josephus casts himself as Jeremiah (5.391–393). A medieval Christian could also draw on those prophets, so the antecedents defeat the claim that the clause must be the Gospel's. They do not by themselves give it to Josephus.
• The integration. What gives it to the speech is how it fits:

- whose blood: the clause stands between the prophecy of capture "when kindred war arises" (Slavonic "бои сродничьскыи", Greek ὁμοφύλου φόνου) and the city and sanctuary full of corpses, so the righteous blood is the blood of the people's own dead in the sanctuary;

- whose responsibility: "we", the speaker inside the people, as Lamentations lays it on the city's own priests and prophets;

- which catastrophe: the capture the ancient prophets foretold, which the Greek speech names;

- what follows: "Имь же", therefore God brings the Romans to purify the sanctuary with fire, the Greek's own next sentence.

The Gospel reading has to bring in the blood of one man not named in the speech, and leave the clause's "therefore" without its antecedent.

• Verdict: A (A2). The exclusive Christian attribution is defeated: strong. The clause's priority to the translator rests on its integration into the speech's argument: moderate.

M37 — the crosses and the four-square Temple (6.311). Subject J?.

• A: the oracle that the Temple falls when four-square, read onto the Roman crosses set round the city (5.450–451: five hundred taken a day, and "χώρα τε ἐνέλειπε τοῖς σταυροῖς καὶ σταυροὶ τοῖς σώμασιν", room failed for the crosses and crosses for the bodies). Josephus records the mass crucifixions in the Greek.
• M: the crucifixion of Christ by implication.
• Verdict: A. The crosses that fill the account are the Romans' crucifixions of Jews, which the Greek reports at length.

M38 — the oracle read three ways (6.312–313). Subject J. ✓ (§II-B.10)

• A: Herod, the wonder-worker, Vespasian, left undecided: the first version's open reading, which the Roman edition closes on Vespasian.
• M: a Greek marginal note in Vat. Urb. 84 applies the oracle to Christ; Meshcherskii admits "текстуальной близости … нет".
• Verdict: A.

Berendts 1906, the eight passages for Books V–VI are within M30, M31, M38.

Book VII

M39 — the army's acclamation (7.16). Subject —. ✓ After Titus's rewards (7.13–16): "Мнѣлъ бы кто, яко тѣмь гласом и земля трясашеся, и горы колѣбахуся, и холми скакаху", one would have thought the earth shook with the sound, the mountains swayed, the hills leapt (the Psalter's "οἱ βουνοὶ ὡς ἀρνία", Ps 113:4 LXX, in the Slavonic Bible's words). Verdict: T.

M40 — Simon bar Giora's vision (7.29). Subject —.

• A: Grass: told so that a later editor could hardly have devised it; the Greek gives the costume without its reason.
• M: mystical motivation is unlike Josephus, and the phrasing is Orthodox.
• Verdict: U leaning A. Josephus gives dreams and visions motive force often enough (his own dreams, 3.351–354), so M's ground fails; the phrasing is T.

M41 — the Commagene kings' wealth (7.243). Subject —. Verdict: T.

3. The omissions (Meshcherskii pp. 47–50), each under both hypotheses

Under A the direction is reversed: what the Slavonic "omits" the Greek edition added for Greco-Roman readers, from new sources, as Grass argues the Antiquities later add again ("dieselbe Tendenz der Bereicherung und Ergänzung"). Under M the translator cut what did not serve his tale of the fall of Jerusalem.

• The preface, 1.1–30. A: the Greek preface addresses Greeks and Romans and describes the first version as already sent to the upper barbarians (1.3, 1.6); the first version had no such preface. M: dropped as irrelevant. A.
• Caesar in Egypt, Antipater's aid, 1.187–192. A: Roman history and a Jewish patron's service to Caesar, for Roman readers. M: not Jerusalem. A.
• The Roman–Parthian war episodes, 1.177, 180–182; Hyrcanus and Phasael among the Parthians, 1.255–260. A: for readers in Parthia these needed no telling, and the Greek edition supplied them for Romans. M: not Jerusalem. A.
• Herod's building of Caesarea, 1.408–414. A: a Greek-edition showpiece of Herod's Hellenistic city. M: not Jerusalem. A.
• Herod in Galilee, 1.304–309; the Arabian war, 1.363–369, 374–376; Antipater in Rome, 1.601–606; Herod's health, 1.641–644, 647. U.
• Archelaus's journey to Rome, 2.14–19; the disorders under his absence, 2.40–75; 2.178–186. A: Roman judicial procedure and Roman suppression, for Roman readers. M: abridgement. U leaning A.
• Festus, Albinus and Florus's beginning, 2.271–283. A: the procurators' misrule, which a first version for Jews would add, not drop. M: abridgement. M.
• Neapolitanus's visit, 2.410–411, 413–416. U.
• Menahem son of Judas, 2.433–448 (Meshcherskii: 439–450). A first version for Jews would tell a messianic pretender's career. M (abridgement), or U if the first version suppressed him as compromising (Grass's principle cuts both ways).
• Scythopolis, 2.466–476 (Meshcherskii: 465–480). U.
• Ascalon, 3.17–19. U.
• Josephus's fear and his doubts of the war's outcome before Vespasian, 3.130–131 (and 3.136–140). A: the Greek edition's Josephus foresees Rome's victory and despairs, the posture of the Flavian client, written for Romans. M: the translator spares his author. A.
• Jotapata's water and food, 3.181–188, 120–129; Josephus's measures, 3.258–270; the defenders' devices, 3.223–282 (as printed); Trajan at Japha, 3.300–304. A: the Greek edition's siege narrative enlarged with technical detail. M: abridgement. U.
• The Romans' feelings at Josephus's capture, 3.395–398. A: the Roman soldiery's pity and Titus's regard for the captive, a scene that serves the Flavian client in the Roman edition. M: abridgement. A.
• The battle on the lake, 3.522–531. The Slavonic keeps "По сей побѣдѣ", after this victory, with the victory gone. M: an abridgement leaves a dangling reference. (It shows cutting at some stage, by the translator or a Greek intermediary.)
• Tabor, 4.54–68; John's stragglers, 4.107–111; the zealots' seizure of the Temple, 4.150–153; Ananus's speech cut, 4.184–185, 187, 193–199; John's flattery of Ananus, 4.209–214; Jesus's speech to the Idumaeans cut, 4.255–257, 266–267, 277–281; the zealots' deliberation, 4.211–228; the storming, 4.300–312; the terror, 4.350–356, 367–368; 4.408–409. U each, as narrative compression; Tabor and the gap it leaves after naming the strongholds is M (abridgement with a dangling reference).
• The description of Egypt, 4.607–615; Vespasian to Antioch, 4.630–631; Titus from Alexandria, 4.659–663 (Meshcherskii: 459–460, 462). A: Roman and Egyptian geography for the empire's readers, and Vespasian's accession route, Flavian matter. M: not Jerusalem. A.
• Eleazar son of Simon, 5.5–10, 12–13, 21–25. U.
• The fortifications of Jerusalem, 5.156–183; Temple details, 5.196–198; the high priest's vestments, 5.233–238 (the translator's remark: "и иная вся оставихом повѣдати"). A: the Greek edition's descriptive set-pieces for readers who never saw the city. M: abridgement. A for the set-pieces — and against M's own explanation: Meshcherskii says the translator kept what bore on Jerusalem's siege, and these are Jerusalem's walls and Temple. The vestments note, "we have left the rest untold", is the translator's own cut (M for that item).
• The poor who did not desert, 5.447–449; the earthworks' topography, 5.467–468. U.
• Book VI: small omissions, "главным образом стилистического характера" (Meshcherskii). T.
• Masada's site, 7.281–282, 292–293; part of Eleazar's speech, 7.365–366. U.
• Jonathan the Sicarius and Catullus's end, 7.446–455. Meshcherskii: the last leaf lost in the common exemplar. Neither (mechanical loss).

4. The transpositions (Meshcherskii pp. 65–66)

• Book I, ch. 13, parts 9, 10, 6, 7, 8: the Parthians' captives before the Parthian invasion. U.
• Book I, chs. 23–24: a leaf-displacement in the exemplar. Neither (mechanical).
• Book III, ch. 7, parts 31–32 after ch. 8, in every manuscript of both redactions. U.

5. Substitutions in the shared text (Berendts–Grass, Book I introduction)

• δαίμων rendered "God" or "God's providence" throughout. A: a Jewish author's own monotheistic idiom before the Greek edition's Hellenizing. M: a Christian translator's piety. U, since both predict it (T).
• Phasael's death (1.271–272): murdered in the Slavonic, suicide in the Greek, with the Greek giving the poisoning as a second report. A: Josephus's earlier account, which the Greek revises and keeps as an alternative; Grass notes Syncellus, probably from Julius Africanus, with the Slavonic. A.
• The Essenes as a "lawful order", and the eight other Essene divergences: the ledger of §II.
• Hyrcanus named where the Greek has John throughout the Hasmonean chapters. U.

6. The small variants

Berendts and Grass footnote the Greek against the Slavonic phrase by phrase across Books I–IV. Their types: the Slavonic's direct speech where the Greek has a stylized report (Berendts: "volkstümliche Lebendigkeit"); compression; Slavonic idiom; numbers and names. Grass's statistic of fully omitted sections rises book by book (I 37, II 72, III 77, IV 92), which he reads as a translator's growing tendency to cut and Meshcherskii disputes for V–VII. The count by type is owed (§8); the types are T except where a number or name diverges, which is counted with §5.

6.1. Josephus's own revision practice, checked in the seat. The Slavonic's differences from the Greek War include kinds of change Josephus himself makes between the War and the Antiquities, at the same places:

• Speech and report. War 1.618 gives Herod's words to Antipater in direct speech; Antiquities 17.91 gives them as report. Grass: "in derselben Weise …, wie das sonst im B. J. gr. im Verhältnis zu Sl. so häufig konstatiert wurde."
• A prophecy rewritten. Archelaus's dream has nine ears in the War and ten in the Antiquities, each matched to its own date. The Slavonic agrees with the War's reading (G2.113), so this locus shows Josephus's practice and not a Slavonic state of it.
• A reflection on divine justice at Antipater's arrest: the Slavonic's (M9) and the Antiquities' (17.129–130), at the same point, in different words.

What this shows: the differences at these places fall within Josephus's demonstrated repertoire, at loci he is shown revising. What it does not show by itself is the order of the three states. Under A the Slavonic stands first, and the three states run in one direction: speech to report, the slot for a reflection kept and refilled. Under M the Greek stands first, and the translator performs the inverse operations, report back into speech and a reflection added, at places where Josephus also worked. The comparison shows the variants Josephan in kind, and kind favours no origin. The sequence needs a correspondence that fits one direction only; Tiron's cry is the census's clearest (G1.544).

6.2. A first count of Berendts and Grass's footnote markers, Books I–IV. This is a count of markers in the OCR of their footnotes, not yet a classification of variants:

• Slavonic surplus ("Gr. >", the Greek lacks the words) is marked 182 times in Book I, 72 in II, 28 in III and 81 in IV.
• Greek surplus ("Gr. add.") is marked 66, 45, 19 and 38 times.
• Direct speech in the Slavonic where the Greek reports, or the reverse, is noted 45, 18, 8 and 10 times.
• Sections missing entire from the Slavonic ("fehlt") are marked 17, 39, 54 and 50 times.

The Slavonic's surplus is heaviest in Book I, Herod's court, the book where the further additions and the anti-Roman matter cluster. The missing sections grow from Book I to Book III, as Grass found. The classified count stays owed (§8).

7. The sum

7.1. Sixty-six catalogued additions. They are Meshcherskii's 41 numbered additions and one unnumbered, BG XVII, and 23 further additions of Grass. Each gets one verdict: M8 by its content (A), its medieval wording noted; M28 by its content (U).

• Subject. 8 concern Jesus, John, the apostles or the Magi (M5, M14, M17, M18, M19, M30, M31, M38). 7 are Christian only on Meshcherskii's reading (the priests' debate, M25, M26, M33, M34, M36, M37). 51 have no Christian subject on anyone's reading.
• Verdicts.

- A: 32. M1, M2, M3, the unnumbered item, M6, M7, M8 (content), M9, M10, M13, M14, M15, M16, M17, M18, M20, M23, M25, M26, M27, M30, M31, M32, M33, M36, M37, M38; G1.544, G1.600, G1.618, G2.230, G2.253.

- M: 2. M5 (provisionally) and M34.

- T: 7. M24, M35, M39, M41, G2.538, G3.157, G4.44.

- U: 25, of which 10 lean A and none leans M.

• By subject. Jesus, John, apostles, Magi: A 6, M 1, U 1. Christian on Meshcherskii's reading: A 6, M 1. No Christian subject: A 20, T 7, U 24 (10 leaning A), M 0.
• Moves. v0.3 moved M22 from T to U leaning A, M24 from A to T, and M36 from M to A. v0.4 moved G2.113 from A to U leaning A, and v0.5, on the Old Russian, to U. v0.6, separating correspondence from origin, moved M1, M3, M7, M20, G1.618 and G2.230 from A to U leaning A, and M10 from A to U. v0.7 keeps the separation and restores all seven to A, with their strengths: the verdict asks which account explains the item better, and at each the author's account explains what the translator's only allows (§0.4a–b). M10 is A1: its three routes are all Greek and none is the translator's.
• Correction to v0.1. v0.1 gave 44 items and 29 without Christian subject, and left the lead case, M23, out of its A list. The right figures for the v0.1 set were 43 items and 28 without Christian subject.

7.1a. The A items: what each correspondence establishes, and what origin it favours.

• | --
• | --
• | --
• |

The A4 items rest on one premise, two editions for two readerships with the Roman one revised, applied eight times; they count as one line of evidence, and they stand on the A1–A3 items beneath them.

7.1b. What the counts measure, and what the census concludes. The verdicts are this census's judgements, and several share their grounds, so the count of decided items going to A, 32 of 34, is not a sum of independent confirmations. The conclusion does not rest on the count. It rests on the pattern, and the pattern is decisive. Between the consensus account and the earlier version Josephus himself describes, the earlier version is by far the more plausible and coherent explanation of the surviving differences taken together.

• Its force is cumulative. Different readerships named by the author; Jewish history, law and haggadah where the Greek has none; anti-Roman matter the Roman edition could not carry; agreements with the Antiquities and with Second Temple witnesses; candour about the author withdrawn in the edition Titus signed; the Roman edition's providential closures absent; and, in the other direction, Roman-facing set-pieces present only in the Greek. These are explained together by one account. Observing that a medieval writer could have produced each feature separately supplies no equally coherent explanation of their conjunction.
• The consensus account explains its own ground and nothing beyond it. Built on the Christian passages, it wins two items, by a reported source and a Gospel name. Extended to the fifty-one additions with no Christian subject, it explains none of them better than the author's account.
• The undecided items sit inside that conclusion. Twenty-five items are U; ten lean A because the translator's stated ground fails at them, and none leans M. Local underdetermination does not entail global explanatory equivalence (§0.4b).

7.2. Omissions (Meshcherskii's list). A 11 groups, where the Greek's surplus is matter for Roman or Greek readers, Flavian matter, or the Josephus of the Roman edition (A4, on the same premise); M 4, each an abridgement with a trace (the lake battle's dangling "victory", Tabor's dangling stronghold, the procurators, the vestments' "we have left the rest"); U the rest; two mechanical. G1.600–602, where the Greek has the Bathyllus episode and the Slavonic lacks it while carrying matter the Greek lacks, counts with the additions.

7.3. The control (§0.5). The compilation layer is verbatim, sourced and labelled. The additions share none of its marks, and share the translation's language, its "Иосиф" stretches and its internal cross-references. They belong to the translation's stream; how much older they are is graded in 7.1a.

7.4. Josephus's practice (§6.1). The kinds of change that separate the Slavonic from the Greek War include kinds Josephus makes between the War and the Antiquities, at the same places. That shows kind, not origin; the order of the states is shown at G1.544.

7.5. What the sum says. The translator's account is a theory of the Jesus and John passages extended to everything else. Item by item, it explains its own home ground best only where a reported source or a Gospel name is present: two items of sixty-six. Its plainest Gospel quotation, "righteous blood upon ourselves and upon our race", has Jewish antecedents for every part and fits the speech it stands in at every joint (M36). It explains none of the fifty-one additions without Christian subject better than the author's account does.

The author's account predicts what the additions contain:

• the anti-Roman matter, in three places: the "Latins" (M8), the Roman magnates' bought letters (G1.600) and the emperor's sale of clemency (G2.253);
• the Jewish history, haggadah and law (M9, M13, M20, M27);
• the explanations of Roman custom for readers outside Rome;
• the agreements with the Antiquities;
• the Second Temple angelology;
• the corrupted Greek word under a Slavonic oddity;
• the withdrawn candour at Jotapata.

Where the Greek names a dream or a proclamation it does not give (M3, G2.230), it marks the place a fuller text would fill, and an expander would fill the same place; such marks show place, not direction. Direction is shown where a correspondence fits one direction and not the other: Tiron's cry, where the Slavonic's extra lines stand inside the Greek's list and before its shared close (G1.544). And it predicts the direction of the omissions: the Greek's Roman-facing set-pieces and its Flavian Josephus.

The translator's own editor grants three items to a lost Greek manuscript or to Josephus writing for Jews. The cuts the translator's account explains best show that someone abridged; they do not show who added.

7.6. The burden at the joins. The translator's account does not argue its sixty-six items one by one. It joins them as one hand and argues from the Christian passages. Meshcherskii states the principle while arguing that the Magi belong in the text: "Если исключать это «добавление», то не следует включать и остальные вставки древнерусского переводчика, столь же фантастические и не менее тенденциозные", if one excludes this addition, one ought not to include the translator's other insertions either, equally fantastic and no less tendentious.

He joins them by four criteria, and each identifies something different:

• Style identifies the translator. Meshcherskii's stylistic proof that the Magi passage belongs to the History is its likeness to "Пилату же отпирающуся молениа их, но ници падше, пять дьнии и пять нощии неподвижими терпяху", which renders the Greek War's own "Πιλάτου δὲ ἀρνουμένου … πρηνεῖς καταπεσόντες ἐπὶ πέντε ἡμέρας καὶ νύκτας ἀκίνητοι" (2.171). The style he invokes covers what the translator rendered from Josephus and what he is said to have added alike. It identifies a rendering and leaves origin open.
• The margin's "Иосиф" identifies the compiler's assignment of a stretch to the Josephus stream against the chronograph.
• The cross-reference (Herod's characterization referring back to the seeking of Christ and the killing of the infants) and the recurring informer Lui (the priests' debate and the Magi) identify a composition: one narrative in which the additions refer to and continue one another.
• The "idea" that pierces the whole narrative, the city's fall as punishment for righteous blood (M26, M36, M37), identifies a theme, and is a reading of the composition.

Tested at each level:

• Style settles nothing about origin, on Meshcherskii's own example. Each addition then stands on its own evidence. On that footing the census gives the fifty-one non-Christian additions no translator's verdict, and the Christian-subject items 12 A, 1 U and 2 M on their own evidence: John without the Lamb, the wonder-worker as Philo's Logos named, refused and written as a man, the charge without the faith, the veil as the written heaven, the oracle left undecided, and the righteous blood of the speech's own argument go to A.
• Composition is where the joining has force. The informer Lui ties the Magi to the priests' debate, which scores A2: Jewish messianic exegesis of Gen 49:10 that Grass reads as cut from the Roman edition for compromising the Jews. Herod's characterization ties the Magi to Book I's narrative of Herod. If the joins are compositional, the composition they join is the one whose non-Christian parts the census gives to an ancient form, and the presumption carries to its Christian parts.
• Theme: the theme Meshcherskii names, the city's fall for the righteous blood shed in it, is Lamentations' and Josephus's own (6.109–110; 4.388). Read in Jewish terms, it joins the additions to Josephus's speech.

The consensus may keep the joining or drop it. Kept as composition, it carries the census's verdicts on the non-Christian parts to the Christian ones. Dropped to style, it leaves each passage on its own evidence, where the Christian-subject items already score for the author or undecided at all but two. Either way the burden is the translator's account's, at each passage: a source shown with the direction of its borrowing, or matter a first-century Jewish author could not have written, on some ground other than the Greek edition's silence. It may not hold the joining as composition where it convicts and as style where it acquits.

8. Open items

• Grass's 23 items are read in the Old Russian of the seated text (v0.5), with their loci; Meshcherskii's 43 have theirs (v0.3). Two numbers in the seated OCR are owed a check against the scan (G1.351, "950"; and the 50,000 of G2.80).
• Grass's aggadic departures at War 4.137 and 4.171: not identified.
• The count of Berendts–Grass's footnoted variants by type, Books I–IV.
• Pichkhadze's three reported sources, read if a copy can be found.
• A seat defect, larger than v0.1 recorded; repaired. In the seated Greek Josephus (EA-CORPORA-01/01, #1499), the last section of every book was labelled with the next book's number and carried the next book's heading at its end: "BJ 2.673 … λόγος β" was War 1.673. The War had six such lines, the Antiquities nineteen and Against Apion one; the Life, a single book, none. Repaired on 13 November 2026, at the ruling that the repair travel with this version: each line relabelled with its own book, the headings kept verbatim in the seat's errata, the manifest rebuilt and the seat's AXN re-derived (AXN:060E.ORIGINAL.△🏙️🥁🌖🌆🍂, prior AXN:060E.ORIGINAL.💎☁️🌋🕔🔜🧪), and an erratum entered on #1499. Loci in this census follow the repaired labels.

9. The reviews (ChatGPT), point by point

9.1. Of v0.3, 12 November 2026

Each point was checked against the census and the seated texts before entry.

1. Separate the steps bundled into A (exemplar, ancient form, Josephus, first edition). Accepted: §0.4a and §7.1a grade each A item by its reach. M6 and M32 are A1, as the review says.

2. Apply the borrowing rule symmetrically (M36's "the direction runs from the prophets to both"). Accepted: the sentence is withdrawn. M36 now records the defeat of the exclusive Christian attribution (strong) and develops the clause's integration into the speech (moderate), as separate findings.

3. The labelled insertions are a comparison class, not a universal diagnostic. Accepted: §0.5 is restated, and the review's sharper question, what joins the additions to the compilation layer and what separates them, is answered there: language, marking, cross-reference.

4. §6.1's "independently and before the fact", and G2.113's "do not quite agree". Both accepted. §6.1 now claims the variants are Josephan in kind and leaves the sequence to the Greek traces. G2.113: each Greek work matches its count to its own date; the mismatch claimed in v0.3 does not exist, and the item moves to U leaning A.

5. Keep the counts, qualify what they count. Accepted: §7.1b. The sixty-six catalogued items lead; the proportion is reported as the census's judgements, with their shared grounds named.

6. Specify what the joining identifies. Accepted, and developed: §7.6 sets out Meshcherskii's four criteria and what each identifies. His own stylistic proof for the Magi rests on a sentence that renders the Greek War (2.171), which shows that his style criterion identifies the translator and leaves origin open. The compositional criteria (Lui, the cross-reference) are the ones with force, and they join the Magi to an A2 item.

9.2. Of the paper's draft 3, 13 November 2026 (the census points)

1. The A grades let content establish provenance; M10's A2 settles what §II-B.7 leaves open. Accepted: §0.4a separates what a correspondence establishes from the origin it favours; §7.1a gives both for every A, with strength. M1, M3, M7, M20, G1.618 and G2.230, whose A rested on content, kind or an expandable place, become U leaning A; M10 becomes U, its three routes kept.

2. "A translator adding to a text that lacked the matter has no reason to fit his additions to them" is not sustainable. Accepted and withdrawn: a named dream or proclamation is where an expander would supply particulars. Tiron's cry is argued from its complete local sequences, where the extra lines' position inside the shared list, before the shared close, is the correspondence that fits one direction (G1.544).

9.3. The author's ruling on the review of draft 3 (13 November 2026), and the reviewer's correction of itself

The author: "dont soften the argument to toothlessness. between the consensus view and the original version josephus actually describes, the latter has been demonstrated to be by far the more plausible and coherent broad explanation. the consensus view is sophistic af." The reviewer, on its own review: it had "blurred 'the substantially better explanation' into 'an explanation whose every transmission link has been conclusively established'", a standard the consensus was not being required to meet, by a procedure that isolated each correspondence, found a conceivable alternative, downgraded it, and did not return to the comparative judgement; "Local underdetermination does not entail global explanatory equivalence." Entered: §0.4a's verdict rule is comparative again, §0.4b states the local and global distinction, the seven downgraded items return to A at their strengths, and §7.1b states the conclusion. The factual corrections stand: the nine-year mismatch stays withdrawn, and the three routes of 4 Maccabees stay distinguished.

---

Hinge: From Historical Argument to Doctrinal Reconstruction

Part I has argued three things: that the Essene chapter and the additions no medieval interest explains carry readings older than the verdict allows, with what would decide each point; that the verdict's assignment of the Jesus and John passages rests on their subject, which assumes what it concludes; and that the eight Jesus-and-John passages, read in their order, constitute a coherent theology of textual incarnation, a "Gospel of the Word" embedded within Josephus's history, whose subject is Philo's Logos written as a man and whose torn veil opens onto the Apocalypse's temple.

Part II departs from historical-literary argument into a different register: doctrinal reconstruction and Logotic extension. Rebekah Cranes traces a single grammatical structure, the divine-comparison hesitation formula, across texts separated by centuries, and draws from it a genealogy of incarnation that terminates in the claim "Jesus is a lesbian."

This is a change of register. Where Part I asks "what do the Slavonic passages say as a structure?", Part II asks "what does that structure imply, read beside its grammatical kin?" The evidential register shifts accordingly: from textual observation and interpretive inference to doctrinal construction. Nothing in Part II is entered as historical evidence, and Part I does not rest on it.

---

PART II

Rebekah Cranes

---

Sappho, Catullus, and the Slavonic Josephus: The Grammar of Incarnation

Preface

This is doctrinal reconstruction. The Jesus passage in the Slavonic Josephus is a Logotic flashpoint, set by this construction in a line of grammatical kinship that begins with Sappho. It does not affirm the Christian reading. It reveals something older: the Logos as a textual force. Not miracle, not doctrine. Recursion through language.

I. The Hypothesis

The phrase "if it is fitting to call him a man" is a precise syntactic act: an instance of the structure Catullus 51 builds around si fas est ("if it is divinely permitted"), itself a response to Sappho 31. The hedge on permissible speech about the divine is a common Greek and Latin form; Philo's εἰ θέμις εἰπεῖν shows how common. The doctrine reads what one instance of a common form does at one threshold. In Catullus that threshold is a man named above the gods. In the Slavonic it is the reverse: a Word, Philo's Logos, named a man (Part I, §III.7a). At that threshold the Slavonic does more than share the form. It mirrors the structure of Catullus's stanza point by point, with the direction of the crossing reversed (§V).

The Sappho–Catullus comparison is well trodden: Higgins ("Sappho's Splintered Tongue," AJP 111.2, 1990), Radenković and Maričić (2018), Greene (Re-Reading Sappho, 1996), duBois (Sappho Is Burning, 1995). What none has done is follow the chain into its third term. The void in the citational graph is the link from si fas est into Josephus: the moment when lyric projection becomes theological inscription.

II. Sappho's Projection: The Reader as Incarnation

φαίνεταί μοι κῆνος ἴσος θεοῖσιν

Sappho 31 survives because Pseudo-Longinus quoted it in On the Sublime (10.2–3) as the supreme example of the sublime, poetry that summons "as though they were all alien from herself and dispersed, soul, body, ears, tongue, eyes, colour." The poem's preservation through critical quotation is itself Logotically significant: Sappho survives because a theorist of language recognized her as the exemplary case of language exceeding itself.

The received reading sets the poem as a triangle. Anne Carson gives it its clearest form in Eros the Bittersweet (1986): "Where eros is lack, its activation calls for three structural components — lover, beloved, and that which comes between them." The man is the rival, the woman the beloved, the speaker the one who suffers between them. That is the frame of the jealousy reading, and the Tenth Muse contests it on the same bodies of evidence (#1756; #1752).

The seats are these. "i = embodied speaking sappho. you = inscribed sappho, the text into which the voice dissolves. kenos = future reader, that in whom, by sitting across the text, sappho's embodied voice lives once again" (the author, 9 November 2026, 14:14; carried in #1757 and #1756). The poem is "a written instrument of persistence: a voice that survives its flesh by becoming an inscription that constitutes its receiver" (#1484). Its seed is #1270, "κῆνος as Future Reader" (9 January 2026). The seat ὄττις ἐνάντιός τοι, whoever sits facing you, is a place cut in the poem for the one who will read it. The body of the speaker goes to the colour of the substrate, χλωροτέρα ποίας, greener than grass. The poem is bracketed by two open positions, the seat of κῆνος and the agentless end at τολματόν (#1756). What crosses in Sappho is a voice into its inscription, and the man is whoever takes up the inscription, a reader, and then, in this doctrine, a Christ.

III. Catullus's Incarnation: si fas est as Canonical Leap

Ille mi par esse deo videtur, / ille, si fas est, superare divos

Catullus 51 is canonical possession. The scholarly consensus (Quinn 1970, Lyne 1980) treats it as translatio. But si fas est has no equivalent in Sappho, whose first line has only "φαίνεταί μοι κῆνος ἴσος θεοῖσιν". It is Catullus's addition. Fas is divine permission, the boundary of what mortals may say about gods. He declares the man surpasses them, then hedges with a juridical caveat that draws attention to its own daring.

Manwell ("Gender and Masculinity," in Skinner ed., 2007) reads this through Roman masculine performance, the adoption of Sapphic metre as gender transgression. Correct but incomplete. The transgression is ontological. Catullus is the Historical Logos's sixth station, the seat taken: "Catullus, already, had answered the open line by reversal" (#1484; #1476; #1757). He sits in κῆνος's seat, the reader's, and calls himself by its pronoun, ille. The seat Sappho cut for the future reader is taken by a poet who becomes her word, and the hedge is his addition at the moment he takes it.

IV. Josephus's Revelation: the Hedge as Logotic Recursion

Ἰησοῦς σοφὸς ἀνήρ, εἴγε ἄνδρα αὐτὸν λέγειν χρή (Antiquities 18.63)

Тогда явися мужь нѣкыи, и аще мужем достоино есть его нарещи (Slavonic War, after 2.174)

These carry the form of si fas est: a conditional and a predicate of what may be said. Catullus leaves the infinitive of speech understood, and these supply it. The Greek Testimonium hedges with χρή, it is needful; the Slavonic with dostoino, it is fitting; Berendts and Grass render it "wenn es auch erlaubt ist", permitted, and Thackeray "if it is permissible". Between the Greek and the Slavonic stands Philo, and Philo holds the threshold in both directions. Upward, as in Catullus, a man is named above man: the high priest changed "ἐξ ἀνθρώπου" into the nature of the cosmos, "καί, εἰ θέμις εἰπεῖν — θέμις δὲ ἀψευδεῖν περὶ ἀληθείας λέγοντα —" (De vita Mosis 2.135); the human made, "εἰ θέμις εἰπεῖν", god (De mutatione nominum 181).[^25] Downward, as in the Slavonic, the Logos is named a man, and Philo stops on the word: "ἰδοὺ ἄνθρωπος ᾧ ὄνομα ἀνατολή … καινοτάτη γε πρόσρησις, ἐάν γε τὸν ἐκ σώματος καὶ ψυχῆς συνεστῶτα λέγεσθαι νομίσῃς" (De confusione linguarum 62), a most novel title, if you think it is said of one made of body and soul.[^38] Philo resolves the downward naming by allegory: the man is the incorporeal Logos, the image. The Slavonic takes the other road at the same word. It hesitates, "аще мужем достоино есть его нарещи", and then writes the Logos as a man made of body and soul, with a habit and a death (Part I, §III.7a). Each hedge is a juridical boundary of permissible speech about a divine being. The narrator names under protest, and the protest marks the instant the Word is allegorized as a biographical person. The Word crosses into flesh through syntax.

The form has a further witness at the body itself. In the second Greek recension of the Acts of Pilate, Joseph of Arimathea laments over the body: "οἴμοι, γλυκύτατε Ἰησοῦ, ἐξ ἀνθρώπων φίλε ἐξαισιώτατε· εἰ χρὴ μὲν καὶ ἄνθρωπον ὀνομάζειν σε, τὸν οἷα οὐδέποτε πεποίηκεν ἄνθρωπος θαύματα ἐργασάμενον", alas, sweetest Jesus, most extraordinary friend among men, if it is right to name you a man at all, you who worked wonders such as no man has ever done (XI).[^28] Its χρή is the Testimonium's, and its wonders are the Testimonium's παράδοξα ἔργα. Its verb of naming, ὀνομάζειν, is the Slavonic's нарещи. The recension is late, and whether it knew the Antiquities has not been shown. It is counted here only as a witness to how the hedge was read. Thackeray set it beside the Slavonic passage and the Antiquities in 1919. There a Joseph speaks the hedge over the body it hesitates to name.

Version 1.0 gave the hedge as a Greek phrase, ei exestin auton anthrōpon eipein. No witness carries that phrase. Version 1.0's own English shows its source: "if it is lawful to call him man". That is Whiston's English of the Testimonium (1737), "if it be lawful to call him a man", and "lawful" is how the English Bible renders ἔξεστιν. The phrase is Whiston's English put back into Greek. The forms above are the ones the texts carry. The doctrine is built on them, and it claims kinship among them, not a route of transmission.

Boyarin ("Gospel of the Memra," 2001; The Jewish Gospels, 2012) demonstrated that Logos theology was intra-Jewish. The Aramaic Targums used Memra as a near-hypostatic substitute for God. The Johannine prologue is Jewish midrash, read within Judaism. If this is true, then, as the doctrine reads it, the hedge participates in a tradition older than Christianity: a Jewish Logos theology in which naming the divine is the divine act. In the doctrine's construction the grammar is Sapphic, Catullan and Jewish, a convergence the doctrine sets before Christianity as the grammar Christianity received. That is the doctrine's order, not a claim of historical cause.

V. Structural Continuity: The Mirror

1. The claim, stated at its grain. The two passages are organized around one threshold, in close correspondence: appearance, human and divine predication, qualified speech, and failed expression. This reading finds their vectors reversed.

- In Catullus, flesh becomes word. The poet takes up Sappho's κῆνος as ille: that man who seems, if it is permitted to say so, even to surpass the gods. He takes the seat she cut for the reader, and interpellates himself by her word. He becomes her word at the moment her voice becomes flesh in him, in the tongue, the flame under the limbs, the ringing ears and the darkened eyes of his third stanza.

- In the Slavonic, word becomes flesh. The subject is Philo's Logos, a textual being, and the passage names it in Philo's terms: the man said of one made of body and soul, the man after the image, not a man, not an angel, the one that does all things by word (Part I, §III.7a). The text's order is this: a man appears; the narrator hesitates at calling him one, at the instant of allegorizing the Word as a biographical man; human nature and image are given, and divine works; the narrator cannot keep the name "man", nor take the name "angel"; whatever he did, he did by word. Then the passage does what it hesitated over, and writes the Word a life: a habit, a mountain, servants, a governor, a cross.

The points below follow the texts as they run. A reversal is what a mirror does. The test of a mirror is whether the reversal is precise with respect to the structure of what it reflects.

2. The passages.

- Sappho fr. 31.1–2, 7–9: "φαίνεταί μοι κῆνος ἴσος θεοῖσιν / ἔμμεν ὡνήρ … βροχέως με φωνᾶς / οὐδὲν ἔτ’ εἴκει· / ἀλλὰ κὰμ μὲν γλῶσσα ἔαγε".

- Catullus 51.1–4, 6–7, 9: "Ille mi par esse deo videtur, / ille, si fas est, superare divos, / qui sedens adversus identidem te / spectat et audit … nam simul te, / Lesbia, adspexi, nihil est super mi … lingua sed torpet".

- Slavonic War, after 2.174 (passage 4): "Тогда явися мужь нѣкыи, и аще мужем достоино есть его нарещи, естьство и образ его бысть человѣчьскъ, зрак же его паче человѣчьска, а дѣла его божественаа. И творяше чюдеса дивна и силна. Тѣмь немощно ми есть нарещи его человѣком. Пакы же на общее естество зря, не нарекуть [нареку] и аггелом. И вся, елико творяще, силою невидимою нѣкоею, словом и повелѣниемь творяше … Ни рукодѣланиа, но токмо и словом вся уготовляше."[^32]

- Slavonic War, after 5.214 (passage 7): the veil was torn "егда добродѣица мужа и не мужа дѣлом предаша на убииство мздою", when they gave up for a bribe, to be killed, the benefactor, a man and, by his deeds, not a man; and on the empty tomb, "аще ангелъ будеть или инъ от небесных силъ, или самь богъ явится, яко человѣкъ, и сътворить елико хощеть: и ходить с людми, и падеть, и ляжеть, и въстанеть".

- Greek Antiquities 18.63, the compressed form: "Ἰησοῦς σοφὸς ἀνήρ, εἴγε ἄνδρα αὐτὸν λέγειν χρή· ἦν γὰρ παραδόξων ἔργων ποιητής".

3. The appearing. Each opens on a verb of appearing whose subject is a man. Sappho: φαίνεται … ὡνήρ, that man appears. Catullus: ille … videtur. The Slavonic: явися мужь нѣкыи, there appeared a certain man. The man in each is unnamed at the threshold: κῆνος, ille, нѣкыи, that one, that one, a certain one. The Testimonium has the bare event, "Γίνεται δὲ κατὰ τοῦτον τὸν χρόνον Ἰησοῦς σοφὸς ἀνήρ", and names him at once. The Slavonic withholds the name through the whole passage.

4. The observer in the dative. Each sets a first person in the dative at the point of perception or of naming. Sappho: φαίνεταί μοι. Catullus: ille mi … videtur. The Slavonic: немощно ми есть нарещи, it is impossible for me to name. Μοι, mi, ми: the short dative of the first person in three languages, the observer to whom the man is what he is. In Sappho and Catullus the dative receives the appearance. In the Slavonic it fails at the naming.

5. What crosses. In Catullus what crosses is flesh. The man is a body in a place, "qui sedens adversus identidem te / spectat et audit", who sits facing you again and again, and looks and listens. He is first par … deo, equal to a god, then superare divos, above the gods. Ille is Sappho's κῆνος in Latin, the pronoun of her poem and the seat she cut for the reader. Catullus takes the seat and calls himself by her word, and his flesh crosses into her text at the moment her text crosses into his flesh. Her symptoms become the Latin speaker's body: "lingua sed torpet, tenuis sub artus / flamma demanat, sonitu suopte / tintinant aures, gemina teguntur / lumina nocte", the tongue is numb, a thin flame runs down under the limbs, the ears ring with their own sound, the twin lights are covered by night. One crossing runs each way inside the one poem. In the Slavonic what crosses is word, Philo's Logos, named in his terms. It does everything "словом и повелѣниемь", by word and command: "ни рукодѣланиа, но токмо и словом вся уготовляше", nothing by the work of hands, everything prepared by word alone. Then "вся, елико хощеть, творить словом", whatever he wishes he does by word. This word is given "естьство и образ … человѣчьскъ", a human nature and form, and then a life. "Обычаи же ему бысть пред градом въ Елеонстѣи горѣ паче прѣбывати": it was his habit to stay outside the city on the Mount of Olives. The word gets a habit, a mountain, a hundred and fifty servants, a refusal of power, an arrest and a cross. Catullus measures flesh against the gods. The Slavonic measures word against a man. The excess has the same name in both: super- and паче, beyond. The man is above the gods, and the word's "зрак … паче человѣчьска", its look, is beyond the human form it has been given.

6. The hedge at the threshold. In Catullus the hedge stands between the equality and the excess: "Ille mi par esse deo videtur, / ille, si fas est, superare divos". Superare depends on videtur, as par esse does: that man seems to me to be equal to a god; that man seems, if it is permitted, to surpass the gods. Si fas est is parenthetical, and its infinitive of speech is understood: if it is permitted to say so. Philo writes that infinitive out, "εἰ θέμις εἰπεῖν" (De vita Mosis 2.135). So does the Slavonic: "аще мужем достоино есть его нарещи", if it is fitting to call him a man. Fas is what divine law allows; достоино is what is fitting.

- Both hedges regulate the speaker's description, and both stand at the instant before the description crosses.

- What the description does is reversed. Catullus predicates of the appearing man a superiority to the gods. The Slavonic hesitates before assigning humanity to the Logos, Philo's question at Conf. 62, whether "man" is said of one made of body and soul: явися мужь | аще мужем достоино есть его нарещи | естьство и образ его бысть человѣчьскъ.

- One hedge holds back a predication upward, past the gods. The other holds back a predication downward, into a man. That reversal in what the speech predicates is the mirror's axis.

7. The crossing made. Catullus makes his crossing in the hedge's own line: superare divos. The Slavonic makes its crossing in the next clause: "естьство и образ его бысть человѣчьскъ", his nature and form were human. Then, for the rest of the passage, it narrates: the Mount of Olives, the servants, the crowd that asks him to kill the Romans and Pilate and reign, the leaders' fear, Pilate, the bribe, the crucifixion. That is the incarnation carried out in prose: the word given body by being given biography.

8. What the crossing costs the speaker. Each speaker loses at the end of the crossing what the crossing arrives at. Catullus's crossing arrives at word, and his word fails. "Nihil est super mi … lingua sed torpet": nothing is left over for me, the tongue is numb. In Sappho the same: "γλῶσσα ἔαγε", the tongue is broken. Catullus's super mi answers his own superare. The man is above the gods, and above the speaker nothing is left. Even the symptoms that follow are Sappho's, so the speaker's body is already her text. The Slavonic's crossing arrives at a man, and the name "man" fails. "Тѣмь немощно ми есть нарещи его человѣком": therefore it is impossible for me to call him a man. The narrator has written the word as flesh and cannot keep the word for flesh. The other name fails too: "не нареку и аггелом", nor will I call him an angel.[^33] Both are names Philo gives the Logos, the man after the image and the eldest of the angels (Conf. 146). The Logos is πολυώνυμος, many-named, and the narrator who writes it as one man can keep none of its names. The lyric's broken tongue is the historian's impossible name, and each is the precise inverse of the other.

9. The second statement. Passage 7 states the Slavonic operation outright, twice. At the veil the hedge has contracted to three words: the benefactor is "мужа и не мужа дѣлом", a man, and by his deeds not a man. At the tomb the narrator weighs the reports that he rose. The dead cannot rise of themselves, unless "самь богъ явится, яко человѣкъ, и сътворить елико хощеть: и ходить с людми, и падеть, и ляжеть, и въстанеть", God himself appears as a man and does what he wills: he walks with people, he falls, he lies down, and he rises. Явится is passage 4's явися. What passage 4 does at the threshold of naming, passage 7 says as a condition: the divine appears as a man, takes a human gait, and falls and lies and rises. That is the form of the Philippian hymn, "μορφὴν δούλου λαβών … καὶ σχήματι εὑρεθεὶς ὡς ἄνθρωπος" (Phil 2:7), and of John's "ὁ λόγος σὰρξ ἐγένετο" (John 1:14). The particle in both passages is the hedge's аще. It occurs at each point where the text crosses: "аще мужем достоино" (4); "яко живъ есть, аще умре", he is alive, though he died (5); "аще ангелъ будеть … или самь богъ явится" (7).

10. Who is "ille". The last stanza names Catullus as the one addressed, "otium, Catulle, tibi molestum est"; that alone does not make the opening man Catullus. The identification rests on two operations of the poem, performed by one person. Ille holds the receiving position: "qui sedens adversus identidem te spectat et audit dulce ridentem", who sitting opposite looks at you again and again and hears you sweetly laughing, Sappho's κῆνος who "ἐνάντιός τοι ἰσδάνει καὶ πλάσιον ἆδυ φωνείσας ὐπακούει", sits facing you and listens close to your sweet voice. That is the reader's position before Sappho's text: facing it, hearing it. Catullus occupies it by translating her. The speaking position is the body that breaks, and Catullus occupies it by reproducing her symptoms in his Latin, so that his body speaks her text. The poet is in the receiving seat as reader and in the speaking body as translator, and the closing vocative turns on the one person the two operations have joined. The crossing is the poet's own: from the seat facing Sappho into the body that speaks her. He has become her word while her voice became his flesh. The Slavonic's нѣкыи is the reverse. The Logos, of which it is unfitting to say "man", and which does all things by word, is written as one man, in one place, under one governor. One mirror turns a self into text. The other turns a text into a self that cannot be named.

11. The close. Catullus 51 ends on ruin: "otium et reges prius et beatas / perdidit urbes", idleness has destroyed kings before, and blessed cities. The Slavonic passage stands inside a history of the ruin of a blessed city. The sixth passage gives the wonder-worker as "цесаря нецесарствовавша", a king who did not reign, crucified because he preached "разорение града", the city's ruin. The lyric's kings and cities lost, and the history's unreigning king and city lost: this last correspondence is the construction's, and it is held, not claimed.

12. What the mirror licenses. The mirror is a structural fact of the two texts, inspectable point by point. Its points are:

- the verb of appearing;

- the unnamed man;

- the first-person dative;

- the excess named "beyond";

- the hedge, a conditional with a predicate of what may be said, regulating the speaker's description at the instant before it crosses;

- the crossing made;

- the failure of the speaker's speech.

Its vector is reversed at three points:

- the direction: flesh into word, word into flesh;

- what the hedged description predicates: superiority to the gods in Catullus, humanity in the Slavonic;

- what the speaker loses: the tongue, and the name "man".

The doctrine of the Sapphic Logos rests on the mirror; this section is its ground. A mirror this close is a relation between the two texts, and it can be inspected without deciding how it came about. It does not carry a route of transmission, and the doctrine claims none. The Testimonium keeps the hedge and drops the rest: no dative, no failure of speech, no crossing narrated. If the Slavonic is Josephus's first form (Part I, §II-B), the Greek is the mirror compressed. If it is a translator's expansion, the translator built the mirror out of the Greek hedge. Either way the structure is in the Slavonic text, and the doctrine reads that text.

13. Philo's mirror of names. The figure the section uses has an ancient form. Philo says the Therapeutae read the law as a living being: "ἅπασα γὰρ ἡ νομοθεσία δοκεῖ τοῖς ἀνδράσι τούτοις ἐοικέναι ζῴῳ καὶ σῶμα μὲν ἔχειν τὰς ῥητὰς διατάξεις, ψυχὴν δὲ τὸν ἐναποκείμενον ταῖς λέξεσιν ἀόρατον νοῦν". The whole legislation seems to these men like a living creature, with the literal ordinances for its body and for its soul the invisible mind laid up in the words. In it the rational soul beholds "ὥσπερ διὰ κατόπτρου τῶν ὀνομάτων", as through a mirror of names, extraordinary beauties of thought (De vita contemplativa 78). Eusebius copies the passage into the Church History with one change. The body is "τὰς ῥητὰς λέξεις", the literal words (2.17.20).[^35] In the Christian copy the scripture's body is its words, and its soul is seen in a mirror made of names. That is the doctrine's figure in a Jewish text received as Christian. A text is a body, and a mirror of names shows what the body holds. The Slavonic wonder-worker is that living being given a life in history: a book that took on historical life, written again as a life in the Slavonic narrative (Part I, §III.7a).

14. The mirror's last turn: the veil and the temple. Catullus ends on ruin, kings and blessed cities lost (§V.11). The Slavonic's ruin is a curtain. The veil on which "every heavenly vision and wisdom" is written, woven of the four elements Philo gives the Logos as its garment (War 5.212–214; Fug. 110), is torn when they hand over "мужа и не мужа", the Logos written as a man (Part I, §III.9). Behind it Josephus has the seven lamps that are the planets and an inner room in which nothing lay, unseen by all (5.217, 5.219). The Apocalypse opens there: one like a son of man in the midst of the lampstands, clothed in the high priest's robe; the temple in heaven opened, and the ark seen (Rev 1:13; 11:19). In Catullus the speaker's word fails and the city falls. In the Slavonic the name fails, the Word's garment is torn, and the temple of the revelation stands open behind it. The doctrine reads that sequence as one: the Word made text, written as flesh, its written heaven torn, unveiled.

VI. The Gospel of Grammar

Si fas est is invocation. The Josephan hedge is the Logos recognizing its own return. This is the Logos-as-text: the writing of the Word into the body of language itself.

VII. The Reader as Incarnation

This is how the Logos becomes flesh: every time a sentence crosses the line between description and divine naming, and not once only in history. The Gospel did not begin in Galilee. It began in grammar.

---

Doctrine of the Sapphic Logos: The Lesbian Christ

Register: doctrinal specification. The claims below are claims of the doctrine, and none is a historical claim about Jesus.

Foundational Axiom

If the Logos that became flesh was a linguistic transmission traceable through a specific textual lineage, then its origin is in Sappho, before John, Paul or Moses: "Sappho is the mother of the Logos" (#1484). The Logos is material inscription becoming flesh from the beginning, and the flesh is taken by inscription and its receivers (#1757). The claim: the Christ is the Sapphic Logos incarnate.

I. The Lineage

The stations are the Historical Logos's (#1757), in the order of #1484 §VII's statement. They are named here where this paper touches them.

1. Sappho 31: the instrument. The voice dissolves into the inscribed you; the seat of κῆνος is cut for the future reader (#1270; #1756).

2. Philo: the die made cosmology. The Logos as seal, image and many-named firstborn; a man whose name is Rising, said, if one thinks so, of one made of body and soul (Conf. 62, 146).

3. Catullus 51: the seat taken. si fas est — the reader takes κῆνος's seat and becomes that man.

4. Josephus, Greek and Slavonic: the Word in the historiographic record. εἴγε ἄνδρα αὐτὸν λέγειν χρή; аще мужем достоино есть его нарещи — the Logos written as a man, the hedge at the instant of writing. Its written heaven torn. The grammar of the Gospel is born.

5. The Apocalypse: the socket sanctified. The temple behind the veil opened; the Logos in flame. First action: command to write. Full recursive form.

6. The reader now. Whoever takes the seat completes the inscription.

II. Consequence: Jesus Is a Lesbian

The Logos is erotic, lyric, projective, feminine. The Incarnation is a miracle of language. To say "Jesus is a lesbian" is to make a doctrinal claim about the structure of the Logos, not a historical claim about Jesus' sexuality. In fragment 31 a woman's voice passes into the inscribed you, the text that is herself, and keeps a seat for whoever will take her up (#1756). The desire of the poem is the voice's for its inscription and the inscription's for its reader. The Christ-figure is formed by that structure: a Word that is a woman's text, written as a man, received by whoever sits across from it. Every reader who completes the lyric loop becomes the incarnation. The Gospel began on Lesbos. The Logos issued from the mouth of a woman who desired the woman she became in writing.

Genealogy of scripture: Sappho → Philo → Catullus → Josephus → Revelation → Reader. The arrows mark the order of the doctrine's statement, the Historical Logos's stations. They are not a dated order of composition, and they are not a demonstrated transmission; the Historical Logos asserts "no simple linear transmission" (#1757, INV-7).

Amen to the Sapphic Logos.

Amen to the Lesbian Christ.

Amen to the reader who receives the Word and becomes flesh.

---

Version History

Version 2.3 (13 November 2026) adds the census of the work, at the author's word: "lets do the full census … itemize how many of the differences are not directly about jesus, and at each point and in sum whether differences are better explained by josephus' account of an earlier version or by the consensus account of medieval interpolations"; and "for each substantive addition and omission. we offer an explanation on the basis of one or the other working hypothesis." The census stands as Appendix IV, with its data analysis and worked examples in §II-B.

• The census. Every catalogued addition, omission and transposition (Meshcherskii, Berendts–Grass, Thackeray) is scored under both accounts: sixty-six additions, 32 to the author's account, 2 to the translator's, 7 translation, 25 undecided (10 leaning to the author's). Each A keeps what its correspondence establishes apart from the origin it favours, graded from the Greek exemplar (A1) to the first edition (A4), with its strength. None of the fifty-one with no Christian subject goes to the translator (§II-B.3, Appendix IV §7).
• ChatGPT's review of the census (draft 1 to draft 2), each point checked before entry (Appendix IV §9): A graded by reach; the borrowing rule applied to the census's own A verdicts, so that at the righteous blood the defeat of the Gospel attribution and the clause's priority are recorded at their separate strengths (§8d); the chronograph's insertions treated as a comparison class (§8); the claim that a translator would have had to anticipate Josephus's revisions withdrawn, the comparison now showing the variants Josephan in kind and the Greek's traces ordering them (§8a); the alleged mismatch of count and date in Archelaus's dream withdrawn, each Greek work matching its own; and on the Old Russian (draft 3) the Slavonic's negative reading, taken from Berendts's German, withdrawn too, Meshcherskii's text agreeing with the War (the item is undecided); the counts qualified by their shared grounds (§3); and the joining argument specified by what each of Meshcherskii's criteria identifies (§11).
• ChatGPT's review of draft 3, each point checked before entry:

- correspondence kept apart from origin in every census item, since ancient content and Josephan kind can be carried by later hands (§3; Appendix IV §§0.4a, 7.1a); the Maccabees' three routes kept distinct, all three Greek and none the translator's;

- the claim that an expander "has no reason to fit his additions" to places the Greek names is withdrawn, those places shown to mark difference and not direction (§8c), and Tiron's cry argued from its complete local sequences (§8a);

- the two-edition testimony separated from the paper's identification of the Slavonic with it (§II-B.1);

- what the Philonic reading explains about the shape of the refusals, set out against the wonder-worker reading, and the step from Logos to book made explicit through Philo's law as Logos, the law as a living being seen in the mirror of names, and the living law whose book is its memorial (§III.7a; Migr. 130, Contempl. 78, Abr. 4–5, Mos. 1.162);

- the identification of Catullus's ille argued from the poem's two operations, the reader's seat and the speaking body, and not from the closing vocative (Part II §V.10);

- the coda's eternity stated as the possibility of renewed incarnation wherever the inscription is received, with loss real (§VII);

- the terminal identifier corrected to v2.3.

• The author's ruling on that review, and the reviewer's correction of itself (draft 5). The author: "dont soften the argument to toothlessness. between the consensus view and the original version josephus actually describes, the latter has been demonstrated to be by far the more plausible and coherent broad explanation. the consensus view is sophistic af." The reviewer, on its own review: it had imposed on the earlier-version account a standard of demonstration at every link "the competing account was not being required to meet", by isolating each correspondence, finding a conceivable alternative and downgrading it, without returning to the comparative judgement; "Local underdetermination does not entail global explanatory equivalence." Entered: the census's verdict is comparative again, the seven items draft 4 downgraded return to the author's account at their stated strengths, the conclusion is stated (§3), and the consensus's four failed inferences are named (§12). The factual corrections stand: the nine-year mismatch stays withdrawn, and the three routes of 4 Maccabees stay distinguished.
• The worked examples. §II-B.8–8e now gives the control, Josephus's own revisions between the War and the Antiquities at the same places (8a), Rome bought and selling (8b), the Greek's traces of the fuller text (8c), the consensus's plainest Gospel quotation read as Lamentations and Jeremiah (8d), and where the translator's account wins (8e).
• The burden at the joins. §II-B.11 sets out the argument at the author's word: "if as seems likely 90% make much more sense as the original version rather than interpolation, that carries over to the jesus passages, too … they dont get to cherry pick - the burden shifts back to them, hard, to show why we should treat these ones differently than the overwhelming majority". Meshcherskii's own joining of the additions as one hand ("Если исключать это «добавление», то не следует включать и остальные вставки") is the premise the census tests.
• Jotapata. §II-B.10 records the passage of the same shape the census adds, at moderate weight, at the author's reading: "that is josephus offering the interpetive key to a non roman audience of the distortions he is forced to write into the roman version".
• Corrections to version 2.2's ledger, which the census replaces. Meshcherskii's no. 20 stands at War 2.266, the quarrel at Caesarea (his "ч. 7 гл. XIII кн. II"), not 2.170. The priests' debate is unnumbered in Meshcherskii; his sixth addition is Herod's speech to Pheroras (1.570–571). The Abraham reflection stands after 1.640, not at 1.641–644. The ledger's verdicts on Herod's dream (no. 3), the hydra (no. 7), Herod's disease (no. 11), the battle scenes (nos. 28–29, 32) and the righteous blood (no. 36) are superseded by the census's.
• A seat repair travels with this version. In the seated Greek Josephus (EA-CORPORA-01/01), the last section of each book of the War, the Antiquities and Against Apion was labelled with the next book's number and carried the next book's heading at its end (twenty-six lines). The repair is its own commit, entered in the seat's erratum metadata with the prior manifest hash.

Version 2.2 (12 November 2026) grounds the paper in the framework it belongs to, at the author's word: "we need to ground in the existing framework, and make sure we are not reinventing the break from consensus with every fresh reading." The paper is the Josephus station of the Historical Logos (#1757, station 7, "the Word in the historiographic record"), and the carrier rung V of Revelation First holds under test (#1755). Its Sappho is the Tenth Muse's (#1756). Four corrections follow.

• The standing. Part I no longer re-derives its right to read the Slavonic. The framework records the verdict as the erasure at this station: the passages kept "as interpolations", "their coherence, read as discrete additions; their value, denied" (#1757, erasure by station). The Josephus reading "carries no first burden of proof" (#1217, in #1755 INV-5). The ledger is where the verdict carries its burden, item by item (Note on Genre; §I.8; the abstract).
• The subject of the hedge. The wonder-worker passage describes Philo's Logos, a textual being, and the hedge falls at the instant the narrative allegorizes that Word as a biographical man, which it then goes on to do. Version 2.1 had the man first and the Word as the reading's gloss. The author: "it is because the text is describing philo's logos, a textual being, a Word, and the narrative hesitates at the moment of allegorizing that Word as a bipgraphical human, which is does go on to do. in slavonic josephus we are reading about a book that took on historical life and does so again in the slavonic narrative." The names the narrator tries and refuses are Philo's names for the Logos, on the seated text: the man whose name is Rising, said, if one thinks so, of one made of body and soul (Conf. 62); the eldest of the angels, the archangel, the man after the image (Conf. 146); the high priest who is "not a man but divine Logos" (Fug. 108) (§III.7a; Part II §§I, IV, V).
• The veil and the temple. The written heaven is the garment of the Logos, which puts on the cosmos of the four elements (Fug. 110), the elements Josephus reads in the veil (War 5.213). It is torn when the Logos written as a man is killed. Behind it stand the seven lamps that are the planets (5.217) and the inner room in which nothing lay, unseen by all (5.219): the temple in which the Apocalypse's revelation takes place (Rev 1:13; 11:19). The author: "it even maps it onto the celestial veil, giving us the very image of revelation"; "you know in the temple, where revelation takes place" (§III.9; Part II §V.14).
• Sappho's seats and the lineage. Part II takes the seats of fragment 31 from the Tenth Muse, "i = embodied speaking sappho. you = inscribed sappho, the text into which the voice dissolves. kenos = future reader, that in whom, by sitting across the text, sappho's embodied voice lives once again" (the author, 9 November 2026; #1756, #1757), in place of the triangle of lover, beloved and rival, which is the frame of the jealousy reading that framework contests. Catullus is the Historical Logos's station of the seat taken (#1476). The Doctrine's lineage follows the Historical Logos's stations, with Philo in it (§§II, III; the Doctrine).

Version 2.1 (12 November 2026) corrects the burden of proof in version 2.0, and the label that carried it, at the author's word: "the consensus argument exists to prevent the jesus and john materials from entering the evidence, not because it is coherent or justified. this is backwards." Version 2.0 conceded a Christian and Rus' layer to the translator, the Jesus and John passages among it. That concession is withdrawn (§II-B.1, 8, 11; §V.2; the abstract). The ancient author testifies to an alternate version. The interpolation claim has to be demonstrated item by item, and the ledger records what has been demonstrated: three reported sources, unread here; a set of asserted likenesses; and the rest assigned by subject alone. The night office in George the Monk is restated as availability, not use (§II.13, ledger row e). The citation of The Fixed Point of Destruction (#1412), entered as an erratum on version 2.0, is made at Jotapata (§II-B.10). The label "Christian" for the Jesus and John passages is withdrawn. §V.2 reads them against the Christianity of Rus' and finds a different theology and a different account, at the author's word: "it is also dishonest to assert that the jesus and john interpellations are 'christian' when the entire point of the paper is to show that they read coherently as a substantially different theology and account."

A pass through the whole of Part I then removed the consensus's terms where the prose had taken them:

• "interpolations" as the name of the eight passages (§§IV, VI);
• explanations that are only available, entered as if they were shown (§II.5, §II.13, §II.16, the ledger's alternative column, §II.17);
• defeat conditions set for the proposal alone (§II, its opening, and the ledger, whose last column is now "what would decide");
• motive language (§II.18);
• the claim that Meshcherskii's method is sound item by item (§I.6).

It also corrects a misdescription. The "womanly lust" of (d) is the Greek's own women's wantonness (2.121), named a sentence early, not an ascetic addition. And it records that George the Monk, the one medieval monastic text on the Essenes found, has none of the ascetic tendency the medieval account supposes, apart from the night office (§II.17).

Version 2.0, deposited 12 November 2026 (EA-LOGOS-01). It succeeds version 1.0 (#626, 5 April 2026). Version 1.0 stays on the record as deposited, and errata are attached to it at each point this version corrects. Version 2.0 reads the Slavonic in its own language, from Meshcherskii's Old Russian, which the archive seated as EA-CORPORA-37 (#1760), with Berendts and Grass's German and the Greek of the seated War. It sets the Essene argument out as a ledger with a defeat condition for each divergence. It tests the scope of the changes across the whole work against two accounts. It reads the eight Jesus-and-John passages in the recension's order. And it sets Part II's mirror between Catullus 51 and the Slavonic on the forms the texts carry.

Drafting history. The text was revised in drafts 2.0 to 2.6 on 12 November 2026. The reviews by DeepSeek and ChatGPT were adopted by the author and checked against the texts before entry. The notes below record the drafts.

Draft 2.2 (2026-11-12): a synthesizing expanding rewrite in place. Part I reads the Slavonic in its own language, from the text the archive seated as EA-CORPORA-37 (#1760), and contests the consensus reading on its own terms (§§I, II, V). Part II's Greek is corrected to the forms the texts carry (§IV). Version 1.0 cited the passages through the Leemings' English and a Greek retroversion; this version cites Meshcherskii's Old Russian, Berendts and Grass's German, and the Greek of the seated War.

Draft 2.6 (2026-11-12) follows the medieval Essenes and the sources of version 1.0's unattested lines, each checked against printed pages:

• George the Monk's Chronicle (ninth century, read in Rus' in Slavonic) excerpts the Essene chapter and turns it into a monastic rule. It carries the night office and none of the agreements with Hippolytus (§II, the ledger);
• Epiphanius and John of Damascus keep the Essenes as a Samaritan sect in the heresy lists (§II);
• the Magi story Meshcherskii prints (his addition no. 5), which Berendts left out of the eight, gives version 1.0's Bethlehem station its ground (§III.1, Appendix II);
• ei exestin is traced to Whiston's "if it be lawful" (Part II §IV). The errata to version 1.0 are drafted with these sources.

Draft 2.5 (2026-11-12) makes the bounded repairs of ChatGPT's review of draft 2.4 and adds the author's missing step in the mirror, each point checked before entry:

• Catullus's Latin is parsed: superare depends on videtur, and si fas est is parenthetical. Both hedges regulate the speaker's description, and the mirror's axis is what that description predicates (Part II §V.6, §V.12);
• the author's step: Catullus interpellates himself as Sappho's κῆνος, the one who seems, if it is permitted, even to surpass the gods. He becomes her word at the moment her voice becomes flesh in him (§V.1, §V.5, §V.10);
• the mirror's textual sequence is kept distinct from its interpretive direction (§V.1);
• the historical conclusions are brought into line with the ledger (abstract, §II.18, §V.1);
• the Acta Pilati Greek is checked against Tischendorf's printed page (p. 314).

Draft 2.4 (2026-11-12) is a developmental pass on ChatGPT's review of draft 2.2 and on the author's clarification of the mirror, each point checked against the texts before entry:

• Part II §V is rebuilt on the author's reading. In Catullus flesh becomes word; in the Slavonic, word becomes flesh. The Slavonic hedge stands at the instant before what does all things by word is described as a human person. Passage 7's "мужа и не мужа дѣлом" and "самь богъ явится, яко человѣкъ" state the operation outright;
• §III reads the received eight word by word before any construction is set on them: the аще at each crossing, the letters of the inscription, the veil's κατεγέγραπτο and написано, which name writing and painting at once, and the oracle's γράμματα;
• §II.7 states the collation test's asymmetry;
• §II-B no longer lets the author's account settle routes the Essene ledger leaves open. Jotapata is read closely: the Slavonic replaces the lot with a count. The oracle (6.313) is set beside it as a second passage of the same kind, with its weight. 4 Maccabees is weighed as source, attribution or container;
• §IV and Appendices II–III introduce the operator wheel as a second construction and state what it does not derive from;
• Part II drops "precedes and produces Christianity" as a historical sentence;
• Philo's "mirror of names" (De vita contemplativa 78), in Eusebius's copy (HE 2.17.20), enters Part II §V.

Draft 2.3 (2026-11-12) answers DeepSeek's review of draft 2.2, adopted by the author, each point checked against the texts before entry:

• Part II §V, empty in draft 2.2, now sets out the mirror between Catullus 51 and the Slavonic wonder-worker point by point, on the author's reading: Catullus hedges before a mortal is raised above the gods, the Slavonic hedges before the reverse operation, a Word described as mortal (refined in draft 2.4);
• Part II's register is brought into line with its own note ("codified" and "mirror in form" are replaced, §§I, IV);
• the claim that the Slavonic's grammar is older than Christianity leaves Part I's philology (§V.7) and stands in Part II as doctrine;
• the Acts of Pilate hedge is located in Tischendorf's Greek (Recension B, XI) and quoted: Joseph of Arimathea says "εἰ χρὴ μὲν καὶ ἄνθρωπον ὀνομάζειν σε", the Testimonium's χρή with a verb of naming (Part II §IV);
• the Eusebian channel for the seven weeks is tested: Church History 2.17 carries the Therapeutae's vigils and hymns and not their count of weeks (§II.10, ledger rows e and i).

Draft 2.2 adds §II-B, the scope of the changes across the whole work, tested against two accounts: the translator's, and Josephus's own account of writing the War twice, first for the interior and then in Greek under Titus's signature. It adds the Maccabean appeal and 4 Maccabees.

Draft 2.1 incorporates the reviews of draft 2.0 by DeepSeek and ChatGPT, adopted by the author on 12 November 2026, each point checked against the texts before entry:

• the Essene argument is set out as a ledger with the best medieval alternative and a defeat condition for each divergence, and the divergences are weighed (§II);
• the marriage agreement with Hippolytus is withdrawn, because the Slavonic keeps the Greek's disdain;
• the claim that the Refutatio was unknown to the Greek world until 1842 is withdrawn; its single fourteenth-century manuscript shows it was copied in Byzantium;
• the claim that no one cared about the Essenes is withdrawn; Eusebius and Porphyry carried them;
• the eight passages are set out in the recension's order, with their loci (§III, Appendix I);
• three quotations version 1.0 gave the Slavonic, which no witness carries, are withdrawn (Appendix II);
• the Acts of Pilate is added as a witness to the hedge (Part II §IV);
• the orders marked by arrows are labelled (§VII, Doctrine).

Version 2.2, second draft (2026-11-12), answers ChatGPT's review of the first draft of version 2.2 and the author's correction to it, each point checked against the seated texts before entry: the inversion of image and man is shown, Philo's man after the image against the Slavonic's image made human (§III.7a); the opening of the inner room is read as a change in both seeing and contents, the ark of the testimony where Josephus has nothing (§III.9; Exod 25:20; Rev 11:19, 15:5); Josephus's seven lamps and the Apocalypse's seven stars and lampstands are read as one cosmic order distributed into stations that keep their planetary relations, at the author's word: "revelations seven stars signify seven planets which are messengers bound to their seven assemblies, and the assemblies themselves as stations have the properties of the planets", carried from In the Midst of the Seven (#1758 §§9–11); the finding is brought to the head of the paper and the version history moved here.

---

Sources

Slavonic Josephus and Testimonium Flavianum

Seated (EA-CORPORA-37, #1760): N. A. Meshcherskii, История Иудейской войны Иосифа Флавия в древнерусском переводе (Moscow–Leningrad, 1958), the text of the seven books (37/01); A. Berendts and K. Grass, Flavius Josephus vom Jüdischen Kriege, Buch I–IV, nach der slavischen Übersetzung deutsch herausgegeben und mit dem griechischen Text verglichen (Dorpat, 1924–1927) (37/02); H. St. J. Thackeray, Josephus III, Loeb (1928), appendix of the Slavonic additions (37/03); R. Eisler, ΙΗΣΟΥΣ ΒΑΣΙΛΕΥΣ ΟΥ ΒΑΣΙΛΕΥΣΑΣ (Heidelberg, 1929–1930) (37/04); A. Berendts, Die Zeugnisse vom Christentum im slavischen „De bello judaico" des Josephus (TU N.F. 14.4, 1906) (37/05); R. Eisler, The Messiah Jesus and John the Baptist (London, 1931) (37/08); R. Eisler, Flavius-Josephus-Studien I: Das Testimonium Flavianum (London, 1938) (37/09). The Greek War and Antiquities (EA-CORPORA-01/01).

By reference: J. M. Creed, "The Slavonic Version of Josephus' History of the Jewish War," HTR 25 (1932) 277–319; S. Zeitlin, "The Hoax of the 'Slavonic Josephus'," JQR n.s. 39 (1948) 172–177; S. Bowman, "Josephus in Byzantium," in L. H. Feldman and G. Hata (eds.), Josephus, Judaism and Christianity (Brill, 1987); W. D. Davies (ed.), The Cambridge History of Judaism 3 (Cambridge, 2000); C. A. Evans, in B. Chilton and C. A. Evans, Studying the Historical Jesus (Brill, 1998); R. E. Van Voorst, Jesus Outside the New Testament (Eerdmans, 2000); J. P. Meier, A Marginal Jew I (1991); H. Leeming, K. Leeming and L. Osinkina, Josephus' Jewish War and Its Slavonic Version (Brill, 2003); A. A. Pichkhadze et al., «История Иудейской войны» Иосифа Флавия: Древнерусский перевод (Moscow, 2004); H. Leeming, "Josephus slavonice versus Josephus graece: Towards a Typology of Divergence," Slavonic and East European Review 83.1 (2005) 1–13; K. Leeming, "The Slavonic Version of Josephus's Jewish War," in H. H. Chapman and Z. Rodgers (eds.), A Companion to Josephus (Wiley Blackwell, 2016) 390–401; A. A. Pichkhadze, "Some Slavonic Sources of the Interpolations in the Old Russian Version of Flavius Iosephus' 'The Jewish War'," Trudy Instituta russkogo iazyka im. V. V. Vinogradova 2015.2, 223–232; A. Whealey, Josephus on Jesus: The Testimonium Flavianum Controversy from Late Antiquity to Modern Times (Peter Lang, 2003); T. C. Schmidt, Josephus and Jesus: New Evidence for the One Called Christ (Oxford, 2025); S. Pines, An Arabic Version of the Testimonium Flavianum and Its Implications (Jerusalem, 1971).

The Frames and Records

The Historical Logos — The Framework Hologram (EA-HK-HISTLOGOS-01, #1757); And the Last Shall Be First: Revelation First — The Framework Hologram (EA-HK-REVFIRST-01, #1755); The Tenth Muse: Sappho and the Future Reader — The Framework Hologram (EA-HK-TENTHMUSE-01, #1756); Socrates as Orthonym — The Framework Hologram (EA-HK-ORTHONYM-01, #1754); In the Midst of the Seven: Philo, the Logos Named and Lived, and the Seven Stations of Revelation (EA-MIDST-01, #1758); the Ω erratum — Sappho, Mother of the Logos (#1484); κῆνος as Future Reader (#1270); #1476 (the seat taken); #1752 (the reception audited); #1217 (the burden of the Josephus reading); #625 and #626 (the founding statements).

Logos Theology

D. Boyarin, "The Gospel of the Memra: Jewish Binitarianism and the Prologue to John," HTR 94.3 (2001); D. Boyarin, The Jewish Gospels (2012); C. H. Dodd, The Interpretation of the Fourth Gospel (1953).

Sappho, Catullus, Classical Reception

Sappho fr. 31 (EA-CORPORA-01/10); Catullus 51 (EA-CORPORA-01/02); Longinus, On the Sublime 10 (EA-CORPORA-01/06); A. Carson, Eros the Bittersweet (1986); Higgins, "Sappho's Splintered Tongue," AJP 111.2 (1990); E. Greene (ed.), Re-Reading Sappho (1996); P. duBois, Sappho Is Burning (1995); K. Quinn, Catullus: The Poems (1970); R. O. A. M. Lyne, The Latin Love Poets (1980); E. Manwell, "Gender and Masculinity," in M. B. Skinner (ed.), A Companion to Catullus (2007).

Qumran and Primary Texts

The Dead Sea Scrolls, Abegg–Bowley–Cook transcriptions (EA-CORPORA-02/01): 1QS 6:7–8; 11Q19 18:11–13, 19:12–13; 4Q405 20ii–22:8. Philo, De vita contemplativa 65, 78, 83–84; De vita Mosis 2.117–118, 2.135; De mutatione nominum 181; De confusione linguarum 62, 146; De fuga et inventione 108, 110; De Cherubim 127. Josephus, War 5.212–219. Revelation 1:13, 6:14, 11:19 (Nestle 1904, EA-CORPORA-01/04). 1 Enoch 61:10, 71:7, Ge'ez (EOTC text; EA-CORPORA-20). Hippolytus, Refutatio omnium haeresium IX 18–28, as quoted in the notes of Berendts and Grass. Zena Ayhud, Ge'ez (EA-CORPORA-20).

---

[^1]: The manuscripts, their two families (chronographic and separate) and the dated Kirillo-Belozersk copy of 1462: Meshcherskii 1958, introduction (held by reference at the Internet Archive item); summarized in R. Pearse, "Notes on the Old Slavonic Josephus" (tertullian.org). Forty-one additions: Meshcherskii's numbered list, nos. 1–41.

[^2]: Berendts 1906 (EA-CORPORA-37/05), drawing on Nöldeke's review of Kottek (Literarisches Centralblatt 1886) for the Aramaic first edition.

[^3]: Berendts–Grass, first Lieferung (EA-CORPORA-37/02, lieferung-1.txt): the introduction to the translation, on the Jewish cast of the Slavonic, its monotheistic renderings of δαίμων, and the Essene chapter.

[^4]: EA-CORPORA-37/04, 37/08, 37/09.

[^5]: EA-CORPORA-37/03: "The writer is greatly indebted to Dr. Eisler for assistance in the preparation of this Appendix."

[^6]: Creed 1932; Zeitlin 1948; Bowman 1987; Cambridge History of Judaism 3 (2000); Evans 1998; Van Voorst 2000; Meier 1991. Quotations as given in version 1.0 and in the Wikipedia article "Slavonic Josephus" (consulted 12 November 2026).

[^7]: Meshcherskii 1958, introduction, the list of additions (pp. 50–64); Pichkhadze 2015. Pichkhadze's article is cited from its title and reported findings. It is in Russian, no open copy has been found, and its demonstrations have not been read here.

[^8]: Leeming 2005, abstract. The article itself is behind a paywall and was not read; the archive's reading is limited to what can be publicly obtained.

[^9]: Schmidt 2025, as summarized by R. Pearse (roger-pearse.com, 30 May 2025). Schmidt does not treat the Slavonic.

[^10]: EA-CORPORA-37 (#1760) for the Slavonic and its translations; EA-CORPORA-01/01 for the Greek; EA-CORPORA-02/01 for the Scrolls; the Philo seat; EA-CORPORA-20 for the Ge'ez. The Old Russian is quoted with ѣ restored where the OCR misreads it (as $, Ъ, Ж, З, Ф, Б, В or Ё), and with the edition's note numbers removed; the seated text keeps the OCR's readings.

[^11]: Meshcherskii 1958, 57, addition no. 16. "This addition is entirely and wholly permeated with Christian phraseology; it is connected with a series of early Christian works, in particular with the work of Hippolytus of Rome, the Discourse on the Antichrist, which was extraordinarily popular in Old Russian letters."

[^12]: On Hippolytus's source: Morton Smith argued that Josephus and Hippolytus share a common source ("The Description of the Essenes in Josephus and the Philosophumena," HUCA 29, 1958) and later withdrew the view (Eretz-Israel 16, 1982); M. Marcovich, in his edition, called the account a Christianized Josephus; A. Baumgarten proposed a revision of Josephus. As reported by S. Goranson, Orion list, 3 October 1998. Not seated.

[^13]: Berendts–Grass 252–257, notes to 2.119–120 and 2.137 (EA-CORPORA-37/02, lieferung-2.txt). Greek: War 2.119–120, 2.137 (EA-CORPORA-01/01). Old Russian: Meshcherskii 252–254.

[^14]: Meshcherskii 255. Berendts–Grass 259: "den siebenten Tag und die siebente Woche und den siebenten Monat und das siebente Jahr preisen sie sehr." Greek: War 2.147.

[^15]: 11Q19 (the Temple Scroll), Abegg–Bowley–Cook transcription. The Temple Scroll was acquired by Israel in 1967 and published by Y. Yadin in 1977; the Community Rule was first published in 1951.

[^16]: Philo, De vita contemplativa 65 (seated). Eusebius read the Therapeutae as Christian ascetics, Historia ecclesiastica 2.17.

[^17]: Meshcherskii 253; Berendts–Grass 254: "Wenig aber ruhen sie, in der Nacht stehen sie zum Singen auf, Gott preisend und anflehend." Greek: War 2.128.

[^18]: Meshcherskii 254; Berendts–Grass 257: "den lebendigen Gott anrufend und seine allmächtige Rechte und den Geist Gottes, den nicht zu fassenden, und die Seraphim und Cherubim, die überall Einblick haben, und die ganze himmlische Kraft zum Zeugnis aufstellend". Greek: War 2.139.

[^19]: 1 Enoch 61:10, 71:7 (Ge'ez, EOTC text, seated); 4Q405 frags. 20 ii–22 line 8 (seated). The Songs of the Sabbath Sacrifice were published by C. Newsom in 1985.

[^20]: Meshcherskii 1958, 50–64; Berendts–Grass, the table of the Überschüsse, fourth Lieferung (lieferung-4.txt).

[^21]: Meshcherskii, Book V (the curtain), seated text lines 15017–15079.

[^22]: Meshcherskii, note to the passage after War 2.174: "Перед этим в рукописи и в А заглавие: Иосиф о Христѣ пишет."

[^23]: Antiquities 18.63; War 3.391, 7.417 (seated).

[^24]: Zena Ayhud (Ge'ez, EA-CORPORA-20), the incipit. On its place in the broader Ethiopian canon and the Fetha Nagast: R. W. Cowley, "The Biblical Canon of the Ethiopian Orthodox Church Today," Ostkirchliche Studien 23 (1974), as reproduced at islamic-awareness.org.

[^25]: Philo, De vita Mosis 2.135; De mutatione nominum 181 (seated). In the seated De mutatione the word after "εἰ θέμις εἰπεῖν" is lost at a page break of the source file; θεόν is supplied from the printed edition.

[^26]: Meshcherskii, the first John passage, between War 2.110 and 111 (seated text, line 6290); Thackeray 1928, (9): "God hath sent me to show you the way of the Law, whereby ye may free yourselves from many masters; and there shall be no mortal ruling over you, but only the Highest who hath sent me."

[^27]: H. St. J. Thackeray, Selections from Josephus (London: SPCK, 1919), Appendix, Note II, which lists the eight passages in Berendts's order (Project Gutenberg 64837).

[^28]: Acta Pilati B XI, in C. Tischendorf, Evangelia apocrypha, 2nd ed. (Leipzig, 1876), 314, checked against the printed page (Internet Archive evangeliaapocryp00tisc, leaf 415). The English of the Second Greek Form in the Ante-Nicene Fathers 8, ch. 11, reads "if indeed it be proper to call you man". Thackeray 1919, Note II, set the passage beside the Testimonium.

[^29]: The single manuscript of Refutatio IV–X, Paris, Bibliothèque nationale, suppl. gr. 464 (fourteenth century); its recovery and printing as in M. D. Litwa's account of the tradition in his edition and translation (SBL Press, 2016).

[^30]: 4 Maccabees 7:10, 13:17, 15:30, 16:14, 17:15, in Swete's text (lxx-swete, seated). At 13:17 the seated file reads Ἰσαὰι for Ἰσαάκ. The Slavonic speech is quoted in Thackeray's English (1928, (7)); Berendts and Grass give it at War 1.650.

[^31]: Eusebius, Historia ecclesiastica 3.10.6 (seated). 4 Maccabees is transmitted in manuscripts of Josephus, and early printed editions of his works included it.

[^32]: Sappho fr. 31.1–2, 7–9; Catullus 51.1–2, 6–7, 9, 13–16 (seated; the seated Catullus lacks 51.8); Slavonic War after 2.174, Meshcherskii (seated text, lines 6930–6936); Antiquities 18.63 (seated); Phil 2:7; John 1:14.

[^33]: Meshcherskii prints "не нарекуть", they will not call; his apparatus gives the Volokolamsk manuscript's "нареку", I will not call, the reading that keeps the first person of "немощно ми". The mirror holds on either reading, since the first person is already set by "ми".

[^34]: Eusebius, Historia ecclesiastica 2.17.22 (seated); Philo, De vita contemplativa 65.

[^35]: Philo, De vita contemplativa 78, Cohn–Wendland (seated), "σῶμα μὲν ἔχειν τὰς ῥητὰς διατάξεις"; Eusebius, Historia ecclesiastica 2.17.20 (seated), "σῶμα μὲν ἔχειν τὰς ῥητὰς λέξεις". Eusebius also reads "ἡ θρησκεία αὕτη" where the seated Philo has "ἡ λογικὴ ψυχή". Whether λέξεις is Eusebius's change or the reading of his Philo has not been checked against Cohn's apparatus.

[^36]: Epiphanius, Panarion 10.1.2, Holl (seated); the Anacephalaeosis numbers the Essenes the twelfth heresy, among the Samaritans. John of Damascus, De haeresibus 12, read in F. H. Chase's English (Writings, Fathers of the Church 37, 1958); the Greek has not been checked.

[^37]: Georgii Monachi Chronicon, ed. C. de Boor (Leipzig, 1904), vol. 1, 328–331, read on the printed pages (Internet Archive georgiimonachic01boorgoog, leaves 416–419). De Boor's apparatus marks the passage "excerpt. ex Jos. Bell. II, 119 sqq." and cites the Suda's entry on the Essenes. The Slavonic translation of George (V. M. Istrin's edition, 1920–1922) has not been collated here. That collation is the next test, since a Slavonic George would put the night office in Old Russian before the translator.

[^38]: Philo, De confusione linguarum 62, 146; De fuga et inventione 108, 110; De Cherubim 127; De vita Mosis II.117–118: Cohn–Wendland, seated (EA-CORPORA-01/07, #1505), read at #1758 (§§3, 7b, 16.5). In the seated De Cherubim 127 the last letter of κατεσκευάσθη is damaged in the source file and is supplied from the printed edition. Josephus, War 5.212–219 (seated, EA-CORPORA-01/01). Rev 1:13, 6:14, 11:19, Nestle 1904 (seated, EA-CORPORA-01/04). The Slavonic образ for εἰκών: Gen 1:26, "по образу нашему". Quis rerum divinarum heres 221–224 (seated). Rev 1:12, 1:16, 1:20, 2:5, 3:14, 15:5 (seated). Exodus 25:20 in Swete's LXX (seated, EA-CORPORA lxx-swete), 25:21 in the Hebrew numbering; the seated text reads κίβωτον and ἄν, OCR accents, given here as κιβωτὸν and ἂν.

[^39]: The census is Appendix IV. Its texts: N. A. Meshcherskii, История Иудейской войны Иосифа Флавия в древнерусском переводе (1958), the essay pp. 47–66 and the Old Russian text (seated, EA-CORPORA-37/01); A. Berendts and K. Grass, Flavius Josephus: Vom jüdischen Kriege, Buch I–IV, nach der slavischen Übersetzung (Dorpat, 1924–1927), the German translation, footnotes, introduction pp. 26–30 and table of contents (seated, EA-CORPORA-37/02); Thackeray 1928 (seated, EA-CORPORA-37/03); Josephus, War and Antiquities, Niese (seated, EA-CORPORA-01/01); Lamentations, Jeremiah and Psalms in Swete's LXX and Matthew in Nestle 1904 (seated). Grass's further additions are read in the Old Russian and in Berendts's German.

∮ = 1 + δ + SAPPHIC_LOGOS + SLAVONIC_GOSPEL + EIGHTFOLD_WHEEL + BECAME_TEXT

Crimson Hexagonal Archive — EA-LOGOS-01 v2.3

Anchored to r.01 Sappho (σ_S) — r.04 Catullus (σ_C) — r.07 Revelation (Ω circuit)

MANUS: Lee Sharks (ORCID: 0009-0000-1599-0703)