Hostname: page-component-74656f9884-j7cmm Total loading time: 0 Render date: 2026-11-06T19:53:56.445Z Has data issue: false hasContentIssue false
  • English
  • Français

A Negative Testimonium?: A Response to Fernando Bermejo-Rubio

Published online by Cambridge University Press:  07 August 2025

Chrissy Hansen Open the ORCID record for Chrissy Hansen [Opens in a new window]
Chrissy Hansen*
Affiliation:
University of Nebraska-Kearney, Kearney, NE, USA
*
Email: cmehans2020@gmail.com
Rights & Permissions [Opens in a new window]

Abstract

This is a response to Fernando Bermejo-Rubio’s arguments that the original Testimonium Flavianum was a negative text. Bermejo-Rubio’s textual analysis ignores a great number of caveats and counterexamples that in the end render it far from certain or probable that the Testimonium Flavianum had a negative disposition towards Jesus. To the contrary, most of the terminology does not even appear to be problematic for a Christian scribe to interpolate. As a result, it is concluded that scholars are still stuck in the quagmire of attempting to arbitrate what is or is not authentic in the Testimonium Flavianum.

Keywords

EusebiusJosephusTestimonium Flavianumhistorical Jesus

Information

Type
Articles
Information
New Testament Studies , Volume 71 , Issue 1 , January 2025 , pp. 56 - 63
Check for updates
Copyright
© The Author(s), 2025. Published by Cambridge University Press.

1. A Negative Testimonium? A Response to Fernando Bermejo-Rubio

The Testimonium Flavianum (TF) is perhaps one of the most contested passages in all of antiquity, and it is too much to catalogue all of the ink which has been spilt on various questions surrounding it.Footnote 1 In recent years, the debate on the authenticity and original timbre of the TF (if it is authentic at all) has been reinvigorated, with several theories now being offered. Among the more intriguing is Fernando Bermejo-Rubio’s article, wherein he proposes that the original TF was perhaps negative in its characterisation of Jesus of Nazareth.Footnote 2 Giving a detailed philological analysis and overview of the various recensions and textual variants present in our copies of the TF, Bermejo-Rubio concludes that it is quite possible (he views it as even probable) that the original TF was in fact more inflammatory.

In this article, I will respond to Bermejo-Rubio’s arguments. I will not defend nor argue for a neutral original form of the text (nor any other reconstruction), but instead seek to evaluate if a negative interpretation is as probable as Bermejo-Rubio intimates. A closer inspection of the various phrases and terms which are offered as evidence of a ‘negative’ tone does not lead us to this conclusion. Instead, much of the language is ambiguous or has been misinterpreted. My ultimate conclusion is that while agreeing with Bermejo-Rubio that a neutral interpretation is unsatisfactory,Footnote 3 his proposed negative interpretation is also not convincing, as (A) the linguistic suggestions for negative terms are all ambiguous in reality, and (B) there is no language which could not be an interpolation either. In the end, it is perhaps futile to attempt to reconstruct an authentic TF of any kind, and perhaps more weight should be given to arguments for the total inauthenticity of the passage.

2. Τις in the Testimonium Flavianum?

Contained in a recension of Eusebius’ Historia Ecclesiastica 1.11 (Codex A) is the pronoun ‘certain’ (τις) after Jesus’ name, thus it reads ‘a certain Jesus’. Bermejo-Rubio argues the τις is original and that this is further corroborated by the Slavonic Jewish War containing a similar phrase (muži nĕkij). There are a few problems with this. Firstly, the Slavonic does not attest to τις by itself as we have it in Codex A, but would instead correspond more closely to ἀνήρ τις or ‘a certain man’. Codex A however has Ἰησοῦς τις σοφὸς ἀνήρ (‘a certain Jesus, a wise man’) instead. Secondly, as the Slavonic has undergone extensive alterations to lessen the overt Christian overtones of various passages (including its version of the TF), we should therefore dismiss the Slavonic’s significance here.Footnote 4

Further, there is no indication that τις must be taken pejoratively here, or in any of the passages which Bermejo-Rubio cites either.Footnote 5 For instance, Bermejo-Rubio cites Jewish War 2.433 for the usage of τις as negative in introducing Menahem, son of Judas the Galilean. However, it is not τις which has any negative character in 2.433. Instead, it is Josephus’ polemical description of him as σοφιστὴς δεινότατος. In none of the cases that Bermejo-Rubio cites does τις carry an inherent negative connotation.Footnote 6 Bermejo-Rubio admits Josephus often uses the term neutrally or even for figures he holds virtuous.Footnote 7 The pronoun τις is used in a positive and neutral sense by Greco-Roman authors.Footnote 8

Further reflecting on the Codex A variant, Bermejo-Rubio claims, ‘It is by far much easier to explain the dropping of a τις [from an original TF] by Christian hands, than its creation by a Christian copyist. The conjectural emendation [amending the TF to have τις as original] has, in fact, been widely accepted.’Footnote 9 However, Bermejo-Rubio seems rather unfamiliar with the wide Christian usage of τις on multiple occasions, including by Eusebius elsewhere. Acts 25.19 utilises this as an introductory term for Jesus placed in the mouth of a non-believer (τινος Ἰησοῦ).Footnote 10 Justin Martyr, Dialogue with Trypho 108 places it in the mouth of the fictional Trypho (Ἰησοῦ τινος). Eusebius, Praeparatio Evangelica 1.2.6Footnote 11 refers to hypothetical Jewish opponents using τις of the Messiah (τις εἴη Χριστός).Footnote 12 As such, it is clear that Christians were more than capable of imagining non-believers using various forms of τις as a way to introduce or talk about Jesus or the Messiah in general (negatively or notFootnote 13 ). Lastly, we know that outside of imagining opponents using the pronoun for Jesus, Christians could use it in a positive sense, see Ps.-Athanasius, Oratio quarta contra Arianos 36 (ὁ Χριστὸς τίς).Footnote 14 Therefore, it is easy to imagine (A) how this could have been part of a Christian interpolation into Antiquities by Christians at some later point, or (B) if it is ‘original’ (meaning it goes back to the oldest version of the TF), how it could have been a part of a wholesale fabrication of the TF (including even a positive or neutral version).

Christians could have interpolated this pronoun. If they saw the Eusebian TF and it appeared too Christian for a Jewish author to write, additions/subtractions could be made. In that case, they might have chosen to de-Christianise the passage with distancing language.Footnote 15 The fact that this pronoun occurs in a single manuscript of Eusebius’ Historia Ecclesiastica, and a text (Slavonic Jewish War) that was tampered with to de-Christianise itFootnote 16 does not raise confidence in its authenticity. In fact, it gives evidence that de-Christianisation is very likely here.Footnote 17

3. Other Negative Terminology?

Throughout the latter half of his article ‘Was the Hypothetical Vorlage of the Testimonium Flavianum a “Neutral” Text?’ (2014), Bermejo-Rubio attempts to find any hint of a negative tonality in several phrases and terms, but the suggestions are not compelling on closer inspection.

He claims that γίνεται δέ carries a negative connotation, as it introduces disturbances in various Josephan passages.Footnote 18 However, a cursory review of the usage of this term shows no such regular usage (Josephus, Ant. 1.213; 2.4; 3.134; 17.14; 20.51; 20.230 where it is used just for general and even positive events, like introducing the birth of children). The phrase γίνεται δέ has no inherently negative connotations at all. There were just disturbances that happened at specific times, which called for the usage of a stock phrase. Even if γίνεται δέ were read negatively, this intonation could derive from Pilate’s execution of Jesus (i.e., the execution was a bad moment in time). As such, this would be in line with a Christian interpolator introducing the tragedy of Jesus’ unfair execution.

Bermejo-Rubio asserts that ἡδονῇ carries negative connotations, but this is not always true either of Josephus (cf. Josephus, Ant. 18.59Footnote 19 ). There are positive usages of ἡδονῇ in Eusebius’ literature as well, which Olson notes (citing De Laudibus Constantini 17.11; De Martyribus Palaestinae 6.6; Commentaria in Psalmos on Ps 67.4).Footnote 20 Bermejo-Rubio is forced to admit that only around half the instances of διδάσκαλος have any negative connotations.Footnote 21 Christian scribes likewise had no particular qualms with this term.Footnote 22 Bermejo-Rubio claims φῦλον is dismissive or negative, but there is no such indication that Josephus exclusively uses it in this sense. Josephus himself uses it in quite neutral and even positive connotations (Ant. 1.221; 3.49; 3.105; 3.219; 3.258; 3.287; etc.).Footnote 23 Similar phraseology is found in Eusebius’ works as well, which illustrates a lack of Christian reservation with this term.Footnote 24 Eusebius’ usage of ‘tribe’ may often be negative, but it is not exclusively used this way.Footnote 25 As such, there is no reason that we must take this in a negative sense here.

Bermejo-Rubio asserts that ἐπάγομαι could carry a negative connotation. Despite his claim that it might be interpreted as ‘led astray’, it often does also just mean to bring people over, or to bring people with oneself in an insignificant way (Ant. 17.219; 20.55; 20.149). Contra Bermejo-Rubio, the verb does not always carry a negative meaning. It is context dependent, and it could go either way, since we have had no clear-cut case of negative language exhibited thus far. While Bermejo-Rubio opines that this negative connotation is ‘overlooked’,Footnote 26 this is because the negative reading is not compelling here.

Bermejo-Rubio’s suggestion that ἀγαπάω may carry a negative connotation in this passage, denoting that Jesus’ followers are blindly following a criminal (this is a bad or faulty ‘love’), is based on a single parallel in Jewish War 1.171, where the term might imply some ironic or false appreciation.Footnote 27 Josephus does use the term with an ironic tone as well (Ant. 3.20). However, Josephus also uses it positively and neutrally as well (Ant. 1.99 where God loves Noah; 2.23 where God’s love will be conditional but is still a positive outcome; 5.342 describing Elkanah’s love of Penninah; 6.206 Jonathan’s love for David, etc.). Whether Josephus’ connotations are negative or not, what is clear is that Christians have a very positive history with the verb ἀγαπάω in their writings.Footnote 28 This would be very much something they could have interpolated without issue. As such, this mere possibility of a negative Josephan meaning does little to undermine the hypothesis of Christian interpolation.

One last suggestion is that the term Χριστιανοί may have carried seditious undertones due to its Roman origin as a group designation.Footnote 29 However, the ‘seditious’ claim on Bermejo-Rubio’s part is not evidenced. Tacitus, Annals 15.44 does not assign a pejorative sense of seditiousness to the term Christianus (or Chrestianus in this case), but instead contends that they are a problem due to their odio humani generis (‘hatred of humanity’). This phrase seems to denote a perceived anti-social behaviour, not seditiousness.Footnote 30 Tacitus, Pliny the Younger and Suetonius describe these Christians as a ‘contagion’ (contagio) and ‘superstition’ (superstitio), but the only one who even approaches a seditious undertone (and not in a violent sense) is Pliny, who describes that they were causing economic problems in the region (Ep. 10.96).Footnote 31 Close to Josephus’ time there is no clear evidence that Romans regarded these Christiani as being a seditious faction, but more just a general public nuisance at worst. Even into the late second century, and in direct contrast to the seditious model, Galen’s fragments on Christians describe them as a philosophical group and even ascribe a level of virtuousness to them.Footnote 32 Given this, we cannot presume there is a seditious element behind the term. For Bermejo-Rubio to be correct, one would have to assume the authenticity of the TF and Χριστιανοί in particular.Footnote 33

This summarises the strongest evidence which Bermejo-Rubio provides on the issue. None of the terminology points to any negative connotations on close inspection and as a result, while Bermejo-Rubio’s reading is possible, nothing makes it probable. Bermejo-Rubio’s final argument that a negative reading best explains why Christians would mutilate the passage later can be met with similar difficulties. Bermejo-Rubio states, ‘To start with, one can wonder why, if the original passage truly had a neutral tone, the Christians would have felt it necessary to tamper with it.’Footnote 34 This response presumes there is an authentic core at all; but more to the point, Bermejo-Rubio ignores the context in which the TF is first attested, Eusebius’ Historia Ecclesiastica 1.11. In this passage, Eusebius wants this non-Christian Jewish source to stand as testimony against those who despise Christ and Christians, and further give no excuse to Jews who refuse Christ.Footnote 35 A neutral passage would be far less effective than a positive one, and so there would be every reason to spruce it up. This is even clearer in Demonstratio Evangelica 3.5, the second known quotation, which again highlights Eusebius’ apologetic concerns in citing this passage.

4. Conclusions

This presents a general problem with TF reconstructions. Bermejo-Rubio’s comments on Meier’s neutral reconstruction, which hinges on omitting certain phrases, are decisive in my view, as it makes less sense of the passage when one omits the portions Meier suggested were interpolations.Footnote 36 But Bermejo-Rubio’s own reinterpretations are far from compelling, and none of his arguments show that a negative reading is very probable. It is possible to read the text negatively, but it is also quite possible to retain positive connotations as well. By far, Bermejo-Rubio’s strongest piece of evidence is τις in Codex A (and possibly the Slavonic), but as noted before, even this does not pass scrutiny. Other models of reconstruction range from unlikely to unreasonable.Footnote 37

Bermejo-Rubio gives precise arguments for disavowing the neutral reconstruction by Meier, but in the absence of a convincing alternative reconstruction, I do not think any model is worth supporting. Given the current state of the TF and our evidence at hand, which can be twisted to support any reconstruction, the simplest ‘reconstruction’, and the one most consistent with our available evidence is an irrecoverable TF (even presuming there is anything authentic at all).Footnote 38 As Steve Mason has remarked in The Jesus Handbook: ‘Trying to figure out what Josephus wrote in the Jesus passage (book 18) is probably impossible now, given that all of our Greek manuscripts (from the tenth century and later) more or less agree on the passage we read in our texts.’Footnote 39 In this way, all reconstruction efforts one way or another are futile. The other alternative, which I am partial to, is that the TF is inauthentic in toto, which in light of the problems with reconstruction, I believe should be re-examined.

Competing interests

The author acknowledges none.

References

1 For detailed bibliography and historiography, see Feldman, Louis H., Josephus and Modern Scholarship (1937–1980) (Berlin: Walter de Gruyter, 1984)CrossRefGoogle Scholar; Whealey, Alice, Josephus on Jesus: The Testimonium Flavianum Controversy from Late Antiquity to Modern Times: 36 (Studies in Biblical Literature; Frankfurt am Main: Peter Lang, 2003)Google Scholar. For a few more recent treatments, see Hopper, Paul, ‘A Narrative Anomaly in Josephus: Jewish Antiquities xviii:63’, Linguistics and Literary Studies: Interfaces, Encounters, Transfers (ed. Fludernik, Monika and Jacob, Daniel; Berlin: Walter de Gruyter, 2014) Google Scholar; Feldman, Louis, ‘On the Authenticity of the Testimonium Flavianum Attributed to Josephus’, New Perspectives on Jewish-Christian Relations: In Honor of David Berger (ed. Carlebach, Elisheva and Schachter, Jacob J.; Leiden: Brill, 2012) 13–30Google Scholar; Allen, Nicholas P. L., Christian Forgery in Jewish Antiquities: Josephus Interrupted (Newcastle Upon Tyne: Cambridge Scholars Publishing, 2020) 98–228Google Scholar; Ivan Prchlík, ‘Ježíš řečený Christos‘ u Iosepha Flavia: Jistota nejistoty’, Antica Slavica (ed. Peter Fraňo and Michal Habaj; Trnava: Univerzita sv. Cyrila a Metoda v Trnave 2018) 77–152 and 280–6; Detering, Hermann, Falsche Zeugen: Außerchristliche Jesuszeugnisse auf dem Prüfstand (Aschaffenburg: Alibri Verlag, 2011) 19–41Google Scholar; Goldberg, Gary J., ‘Josephus’s Paraphrase Style and the Testimonium Flavianum’, Journal for the Study of the Historical Jesus 20 (2022) 1–32CrossRefGoogle Scholar; Tibor, Grüll, ‘Josephus és a Messiás: A testimonium Flavianum eredetiségének kérdése’, Studia Biblica - Bibliai Tanulmányok 2 (2020) 9–63.Google Scholar

2 Bermejo-Rubio, Fernando, ‘Was the Hypothetical Vorlage of the Testimonium Flavianum a “Neutral” Text? Challenging the Common Wisdom on Antiquitates Judaicae 18.63–64’, Journal for the Study of Judaism 45 (2014) CrossRefGoogle Scholar. He has since reasserted it was a negative text in They Suffered under Pontius Pilate: Jewish Anti-Roman Resistance and the Crosses at Golgotha (Lanham: Rowman & Littlefield, 2023), Kindle edition, location 1026.

3 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 331–47. See also his incisive remarks in Fernando Bermejo-Rubio, ‘La naturaleza del texto original del Testimonium Flavianum: Una crítica de la propuesta de John P. Meier’, Estudios Bíblicos 72 (2014) 257–92.

4 N.A. Meščerskij, ‘Introduction’, Josephus’ Jewish War and its Slavonic Version: A Synoptic Comparison of the English Translation by H.St.J. Thackeray with the Critical Edition by N.A. Meščerskij of the Slavonic Version in the Vilna Manuscript translated into English by H. Leeming and L. Osinkina (ed. H. Leeming and K. Leeming; Leiden: Brill, 2003) 1–106, at 19. The Slavonic’s TF is cribbed from George Hamartolos, who in turn utilised Eusebius. As such, we should dismiss its utility on this count, as it may not present an independent attestation of (ἀνήρ) τις being authentic but could just be indirectly relying on Eusebius instead.

5 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 358 n.152.

6 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 358 n. 152 incorrectly cites the Menahem passage as ‘Ant. 17.433’. There are multiple defective citations in this footnote. Ant. 19.273 at no point refers to Simon (contra Bermejo-Rubio), nor does τις occur anywhere there. Instead, Bermejo-Rubio must be referencing Ant. 19.332, where again τις is not being used negatively. It is when Josephus describes the insolence of Simon after that any negative intonation appears in the text. In general, it is rather neutral toward Simon. Bermejo-Rubio cites Ant. 20.97, never giving a name. Here he is referencing Theudas, but here τις is not negative by itself but is made negative because of Theudas being characterised as a ‘magus’ (γόης τις), a term with often negative connotations, see Eric Vanden Eykel, The Magi: Who They Were, How They’ve Been Remembered, and Why They Still Fascinate (Minneapolis: Fortress, 2022) 42–56. As Philip Edward Jewell, ‘Magic in the Works of Flavius Josephus’, (PhD Dissertation; University of Southampton, 2006) 107 shows, most instances of γόης in Josephus’ works are negative. Bermejo-Rubio references a few other places in Jewish War. He cites Jewish War 2.60 where Athronges is called ‘a certain shepherd’ (ποιμήν τις). This is not a negative usage of τις. He then cites Jewish War 2.433, which we have discussed above (see errant citation of a ‘17.433’). He cites 2.599 and a ‘certain Jesus’ (Ἰησοῦς τις), but τις is introductory, and the negative intonation of the passage arises from the description of Jesus and another John as agitating the crowds (παρώξυνεν δὲ τοὺς πολλούς). Jewish War 3.229 works against Bermejo-Rubio’s thesis of τις carrying negative connotations because the whole of 3.229–33 is describing the valour and courage of a ‘certain man’ (ἀνήρ τις) and his brothers. None of Bermejo-Rubio’s cited examples demonstrates his point, and one of them disproves his supposition.

7 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 358. For just a handful of counter-examples, see Josephus, Ant. 9.10; 10.67; 14.22; 16.231; 16.387; 18.91; 18.195.

8 Pindar, Olympian 1.100; Plutarch, Life of Agis 5; Herodotus, Histories 3.34; Sophocles, Philoctetes 440.

9 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 358. The few examples that Bermejo-Rubio cites do not indicate this pronoun has been ‘widely accepted’ as authentic, and most of the citations given are to texts from over twenty years ago (many of them forty or more years older). Thus, it does not reflect current trends.

10 Bardet, Serge, Le Testimonium Flavianum: Examine historique considerations historiographiques (Paris: Les Éditions du Cerf, 2002)Google Scholar, 90 considers Acts 25.19 to refute the previous theories of Eisler (and thereby Bermejo-Rubio), who proposed the authenticity of τις on the basis of Codex A and the Slavonic. Bardet states, ‘Il est à craindre qu’il ne se trompe une fois encore, si l’on considère les Actes des Apôtres (25.19) […]’ (90). Bermejo-Rubio while citing Bardet on the usage of τις (‘Was the Hypothetical Vorlage’, 358 n. 15) ignores the implications of Acts 25.19 for his argumentation and does no further survey of early Christian literature on the issue.

11 Thanks to Ken Olson for this reference.

12 Other examples include the Acta Thomae 23 (Ἰησοῦ τινος) where Jesus appears to a merchant and sells his brother Thomas as a slave. For discussion, see A.F.J. Klijn, The Acts of Thomas: Introduction, Text, and Commentary (NovTSup; Leiden: Brill, 20032) 68–9. See also the Acta Philippi 46 (Καὶ διδάσκει πιστεύειν εἰς ὄνομά τινος Ἰησοῦ). In François Bovon and Christopher R. Matthews, The Acts of Philip: A New Translation (Waco: Baylor University Press, 2012) 58 this is numbered Acta Philippi 5.5. For others using ‘Christ’ instead, see Ps.-Athanasius, Oratio quarta contra Arianos 36 (ὁ Χριστὸς τίς), for commentary and German translation see Markus Vinzent, Pseudo-Athanasius, Contra Arianos IV: Eine Schrift gegen Asterius von Kappadokien, Eusebius von Cäsarea, Markell von Ankyra und Photin von Sirmium (Leiden: Brill, 1996) 117–18. In a more general sense, also see Eusebius, Generalis elementaria introductio 3.46 (ἐν ᾧ χριστός τις καὶ ἀρχιερεὺς κατὰ διαδοχὴν ἡγούμενος τῶν τοῦ λαοῦ προστήσεται πραγμάτων). We can also note that this was used to introduce the prophet Joshua as well, see Gregory Nazanius, Oration 2; Justin Martyr, Dialogue with Trypho 115.

13 Dave Allen, ‘How Josephus Really Viewed Jesus’, Revista Bíblica 85 (2023) 333–57, at 345–6 has attempted to rescue the term, claiming ‘While this word tis made no difference to Christian scribes who use it for heroes or villains, it just so happens that Josephus often used this descriptive to say somebody was unimportant. A certain so and so. It was probably common knowledge in Justin Martyrs time that Josephus did in fact use tis.’ However, this is also untrue. The term is simply introductory. He likewise introduces kings and rulers in this way, see Ant. 1.158; 5.182–4; 5.257; 5.276; 6.45; 7.101. Perhaps one of the most noteworthy is from Ant. 8.205, where it is used to introduce the future king, Jeroboam.

14 Here Ps.-Athanasius is declaring that there is just one Christ, thus, identifying this as Jesus. For commentary and German translation, see Vinzent, Pseudo-Athanasius, 117–18.

15 And this would not be unheard of in recensions of Josephus’ works. As noted above, this was done in the Separated Edition of the Slavonic Jewish War. Likewise, as Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 340–1 relays that there are potential issues with Agapius’ text including that it may have downplayed Christianised language due to the Christian-Muslim debates, and that there were errors in transmission as well. One can also compare the Sozomen’s Historia Ecclesiastica 1.1 which dampens all of the TF (deriving it from Eusebius). See also John Malalas, Chronographia 10.26 and Cedrenus, Compendium Historiarum.

16 Meščerskij, ‘Introduction’, 19.

17 A similar issue may be behind the ‘believed/thought to be the Christ’ recensions in the Latin and Slavonic traditions, where the original passage reads ‘he was the Christ’. Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 335–6 takes this as the authentic reading. For detailed discussions of these texts, see David B. Levenson and Thomas R. Martin, ‘The Latin Translations of Josephus on Jesus, John the Baptist, and James: Critical Texts of the Latin Translation of the Antiquities and Rufinus’ Translation of Eusebius’ Ecclesiastical History Based on Manuscripts and Early Printed Editions’, Journal for the Study of Judaism 45 (2014) 1–79 and Whealey, Alice, ‘The Testimonium Flavianum in Syriac and Arabic’, NTS 54 (2008) CrossRefGoogle Scholar. As Levenson and Martin, ‘The Antiquities’, 25–6 note, Rufinus’ text originally read ‘he was the Christ’ but was altered to correspond to Jerome’s version later, an overt choice to de-Christianise the Latin tradition. Jerome’s version (the source of the Rufinus alterations) derives from Eusebius. It is quite possible Jerome’s version was also contaminated by influence from Ps.-Hegesippus’ De Excidio 2.12, see Richard Matthew Pollard, ‘The De Excidio of “Hegesippus” and the Reception of Josephus in the Early Middle Ages’, Viator 46 (2015) 65–100, at 83. De Excidio likewise derives from Eusebius, see David J. DeVore, ‘On the Fourth-Century Reception of Eusebius’s Ecclesiastical History’, Church History 92 (2023) 644–50, at 648; Agnès Molinier-Arbo, ‘Crime et châtiment des Juifs. Réminiscences d’Eusèbe de Césarée dans les histoires du pseudo-Hégésippe’, Revue des études latines 99 (2021) 161–81. The ‘he was possibly the Christ’ found in Michael the Syrian and Agapius were de-Christianisations, perhaps influenced by Muslim courts and pressures of the time, see Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 340–1. The Syriac translation of Eusebius reads ‘he was the Christ’, which means the version that went into Semitic translations was then altered to de-Christianise it in quotations by Michael and Agapius. Origen’s claim that Josephus did not believe Jesus was Christ (Contra Celsum 1.47) does not necessitate even seeing a TF, as he makes similar claims of Jews and other non-Christians in general, cf. Contra Celsum 2.9, 2.38, 6.41. De-Christianisation of this passage also occurred in Cedrenus, Compendium Historiarum, who renders the passage as: ὁ Χριστὸς οὗτος ἦν to πολλοὺς γὰρ καὶ ἀπὸ Ἑλλήνων ἠγάγετο Χριστός.

18 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 353.

19 Curran, John, ‘“To Be or to Be Thought to Be”: The Testimonium Flavianum (Again)’, Novum Testamentum 59 (2017) 71–94, at 89 n. 83CrossRefGoogle Scholar asserts that there is a single positive usage of ἡδονῇ in Josephus (Ant. 18.59). However, further positive or at least neutral usages can be found as well. For instance, 18.274 where it refers to the courage of the Jews in being unwilling to let their laws be usurped; 18.289 where it refers to Agrippa’s ability to gain favour with Gaius, which is doubled in 18.291. Earlier in Antiquities, there is also 11.167 which refers to Nehemiah’s pleasure he received from the king, which dissipated his sadness. See also another positive instance in 12.324.

20 Olson, Ken, ‘A Eusebian Reading of the Testimonium Flavianum’, Eusebius of Caesarea: Tradition and Innovation (ed. Johnson, Aaron and Scott, Jeremy; Cambridge: Harvard University Press, 2013) 97–114, at 104.Google Scholar

21 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 354.

22 There are dozens of instances of this term appearing in Eusebius’ works alone. Eusebius, Demonstratio Evangelica 1.6 in fact uses this of Jesus (τίνες θ’ ἡμεῖς καὶ ὁποῖος ὁ τῶν τοιῶνδε λόγων τε καὶ μαθημάτων διδάσκαλος, αὐτὸς ὁ σωτὴρ καὶ κύριος ἡμῶν Ἰησοῦς ὁ Χριστὸς τοῦ θεοῦ). For a more thorough discussion, see Olson, ‘A Eusebian Reading of the Testimonium Flavianum’, 104–5. Another notable parallel is Acta Philippi 46 which has at least three terms of overlap with Codex A of Eusebius’ Historia Ecclesiastica 1.11.

23 He also uses it in ways that have little relation, such as a ‘tribe of locusts’ for one of the Egyptian plagues (Ant. 2.306). In this case, the devastation of the locusts is more the negative trait here. The ‘tribe of locusts’ is itself an act of God.

24 For other non-negative usages, see Demonstratio Evangelica 5.3 where he says that Christ did not derive from the priestly tribe (φυλῆς). See also 1.3.45 where Eusebius describes Christ deriving from the tribe (φυλῆς) of Judah. A more neutral usage of φῦλον is also in Eusebius, Commentary on the Psalms 23.565.1 (numbering according to Thesaurus Linguae Graece database, which is using Migne’s edition here). Eusebius wrote: Αὕτη δὲ ἦν ἡ πάλαι ἀφωρισμένη τῷ Ἰωσὴφ μερὶς κατ’ ǰξαίρετον, ἐφ’ ᾗ καὶ μέγα ἐφρόνουν πρότερον μὲν Ἰουδαίων παῖδες, νῦν δὲ τὸ Σαμαρειτῶν φῦλον. Here is Σαμαρειτῶν in combination with φῦλον in the genitive plural. Alice Whealey, ‘Josephus, Eusebius of Caesarea, and the Testimonium Flavianum’, Josephus und das Neue Testament (ed. Christfried Böttrich and Jens Herzer; Tübingen: Mohr Siebeck, 2007) 73–117, at 99 emphasises the genitive-plural as carrying a particular negative connotation for Eusebius, and as such Eusebius would not have fabricated a TF with this construction. She neglects this counter-example. Whealey’s attempt to find negative meanings inherent to specific grammatical forms in Eusebius is also questionable, see Olson, ‘A Eusebian Reading of the Testimonium Flavianum’, 104 n. 27. Whealey and others cannot exclude counter-examples due to them being in the genitive singular, as much as they would try. The form φυλῆς is frequent in the New Testament as well (Luke 2.36; Rom 11.1; Phil 3.5; Heb 7.13, etc.).

25 As Olson remarks, Eusebius was not ‘polemicizing against stars when he refers to their “tribes and families” in Preparation 7.15.12’, see Olson, ‘A Eusebian Reading of the Testimonium Flavianum’, 109 n. 43. Allen, Dave, ‘A Model Reconstruction of What Josephus Would Have Realistically Written About Jesus’, Journal of Greco-Roman Christianity and Judaism 18 (2022) Google Scholar makes a similar error here as Bermejo-Rubio. Whealey, ‘Josephus’, 97 argues that Eusebius’ usage of φῦλον for Christians stems from Tertullian and is therefore immaterial (i.e., Eusebius would not use the term normally, but does so because he is utilising Tertullian in this instance). Whealey’s argument is unconvincing. That Eusebius uses φῦλον under the influence of Tertullian shows that he had an exemplar and that under such influence he was fine using it. It is irrelevant where it stems from, Eusebius still finds it acceptable. Further, even if Eusebius’ usage of φῦλον is negative, this does not compel us to think he would not have used it as a basis for interpolation. To think that Christians could not invent unfavourable portrayals of themselves for their own apologetic purposes is unevidenced, and in fact contradicted by the way they construct fake opponents using such derogatory or distanced language, such as Justin’s Dialogue with Trypho.

26 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 355.

27 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 355–6.

28 The literature on this subject is beyond my ability to summarise here. Recently, we can point to just the several studies being done on the Gospel of John’s usage of this, see Paul Aaron Hines, ‘Loving Wisdom: The Ἀγαπάω-Φιλέω Exchange in John 21:15–17 as an Allusion to LXX Proverbs 8:17’, Bulletin for Biblical Research 30 (2020) 379–402; David Shepherd, ‘“Do You Love Me?” A Narrative-Critical Reappraisal of ἀγαπάω and φιλέω in John 21:15–17’, Journal of Biblical Literature 129 (2010) 777–92.

29 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 360.

30 For discussion of this phrase, see Barrett, Anthony A., Rome is Burning: Nero and the Fire That Ended a Dynasty (Princeton: Princeton University Press, 2020) .Google Scholar

31 That Seutonius’ Chrestus refers to Jesus is doubtful, see Williams, Margaret H., Early Classical Authors on Jesus (London: T&T Clark, 2022) Google Scholar. See also, Richard Carrier, ‘The Prospect of a Christian Interpolation in Tacitus, Annals 15.44, Vigiliae Christianae 68 (2014) 264–83.

32 For discussion of the fragments of Galen’s work, see Flemming, Rebecca, ‘Galen and the Christians: Texts and Authority in the Second Century AD’, Christianity in the Second Century: Themes and Developments (ed. Paget, James Carleton and Lieu, Judith; Cambridge: Cambridge University Press, 2017) .Google Scholar

33 See above notes on Eusebius’ use of φῦλον.

34 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 361–2.

35 Thus, Eusebius, Historia Ecclesiastica 1.11.9 remarks Jews earn their derision since even Josephus, one of them, admits the truth of Christ.

36 Bermejo-Rubio, ‘Was the Hypothetical Vorlage’, 331–47.

37 Victor, Ulrich, ‘Das Testimonium Flavianum: Ein authentischer Text des Josephus’, Novum Testamentum 52 (2010) 72–82CrossRefGoogle Scholar proposes it was all authentic, but the arguments are strenuous at best. Partial to this is also van Henten, Jan Willem, ‘Testimonium Flavianum’, The Reception of Jesus in the First Three Centuries (ed. Keith, Chris, Bond, Helen K., Jacobi, Christine and Schröter, Jens; 3 vols.; London: Bloomsbury, 2020) Google Scholar. By far the least likely proposals have been from Dave Allen, see Allen, Dave, ‘An Original Negative Testimonium’, Journal of Higher Criticism 15 (2020) 67–90Google Scholar. He has since altered his reconstruction multiple times, see Allen, ‘A Model Reconstruction’ (cited above) and ‘How Josephus Really Viewed Jesus’. Allen’s reconstruction however is more a series of strenuous thought experiments, see Hansen, C.M., ‘A Response to David Allen’s “A Model Reconstruction of What Josephus Would Have Realistically Written About Jesus”’, Journal of Greco-Roman Christianity and Judaism 19 (2023) 94–103.Google Scholar

38 Several scholars have posed this issue, see Sanders, E.P., ‘Jesus Christ’, Eerdmans Dictionary of the Bible (ed. David Noel Freedman; Grand Rapids: Eerdmans, 2000) Google Scholar; Joseph Hoffmann, R., Jesus Outside the Gospels (Buffalo: Prometheus Books, 1984) 54Google Scholar; T, R.. France, The Evidence For Jesus (Vancouver: Regent College, 2006) 30Google Scholar; Williams, Early Classical Authors on Jesus, 54.

39 Mason, Steve, ‘Non-Christian Texts’, The Jesus Handbook (ed. Schröter, Jens and Jacobi, Christine; trans. Brawley, Robert L.; Grand Rapids: Eerdmans, 2022) , at 165.Google Scholar