Historicity News: Thallus et Alius

I have a slew of things to report. I was thinking of doing some book reviews, for example, but I am not going to have the time. With my England trip coming up and my push to hunker down and finish On the Historicity of Jesus Christ, I will have much less time for blogging over the next two months. So I’m just going to summarize some things of late, including a new publication of mine, new books by others, and major events in the field, over the course of three posts.

First, my peer reviewed paper on Thallus has just been published (my paper on Josephus is soon to follow). The full citation is Richard Carrier, “Thallus and the Darkness at Christ’s Death,” Journal of Greco-Roman Christianity and Judaism 8 (2011-2012): 185-91. [It was available online, as part of Volume 8, as a downloadable PDF, but only until it appeared in print]. The conclusion is that Thallus never mentioned Jesus in any capacity, and must therefore be removed from all lists of authors attesting to Jesus. In fact, we have what is certainly a direct quotation of what Thallus said in Eusebius: that in the year 32 “the sun was eclipsed, Bithynia was struck by an earthquake, and in the city of Nicaea many buildings fell.”

If anyone wants to update Wikipedia’s article on Thallus to quote and/or cite this peer reviewed article, please feel free. It currently quotes my very old online essay on the matter; whereas the new paper is not only peer reviewed, but contains additional arguments confirming the conclusion, improves various points, and skips over unnecessary digressions.

Second, in yesterday’s post “Understanding Bayesian History” I responded to a scientist’s critique of my book Proving History, and he posted a well written reply in comments there, which I much appreciated and to which I have responded in kind, and that exchange makes a lot of things clearer, especially as to my objectives in writing PH and how to improve upon it, and regarding what his concerns actually were. I consider this a model of constructive dialogue, so it’s worth looking at.

Next I’ll report on two new books I’ve read that relate to the question of historicity.


  1. Sili says

    Is Mark Goodacre’s new book on Thomas relevant?

    If really does manage to show thatThomas is dependent on the synoptics, that further puts the lie to there being an early Jesus tradition. And of course, it destroys the ‘argument’ that “sayings gospels” were common and therefore Q is real. If we believe hard enough, of course.

    • says

      That’s true. But there are already a lot of experts who agree that Thomas is not independent of the synoptics, so I often don’t think to mention this, as it’s not as controversial as suggesting even Q didn’t exist. I’ve also not yet read it. But yes, if anyone thinks GThom supports the existence of Q, I’m told by people of good judgment who have seen advanced drafts that Goodacre’s book on that specific subject will correct them (it’s available for pre-order now: Thomas and the Gospels).

    • Sili says

      My copy arrived just this Thursday. I hope to read it on my way to New York next week.

      I’ve greatly enjoyed reading Goodacre since you pointed me to him, and I’ve found Goulder to be engaging as well.

  2. says

    I consider this a model of constructive dialogue, so it’s worth looking at.

    If only all discussions were so easy.

    But really: kudos to both you and Ian. I recognize that this is essentially a dialogue between two folks who agree, and merely have some issues in the details, but really, this is, as you say, a model of constructive dialogue.

    Now. If only you could get Ken Ham to be as agreeable as Ian. Then maybe we could get somewhere.

  3. malcolms says

    I know your purpose was to show that there was almost certainly no mention of Jesus in the original Thallus chronicle, but it seems that one could even go further: It looks like Phlegon is not an independent source but rather has just copied Thallus, with some embellishment about the eclipse. So ultimately all we are left with is one line in Thallus about an eclipse and earthquake in the same year (that being 32 CE if Phlegon is being accurately quoted by Eusebius by way of Syncellus et al.). Since Thallus was writing a history of a long period of time in only 3 volumes, it may even be the case that these 2 events were not even proximate to each other but just happened to occur in the same year.

    I also wonder if perhaps Syncellus is the source of the interpolation into Africanus, or rather, that it wasn’t an interpolation into the text per se but just Syncellus’s own parenthetical remark, not intended to look like part of the quote of Africanus. Is this a possibility when looking at the original Greek?

    • malcolms says

      Sorry, when I read your paper I somehow missed where you said that it appears Thallus had copied from Phlegon. I also agree that it seems more likely that Thallus would’ve abbreviated a longer description of the eclipse than that Phlegon would’ve enlarged a shorter one, since there is no apparent motivation for Phlegon to make something up (unlike, say, with sectarian religious writings), but I couldn’t rule out the possibility that the copying ran the other way. Just out of curiosity, from where did you get that estimate of Phlegon having written around 120-130 CE?

    • says

      All true.

      The date for Phlegon comes from standard references in the field and is a ballpark. I have not checked its merit, I just assume it’s the best estimate reached by the experts who have. It is asserted by a Byzantine encyclopedia and corroborated by many instances of his referring to contemporary persons and events of Hadrian’s time. One could get particular about which book was written when, however, and thus dates could range wider than 120-130, but not by more than a decade or two, and much less likely earlier dates than later. So I don’t rule out the possibility of proposing a different date for a particular reference. One would have to make an argument, though.

    • says

      I think it’s more likely the other way around (that Thallus is summarizing Phlegon). Although it’s possible Phlegon is exaggerating Thallus, it’s more common for a “brief compendium” covering a thousand years in three rolls to abbreviate prior chronologies, than for someone to embellish a chronological entry the way Phlegon would have (I’m looking at the historiography of chronography; although I grant, “more common” does not mean “always,” as I can adduce some counter-examples).

      You are right, though, that these events might not have been conjoined. I’d have to check the eclipse tracks for this period (they have been published, based on reverse calculations), but I seem to recall that this eclipse did occur in Turkey, where the earthquake is also placed, and if that’s correct, then it was a conjunction at least of region (Nicea was a metropolis of Bithynia, northern Turkey, although the description of the eclipse in Phlegon is probably literarily exaggerated–a scholar some time ago published a paper demonstrating authors tended to do that–in fact he also showed a tendency to align eclipses with earthquakes, supporting your suspicion that they were not chronologically conjoined).

      Not that it matters. Jesus isn’t mentioned by either author.

      As to how the Phlegon comment became interpolated into the quote of Africanus in Syncellus: that’s a possibility, and there are ways to evaluate your hypothesis (depending on what evidence survives); I just haven’t looked into it. I’ve seen an actual 17th century printed edition of Syncellus, but obviously that was many copies removed from his original.

      It was actually quite common for marginal notes exactly like you describe getting inserted into the text over the course of copying (we have hundreds of confirmed examples across the whole gamut of literature; I cite the scholarship on this in my next article, too, which proves this occurred in Josephus, and that represents the kind of research one would have to do on the text of Syncellus to test the same hypothesis here). But that could have happened to the text of Africanus used by Syncellus; it wasn’t necessarily Syncellus who produced the note. It could also have happened after Syncellus (if some monk annotating his copy of Syncellus put that in the margins, a later copyist could have inserted it in the transmission of Syncellus).

      I can only confirm that if there is still a manuscript of Syncellus in which the note is in the margins and not the text, it has not been consulted by any scholar producing critical editions of Syncellus (since they would remark on such a thing). That is unlikely, but not impossible. There are many texts for which we know manuscripts exist that have not been used by critical editors. I found this was particularly a problem with Epiphanius, for example, where I was able to get microfilms of specific manuscripts and confirm readings not noted in modern editions. And we know that sometimes there are manuscripts sitting in libraries or collections that aren’t even known to be there, although that’s rare (e.g. I think some new mss. of Josephus turned up recently, although they were judged to be derivative) and usually when that happens, it’s a palimpsest, i.e. an erased text underlying the ink of another text that we have (e.g. the Archimedes codex, or the famous case of the Syriac Sinaiticus). So one would have to first check if any manuscripts of Syncellus exist that weren’t consulted for critical editions, and even then another could turn up later. And to make matters more vexing, I have caught variants in manuscripts that were even used by critical editors yet those variants were not mentioned (sometimes, I have to conclude, deliberately, e.g. standard New Testament critical editions omit many readings in Codex Bezae for some reason, even very important ones). But something like this I cannot imagine being overlooked.

    • malcolms says

      Actually, what I had in mind with regard to Syncellus was not even a marginal note that got incorporated into the text (although that is of course a possibility, too), but just a parenthetical remark that he intended to place within the text. In modern books one would use brackets for this. My point was that Syncellus wasn’t producing a copy of Aficanus’s work but just quoting him. Presumably the text before and after the excerpt was Syncellus’s own writing. Why couldn’t he have added his thoughts in the middle of the excerpt as well?

    • says

      Presumably the text before and after the excerpt was Syncellus’s own writing. Why couldn’t he have added his thoughts in the middle of the excerpt as well?

      I see what you mean. Yes, that happens a lot, too. But not usually in the middle of a long quotation like this, and Syncellus especially not. Syncellus is producing a book full of excerpts, not making an argument and quoting others to support it (so he would not interfere with interjections); and indeed this excerpt just happened to mention a conjunction argument, which is not an argument Syncellus himself was making (it’s only an argument Africanus is making), so Syncellus would not interject to say this. Also, whoever inserted the remark got the facts wrong, whereas Syncellus elsewhere cites Phlegon correctly.

  4. says

    Richard Carrier: “I’d have to check the eclipse tracks for this period”.

    NASA has already done the hard work for us. This map shows all total and annular solar eclipse paths from 0021-0040. The blue paths are the total eclipses (“turning the day into such darkness of night that the stars could be seen in heaven“).

    There’s one that fits Phlegon’s description pretty well, since there was a total eclipse in modern Turkey 0029 Nov 24. You can follow this eclipse on Google Maps and see how it crossed Turkey and was close enough to Israel to cast darkness over all the land from the sixth hour. Of course it didn’t happen during Easter and it lasted only 1m59s, but then again Flegon never said differently.

    There was another eclipse with a date more appropriate to “the fourth year of the 202nd Olympiad“, namely 0033 Mar 19 (that’s also closer to Easter!), but this eclipse took place in the ocean close to Antarctica, so this is presumably not the one Phlegon had in mind. Kudos to this page for pointing me in the right direction).

    So it seems you are right in saying that Phlegon has chronologically conjoined two separate events.

    • says

      Oh that’s beautiful, thank you. I didn’t know they had done that. So sending that link is much appreciated. (That site can be supplemented with this one; and the map you link to is the easiest to read.)

      That eclipse in 29 would not cast Palestine into darkness. If you’ve ever seen a partial eclipse (even of 0.8 magnitude) you know it doesn’t really darken your location as much as you think and might even go unnoticed by someone not looking up (since it looks and feels just like a cloud passed in front of the sun, a common occurrence that doesn’t produce a nighttime experience or anything close). And of course it lasts only a few minutes (in this case, apparently just two). And as you also note, this one occurred in Winter (nowhere near Passover, which was always timed to the equinox).

    • Roger Viklund says

      I have dealt with all this in a Swedish blog post http://tinyurl.com/cv2ocak – and in machine translation into English here: http://tinyurl.com/d68mrhz . The most interesting illustration would be this image http://rogerviklund.files.wordpress.com/2010/12/bithynien2.jpg which I have arranged. That shows the only actual time for the eclipse reported by Phlegon and this would be on November 24 (according to the Julian calendar) in 29 CE. The red line represents the zenith of the eclipse (the place where the eclipse is most total and also lasts the longest), while the blue lines show the boundaries of where the total eclipse passes into partial. I have marked both the ancient Roman province of Pontus et Bithynia (of which Phlegon reported) and Jerusalem with its vicinity (where this eclipse was partial).

    • says

      Jobjorn, thank you for bringing this to my attention. I will compose this analysis in the third person for the benefits of my readers:

      I have now read Jobjorn Boman, “Comments on Carrier: Is Thallus Actually Quoted by Eusebius?” Liber Annuus 62 (2012): 319-25.

      Boman’s thesis is that rather than the Armenian end-date for Thallus being in error, Africanus was in error in claiming his reference came from Thallus, that instead he confused Phlegon with Thallus (he agrees with my conclusion that the passage referencing Phlegon in Africanus now in the text is a later interpolation), thus Thallus did not write anything about the first century AD, much less on Christ or eclipses or earthquakes of the period, that instead his three volume compendium ends with the second century BC (as the present Armenian text claims, and as all other references to Thallus corroborate though do not prove).

      This is an interesting alternative argument to the same conclusion (that Thallus affords no testimony to Jesus), one that grabs the other horn of the dilemma than I did.

      My argument was that if Thallus mentioned the event, Eusebius could not fail to have mentioned that fact in the Eusebian passage in question. Boman agrees. So there are only two ways to explain the text of Eusebius: (1) that my antecedent is true and therefore Eusebius must be quoting the line in Thallus in the Eusebian passage in question, therefore the brief remark he attributes to “other compendiums” must be a quotation of Thallus (or of a number of treatises, including Thallus, that repeat that same line); or (2) that my antecedent is false and Thallus never even mentioned this event (and that is why Eusebius doesn’t mention him in this location).

      Boman thus uses my same argument to take a different tack: if Eusebius is not quoting Thallus here (alone or among a number of chronologies that duplicate the same line), then Thallus did not mention the event at all (and therefore Africanus was mistaken in saying he did). So while I take the first horn (and assume Africanus was not mistaken), Boman takes the second horn, and concludes Thallus never mentioned the event at all (that in fact he never even covered any events in the first century AD).

      It’s a clever argument, and benefits from Boman’s analysis of the Armenian, which importantly updates our published knowledge of the material, making his paper required reading on the Thallus matter. It also picks up an argument I suggested myself once (that Africanus’ “third book of Thallus” was an error for “thirteenth book [of Phlegon],” since this so neatly explains why no one else ever cites Thallus for first century events and doesn’t require positing an error in the Armenian text’s dating of his volumes) and makes perhaps the best case one could ever make for it. I abandoned it myself because I couldn’t see any effective argument for it (sometimes the beautiful thesis just isn’t the strongest or most likely, and it’s hard to give it up). I ended up finding a much stronger case for the argument I eventually published instead.

      I have only two significant objections to Boman’s argument, one major and one minor:

      Minor issue:

      Boman makes an argument from silence that is unclear in its intended conclusion. The premise of which is that there are no references to Thallus in medieval Christian literature at all (which Boman implies is evidence Thallus did not mention the event), but he concedes this reference to Thallus had been in Africanus since the 3rd century, and was well enough known to Syncellus to have included it (a medieval author, which challenges the premise that no medieval authors referenced Thallus), thus the silence of later sources cannot argue for Thallus not having mentioned the event, since that silence existed even in the face of the fact that Africanus said he did, which is just as inexplicable (and any explanation of that silence will operate just as well for explaining it on the theory that Thallus did indeed say what Africanus says–after all, if no one noticed Africanus saying it, they can just as well have not noticed Thallus saying it, the more so as Thallus won’t have mentioned Christ in connection with the event anyway, thus his entry could easily be overlooked).

      Major issue:

      Boman’s thesis requires re-translating the Greek of Eusebius to the effect that he is summarizing Phlegon and then quoting Phlegon, rather than quoting “other compendiums”/”Thallus” and then quoting Phlegon. I find this wholly untenable. And that for three reasons.

      First, Boman proposes “the word “other” (allois) could have been written by Eusebius to emphasize that there were other testimonies than the Christian” but there is neither here nor before this any reference to Christian testimonies, so that conjecture is grammatically untenable. He proposes that the reference to “prophecies” foretelling the year of Christ’s passion is such, but one would not say “other” in respect to that unless you meant other prophecies. Eusebius would instead say simply “in Greek chronologies it is said…,” he would not say “in other Greek chronologies it is said…”

      Second, the passage says in the original Greek of Eusebius (preserved by Syncellus) that in these “other” chronologies “we find it recounted with this phrase” heuromen historeumena kata lexin tauta, which cannot then be followed by a summary or a paraphrase. This is an introduction of an exact quotation (kata lexin is an idiom for “as the phrase goes,” and with the modifier tauta [“this”] a specific phrase is being indicated). Which cannot therefore be a quotation of Phlegon.

      Third, after giving that reference, the text goes on to say graphei de kai Phlegôn ho tas olumpiadas peri tôn autôn en tôi ig´ rhêmasin autois tade, which translates to [emphasis added] “And also Phlegon, one who covered olympiads, wrote the following about these same things in the thirteenth volume in these words:…” [followed by the quotation of Phlegon]. The transition de kai and the fact that Phlegon is only now named and identified (as a recorder of olympiads, a different term than is used of the “others” before, who just write “records,” hypomnemata) entails Eusebius is not elaborating on what preceded but adding something to it (“Phlegon, too, said this”), likewise the other wording (what follows is “about the same things,” which means we were not hearing about Phlegon in the previous line but someone else who wrote of them; and “in these words” parallels “with this phrase,” stylistic variatio which indicates he is saying one person or group said the first exact quote, and Phlegon said the following exact quote, in each case separate exact quotations, not a summary/paraphrase followed by a quotation of what was just summarized/paraphrased).

    • J. Boman says

      Thank you very much, Dr. Carrier, for the review.

      Regarding the objections, I have the following comments.

      On the minor issue: I do not count Syncellus as ‘a medieval author referencing Thallus’, since he is only quoting Africanus verbatim (what we know). Just as Eusebius quoting Origen who “quotes” Josephus on the destruction of Jerusalem due to the slaying of James, does not prove that Eusebius knew the Josephus passage in question, Syncellus’ quote of Africanus doesn’t prove any medieval Christian knowledge of any Thallus passage about Easter miracles. It is still a fact that the only author explicitly referencing Thallus in this context, is Africanus, although other Christian authors had read Thallus work – only three volumes – and thus, arguably, would have had read the passage Africanus refers to.

      On the major issue:

      A problem with depending on the Greek is that Syncellus’ exerpt is more recent than the Latin translation of Jerome.

      Regarding your first reason, I say: I cite Lardner, who thinks that Eusebius “observes, that beside what is said by prophets, and by the evangelists, there are heathen authors who have born testimony to this darkness”. You say that “there is neither here nor before this any reference to Christian testimonies”, but that I believe is incorrect.

      The entry on the 202nd Olympiad begins (in Jerome): “Jesus Christ the son of God, preaching the way of salvation to all, performs the miracles which were written in the Gospels [“in Evangeliis scripta sunt”]. Jesus Christ the son of God, imparting the divine sacraments to his own disciples, commands that they announce (the opportunity and need for) conversion to God, to all peoples”. Then directly follows: “Jesus Christ, according to the prophecies, which had been spoken about him beforehand, came to the Passion in the 18th year of Tiberius, at which time also we find these things written verbatim in other commentaries of the gentiles [“quo tempore etiam in aliis ethnicorum commentariis haec ad verbum scripta repperimus”]” (quoted from the Tertullian Project).

      Ergo, there is a direct reference to the Christian Gospels just (not even 40 words) before the reference to “other” testimonies. I think we agree that to write “other commentaries” (aliis commentariis) wouldn’t make sense if nothing preludes the “other”, so I can only think this is a reference to the beforementioned Christian sources, i. e. the Gospels (in which “the prophecies which had been foretold about him [Jesus] beforehand” are noted).

      I still don’t believe that Lardner’s theory, according to which Eusebius’ “other heathen commentaries AND ALSO Phlegon” would be regarded as retorical, is sufficiently confuted.

    • J. Boman says

      I would like to elaborate my claim that “Eusebius’ ‘other heathen commentaries AND ALSO Phlegon’ would be regarded as r[h]etorical”.

      A main question must be: What was Eusebius’ purpose when he wrote about the 202nd Olympiad? To me, it is apparent that Eusebius wanted to validate the Christian testimonies (i.e. the Gospels), and his own dating of the Passion, by refering to non-Christian texts. The (Christian) readers would think “if even the heathens say that our Gospel is true, it must be”, and that was Eusebius’ intent.

      Eusebius’ text can be summarized as follows: 1) Jesus preached and performed miracles, according to the Gospels, 2) Jesus commanded the disciples to convert all peoples (according to the Gospels; cf. Matt 28:19-20), 3) Jesus came to the Passion, “in accordance with the prophecies” (which were noted in the Gospels), in the 18th year of Tiberius, 4) also in other commentaries (than the Gospels), written by pagans, it was written that an eclipse happened and that Bithynia was shaken by an earthquake and that many buildings in Nicaea collapsed, “all of which agree with what occurred in the Passion of the Saviour”, 5) also Phlegon, “an excellent calculator of olympiads”, writes about this (and then the quote from Phlehon), 6) that the Passion happened in this year is however proven by the Gospel of John, “in which it is written that after the 15th year of Tiberius Caesar [Luke 3:1], the Lord preached for three years”, 7) which is attested by Josephus, who “attests that around that time on the day of Pentecost, the priests first perceived an earth tremor and certain (loud) sounds”.

      I never ment to propose that “that the reference to ‘prophecies’ foretelling the year of Christ’s passion is” in itself a “reference to Christian testimonies”. I wrote that Eusebius wanted “to emphasize that there were other testimonies than the Christian, i.e., the testimonies which were believed to concur with the Old Testament prophesies [, …] thus confirming the Gospel story”. Lardner, whom I quote, refers to “what is said by prophets, and by the evangelists”. The explicit mention of the Gospels in the beginning of the text about the 202nd Olympiad, must be considered important to interpret the text. The Gospels must be considered as “historeumena” in Eusebius’ mind.

      Since the purpose of Eusebius was to “prove” that the Gospels and his dating of the Passion were correct, he needed to refer to as many sources as possible (“other” commentaries, Phlegon, Josephus). 4 above is therefore just a rhetorical summary of what is to come in 5, i.e. the “testimony” of Phlegon. It is analogous to a Christian scholar saying “There are many non-Christian sources in which it is written that ‘Jesus was crucified’ (see Josephus, Testmonium Flavianum)” – “many” sources are implied, but only one is actually cited. This kind of rhetorical bending of the truth, must be considered common. There would be no reason for Eusebius to omit references to Thallus or other pagan writers here, since Eusebius obviously wanted the readers to see that the Gospel story was well attested by non-Christian evidence.

      The simplest explanation is, I believe, that Eusebius was rhetorical, and did not cite any actual existing source, even if he used “heuromen historeumena kata lexin tauta”, and even though he said that “also” Phlegon wrote about the eclipse and other phenomena. “Also” I believe could also refer back to the beforementioned Gospels.

  5. Steve Watson says

    Hi Dr Carrier,

    I don’t know if you have mentioned this elsewhere but a PDF of your Thallus article can be downloaded from the Wikipedia Thallus entry you referenced, at note 8.

    8 ^ Carrier, Richard (2011–2012). “Thallus and the Darkness at Christ’s Death”. Journal of Greco-Roman Christianity and Judaism. Retrieved 14 August 2013.

    I am quite in awe of you, Moscow might be glad the Cold War has ended; finding you in his baffles would have quite ruined a Soviet submariner’s whole day!

    Chin Chin,
    Steven C Watson