I just published a major new article in a peer-reviewed journal, the Canadian-American Theological Review: “Did Any Pre-Christian Jews Expect a Dying-and-Rising Messiah?” It will be the lead article in vol. 14, their delayed 2025 issue (print copies are going out to that journal’s subscribers in a few weeks, and digital versions will be available in maybe a few months). In twenty pages it finds the answer is pretty much yes—and that the vast majority of the dozens of experts who’ve explored the question agree. The folks who erroneously believed a certain YouTube guy with a goatee and a tuque will be outraged and overwhelmed by the support I found for Element 5 of my original study On the Historicity of Jesus: a dying messiah was a live option within certain fringe groups of Jews, out of which Christianity most likely arose, not with any radical new idea, but with the slickest implementation of an already old one.
That these messianic expectations definitely or plausibly existed for at least a century prior to Christianity (alongside more usual ideas of an invincible conqueror) is unquestionably the mainstream consensus now—when measured by the proportion of actual specialists investigating the matter and publishing their results or actually reading them (and not those who didn’t). I’ll be discussing it all with Godless Engineer on his show this Friday (7 August 2026), and we’ll be taking questions from the audience, so join us and hear all about it!
Discover more from Richard Carrier Blogs
Subscribe to get the latest posts sent to your email.
Mazel Tov on your new paper publication and your upcoming talk with GE.
Congrats!
Why didn’t this come up in Josephus? He is quite thorough about Jewish sects that where around that time.
Josephus conspicuously avoids ever discussing even the concept of a messiah. So he certainly wouldn’t have gotten into the details of any messianic ideology.
Of course, he was also a mainstream Pharisee, and so would not likely have favored anti-nationalist messianism anyway. But he does describe several coded examples of messianic movements (see Element 4 in OHJ), always with despite. Which looks far too much in service to the interests of his Roman captors and benefactors.
One point of comparison is how Josephus hugely oversimplifies the sectarianism (he really only describes three sects, the Essenes, Sadducees, and Pharisees), and avoids nuance entirely (e.g. we know from Hippolytus that the Essenes were split into some nine sects and actually ideologically descended from Samaritans; we know from many other sources that the Pharisees were split into at least two major factions at ideological war with each other, the liberals and conservatives; and we know of some several dozen other sects he never even mentions much less describes).
Speaking of Essenes… I see claims that the Dead Sea scrolls are nowadays not considered the product or property of an insular Essene commune, but rather (based on analysis of physical media, particularly animal skins) come from all over Palestine, where documents with the name (or words of?) the deity had to be disposed of by depositing them somewhere designated for the purpose. That would make the scrolls more or less specialized trash. Is this a fever dream, or mainstream archaeology?
How well do we know Simon b Kochba/Kosēba? His Messiah chops seem better then Jesus’. Was he considered a dying/rising one too – this particular belief must’ve matured and stabilised by his time, mustn’t it?
Bar Kochba went all-in on the militarily victorious messiah model, so his followers could not rescue his movement by pivoting to the dying and rising model.
Hence we see he is declared a false messiah in the Talmud, reflecting the overwhelming internecine propaganda against him after his promises fell through.
If Jesus existed (and especially if he didn’t) he does not appear to have ever made any such promises of rapid military success (see my debate with Fernando-Rubio for why attempts to claim otherwise aren’t credible), so he could more easily be pivoted into a martyr rather than a failure.
I also wonder if the size of the groups would have something to do with it.
A more mainstream group with more concrete promises like that of a bar Kochba is going to find it harder, as a whole group, to innovate, and also are going to have fewer people who are cult-brained enough to need to do that.
In contrast, Christianity was a doomsday cult, attracting the small, fanatical personalities that always are attracted by those cults. Those folks are always more able to innovate.
What I would then say as regards the zealot hypothesis is precisely that it actually makes less sense at that point that the cult was a weird Zealot splinter cell. Their profile as a group seems much closer from the evidence to the kind of doomsday cult that attracts personalities that are less violently militant and more willing to embrace a more abstract, spiritual salvation from their problems.
When I first came across your work, I was mostly convinced but still waiting for a retort. So I was still using cautious language when discussing Jesus’ historicity. The scholar you mention (he’s surprisingly more than a YouTube guy) is the one that made me give up hope that mainstream scholars would deal fairly with your work and I have become much less cautious in endorsing mythicism.
Fair.
Note this present subject does not verify mythicism, though. It’s not evidence either way. But it threatens traditional exceptionalist and triumphalist narratives of Christianity still being sold even by secular scholars now. If Christianity was built on an already existing trend of inventing dying messiahs (as the evidence well enough shows), then it was not exceptional and didn’t innovate in any remarkable way. That threatens dogmas other than whether he existed.
Although some scholars do try to argue that he must have existed because how else would Jews think of a dying messiah unless he actually died, and this kicks the legs out from under that one argument, it was never a logically sound argument to begin with. Any idea to invent a scriptural justification post hoc demonstrates the ability to have invented it pre hoc (since you can see the ideas in the very scriptures they cite, even if they were the first to “notice” that). So it is fallacious to say they “needed” an actual dead guy to invent one, even if that was a remarkable innovation of theirs.
Self-sacrificing martyrs were an already-established tradition in Judaism and paganism (Element 43 of OHJ, Ch. 5) and had uses at the very time in question (Elements 23–28 of Ch. 5). Likewise dying saviors in mystery religion (none of whom existed either, so clearly no real death was needed to inspire them: Element 31). So the idea was available regardless. It is thus to no avail to insist they could only think of this if forced to by their actual guy getting nicked. For more on this point see Dying-and-Rising Gods: It’s Pagan, Guys. Get Over It.
Yeah, this was always what bugged me about the argument since I encountered it from Craig (and then was shocked to hear secularists accepting such an obviously wildly fallacious argument). As we’ve discussed before on this blog, if Christianity innovated a dying-and-rising messiah, that’s not at all shocking, because they were an innovative, counter-cultural sect. The Craig-type argument depends on the Christians being mainstream Jews who would never depart from any orthodoxy, and that’s so wildly false, and so unrealistic from the perspective of the history of religion in any case (that no one in a culture ever innovates – if not then where the hell do we ever get new religions from), that it baffles the mind to me that it was ever taken seriously.
We would always predict that some weirdos at the fringes of any culture would theologically innovate. Christianity theologically innovated. Whether they did so out of shock over their guy dying or beforehand is irrelevant. Both are so actually plausible when you don’t have bullshit Christian faith assumptions, and both consistent enough with the evidence we have, that they’re reasonable competing hypotheses.
A good example of this is Mormonism. Imagine scholars 2,000 years from now arguing that Jesus must have really visited the American continent since no Christian group before 1800 believed Jesus traveled outside of the Middle East.
Yep. And that’s a great example because the explanation is so clear: it was an adaptation of an old idea to appeal to a new audience, in this case by flattering nationalistic American egos.
Hello Dr. Carrier. You say that “this present subject does not verify mythicism.” I used to think that it was a piece of background information that boosts the prior probability of mythicism by creating a proof of concept for a mythical Christ figure. But now I think I had that wrong since the mythical model of a sacrificed messiah equally explains how the Christians could have so easily overlaid it on an earthly historical figure that was executed, had he existed. I just want to know if I have this reasoning correct. If not, could you comment on how this idea operates within your Bayesian framework? Thanks.
Good question.
It removes a defeater. So what it does is keep the prior where it is, rather than raise it. One can also frame it as be evidence against rather than a prior operator; but that’s functionally the same (since adding evidence updates priors; there is a note to this effect in OHJ).
So the mathematical logic goes like this:
The prior is simply the frequency with which characters like this tend to be historical.
One can then say “but in this case, we are looking at a different reference class, where ‘Jews would never have done this’, so that becomes evidence against any usual trend having happened here,” and so you’d get a likelihood ratio favoring historicity, multiply that by the prior, and you get a new prior, one much higher for historicity (I discuss this concept using a different example in in ch. 6.5 OHJ).
But since ‘Jews would never have done this’ is false, the evidence claimed doesn’t exist. And evidence that doesn’t exist generates no likelihood to change the prior with.
So the result us no effect. The probability of historicity is simply the same as it always would have been.
This is because undoing this evidence does not create evidence against historicity either. Historicity is exactly as compatible with the actual fact (that the idea was already popular enough to adapt to any movement at the time) as mythicism is. So the likelihood ratio for the actual fact is 1/1, or just 1, and anything multiplied by 1 stays the same value.
Hi Richard,
I just watched your video with GE.
I broadly agree with all of your conclusions on this subject, with one exception: the possibility of a messianic reading of Wisdom 2 and 5.
As far as I know, I am not aware of any Jewish messianic text that refers to the “son of God” in Wisdom 2 and 5.
This seems to make it rather difficult to argue that this passage was understood as messianic by certain Jewish sects. But it is possible that I am overlooking a source, so if you know of one, I would obviously be interested in seeing it.
I do, however, completely agree that Wisdom 2 and 5 refer to Isaiah 53. Where I would be more cautious is regarding the nature of that reference.
It seems to me that this is rather a collective reading of the Servant, applied to the righteous, somewhat comparable to the way Daniel 12:2–3 also appears to draw on motifs from Isaiah 53, particularly the vindication of the righteous, the justification of the “many,” and their subsequent exaltation.
There seem to be quite a number of indications of this in Wisdom 3–5:
Wis 3:1: “the souls of the righteous”Wis 3:4: “they were punished”Wis 3:5: “God tested them”Wis 3:6: “he accepted them”Wis 3:7: “they will shine”Wis 3:9: “his holy ones”Wis 5:15: “the righteous live for ever”Wis 5:16: “with his arm he will shield them”In other words, although Wisdom 2–5 clearly draws on the motifs of Isaiah 52–53—the suffering of the righteous, their rejection by the wicked, and their ultimate vindication and exaltation—the subject seems to be the righteous collectively, rather than an individual figure.
This is actually what I find particularly interesting for the question of the Servant: the text shows that a collective interpretation of the motifs of Isaiah 53 existed in Second Temple Judaism, without requiring us to understand it as a messianic prophecy in the strict sense.
I am not sure what your reasoning is. Because we agree with that (and you’ll find I even explicitly note the same point for Isaiah 53 in the article).
Perhaps you are confusing original intent with later reinterpretation again?
Most passages regarded as messianic (and which those were varied sect by sect, exegete by exegete) were never written to be. They “became” messianic in “inspired” readings (which is called pesher—where the holy spirit unlocks “hidden messages” in scripture, thereby producing the whackadoo messianic readings littering 11Q13 for example).
What the literature argues is that Wisd 2/5 is a riff on Isaiah 53. Which entails that as soon as anyone started reading Is 53 as messianic, that would unlock the “secret” meaning of Wisd 2/5 as also messianic, because “now” it’s talking about the same guy—the messiah delineated in Is 52–53 (and thence Dan 9, Zech 3/6/12, etc.). Which I am sure “felt” amazing to realize (that moment of synchronicity and discovery), which feeling would be misread as the Holy Spirit telling them they were right, that they “figured it out,” that their red string board is coming together nicely (so, Hengel, 129-32).
Also, be aware, translations of Wisdom often fudge the plural and singular to suit the interpretive desires of the translator. This goes way back even to the Latin Vulgate. And likely predates it. So you have to sort through that mess first. But the result is as noted.
Hello Richard,
Thanks for your answer.
Unfortunately, I haven’t read your article yet, so I assumed that you were using Wisdom 2 & 5 to argue that some Jews were expecting a suffering messianic figure.
I understand your reasoning, since I agree with you about the concept of pesher and about the fact that Wisdom 2 & 5 allude to Isaiah 53.
Hengel, whom you cite in your response, demonstrates this as well, but then concludes:
“The book of Wisdom therefore presents a universally applicable religious experience, which it supports by manifold allusions to the Scriptures, not least to (Deutero-) Isaiah. But Isaiah 53 loses its unique traits almost completely. Moreover, it should be stressed that the motif of vicarious suffering to overcome the guilt of others is entirely absent. Wisdom deals only with the basically conventional idea of the justification of the righteous and the punishment of the wicked, and even the immanent eschatological expectation essential to Daniel, Qumran, and the Septuagint of Isaiah 53 falls away.”
I therefore think we agree that Wisdom cannot demonstrate that pre-Christian Judaism interpreted Isaiah 53 messianically.
But it is certainly possible that some Jews may have read Wisdom 2 and 5 as referring to the Messiah. That simply cannot be demonstrated from the text itself or from its reception.
One more important question: do you discuss the Apocryphon of Levi (4Q541) and the Self-Glorification Hymn (4Q471b/4Q491) in your article?
In my view, these provide some of the strongest evidence for a concept of a suffering messianic figure in the intertestamental literature, alongside 11Q13 and the Parables of Enoch (which associate the Messiah with Isaiah’s Servant, even though suffering and death are not mentioned there).
Best regards,
I discuss it. But it’s really discussed by the authors I cite and summarize, like Hengel. This is also the case in OHJ.
Right. The logic goes the other way: once we have evidence of their reading Is 53 messianically, we get the reading of Wis 2/5 messianically. Daniel 9 also plays a role in understanding this but indirectly. Dan 9 is explicitly reading Is 53 messianically (possibly even inventing the entire concept of an eschatological messiah, though that would be under the influence of Persian Zoroastrianism whose idea it was from long before). So it establishes proof of concept: if they saw that so as to produce it, then it’s easy for the same kind of thinkers to see it in Wisdom; it basically falls out as an inevitable consequence of the same pesher logic.
Not as such. The article is a survey article, so it is mainly surveying what others discuss. 4Q541 comes up in several of the scholars cited (more even than I mention do in the article), but most prominently (as mentioned in the slides portion of the GE show, which is in the article), George Brooke—Kipp Davis’s dissertation advisor.
Hengel was the first to suggest it though (so far as I know). So the idea has been around since the DSS were published. But Brooke has the most extensive case and he is literally the world’s leading expert on Apoc. Levi. See The Dead Sea Scrolls and the New Testament.
The first scholar to explicitly identify a connection between 4Q541 and Isaiah 53 was Émile Puech, the French Dead Sea Scrolls scholar who was assigned to study the text in 1991.
In his substantial study, “Fragment d’un apocryphe de Levi et le Personnage eschatologique,” Puech argued that 4Q541 provided early evidence for the existence of a suffering messianic figure prior to Christianity.
George Brooke subsequently followed this line of interpretation in 1993.
Now, regarding John J. Collins, I think there is an important point that is often overlooked. I watched your video with GE, and I’m not sure whether you are aware of this. I thought it might be interesting to you.
In the excursus of The Scepter and the Star, published in 1995, Collins rejected the interpretation of 4Q541 as referring to a suffering servant.
He did not identify any connection between the figure in 4Q541 and Isaiah 53. When the second edition of The Scepter and the Star was published in 2010, this excursus remained unchanged.
Nevertheless, Collins’s position appears to have evolved.
In a study published in 2000—only five years after the publication of his excursus—Collins conceded that particular terms in 4Q541 was unusual enough to give some credibility to the view that the author had Isaiah 53 in view.
Moreover, he argued that the eschatological priest in 4Q541 was most likely a messianic figure, namely the Messiah of Aaron
attested in the Dead Sea Scrolls.
That creates a rather interesting situation.
On the one hand, Collins’s 1995 excursus is sometimes cited as evidence against the existence of a suffering-servant tradition at Qumran.
On the other hand, in his later treatment of 4Q541, Collins himself acknowledged that particular terms in the text lends some credibility to an Isaiah 53 connection, while also identifying the eschatological priest as most likely a messianic figure.
Those who reject the idea of a suffering Messiah before Christianity should therefore be cautious about citing Collins as evidence against it. His later work points to a considerably more nuanced position.
Those are all around the same time, but yes. It’s fair to give Puech first credit. Hengel even cites him.
And wow, yes, I checked both editions and knew about how Collins approached that thee. But I was unaware of his different take in 2000. Can you give the full citation? I’d love to check into it and readers here can add it to my bibliography.
Collins’s article dates to 2000 (five years after he wrote his excursus on the suffering servant in The Scepter and the Star).
It was published in a French journal, which may explain why it has somewhat flown under the radar.
The exact reference is:
John J. Collins, “Teacher and Servant,” in Revue d’histoire et de philosophie religieuses, 80th year, no. 1, January–March 2000. Hommage à Marc Philonenko.
Here is the relevant quotation concerning 4Q541:
“There are however some other points that still lead scholars to assume a relationship with Isaiah 53. Chief of these is the word מכאבין, afflictions, which occurs twice in fragment 6 and once in fragment 2. This appears to be a Hebraism. The same word is used in Isa 53:3-4 (איש מכאבות, a man of sorrows, and מכאבינו סבלם, he has borne our afflictions). This term is unusual enough to give some credibility to the opinion that the author had Isaiah 53 in view. The text is too fragmentary, however, to allow a confident conclusion as to how the prophetic text was interpreted. If the figure who makes atonement in fragment 9 is cast in the role of the Servant, as is possible but by no means certain, then the Isaianic passage is read to refer to a priest, who makes atonement not by his sufferings but by fulfilling his priestly functions and enlightening people with his instruction.
Again here there is similarity to the career of the Teacher, who also complains of opposition. The more likely referent, however, is the messiah of Aaron, who is also said to atone for the sin of the people in CD 14:19.”
Thank you. I love super helpful bibliographic comments like that!
Though this reveals that Collins is not here backtracking after all. He is insisting there is no suffering figure in the scrolls, the very dogma proponents extensively refute. He seems here to be just leaning on his other work and assuming he successfully rebutted what in fact he never addressed, and thus just hat tipping the existence of opposition rather than taking a more positive stance on it. Which is typical behavior in this field I am finding out.
It’s already significant that he acknowledged that Isaiah 53 is clearly alluded to in 4Q541.
I think he acknowledged it somewhat reluctantly, since most scholars who have written about this text have recognized the allusion.
He didn’t really have much choice.
I wonder whether his position has changed since 2000. The idea of a suffering Messiah in pre-Christian Judaism has become much more firmly established since then, especially thanks to Martin Hengel’s study, which is still an important reference today.
Richard,
Like most I haven’t been able to read your article yet, but in the chat with GE you mentioned that later (1st century) apocalypticists would necessarily read the Dan 9 prophecy as referring to one single anointed, rather than two.
But, it’s always seemed to me that Daniel 9 (like much prophecy) was purposefully written to be confusing and muddy. Its flexible and difficult to pin down and easy for a later interpreter to play with. So, lets take for granted minimal historicity plus Dan 9.
As such, it seems to me an interpreter like Mark (or Peter, or Jesus, etc) could take it to mean one or two anointed.
So, when I read Daniel 9, whether through the eyes of Jesus or Mark (i.e., this still needs to be fulfilled in the future), I’m thinking the “anointed prince” of 9:25 and 9:26 who is destroying the city is probably the same guy and maybe referring to some Roman official.
Then Mark sees the “anointed one cut off” of 9:26 is Jesus (or, if I’m Jesus, I’m thinking that’s gotta be me lol). Then in 9:27 it picks back up with more of what Jesus needs to do with making a covenant with many. But “the desolator” of 9:27 seems to be referring back to the “prince who is to come” who destroys the city and sanctuary in 9:26.
Thoughts? What exactly do you imagine the earliest Christians were “story-booking” out was going on with Daniel 9 in their own time with Jesus and the Romans? Obviously, they don’t need it to fit perfectly point-by-point, just good enough. But it does fit suspiciously well for their theology.
Thanks!
No, I said the author of 11Q13 must have. Because their reference to it is in a list of the final years of the apocalypse. If they thought it was two, their reference to the first would be several sections earlier (if the scroll even covered earlier sections of the timetable; all we have is a section about the last few years of it).
But yes, Daniel is deliberately dodgy and easily reinterpreted many ways. Like most prophecy was. Indeed that feature basically caused the entire pesher genre 11Q13 was written in.
However, your proposal won’t work, because if we read Daniel 9:25 as meaning a different messiah than the one in 9:26, that entails chronologically a messiah at the seventh seven of years, and the next messiah at the sixty second seven, and whether you overlap or add those periods, either way, it’s a difference of hundreds of years.
This is why, as I noted in the talk, the original intent appears to have been to refer to the bookends of messianic priests: the first of the second temple (in legend named Jesus, who inaugurated the second temple, built by Cyrus the Great in the late 500s BC, and is the one originally meant by the Jesus in Zech 3 and 6) and the last of the second temple (Onias III in the 160s BC being believed at the time to be the last legitimately anointed high priest, and the end of the world predicted to occur seven years later, ending the 70th seven). For background see How We Know Daniel Is a Forgery.
Kipp Davis has been thoroughly dishonest throughout this whole debate.
Not only has he continued misrepresenting your arguments and evidence after being corrected several times, but also he seems not to be able to do discuss anything you do like a grown adult. I really lost respect for him after this, I occasionally listened to him online.
I think the sensible historicist model needs at least some Jews to be ok with a dying messiah, otherwise you would never get Christianity at all. I don’t think one random zealot Who got crucified would be accepted by anyone as a messiah unless his core followers already had that belief.
At first read, I thought it was a typo for toupee (probably why the tuque). 😉
>>The folks who erroneously believed a certain YouTube guy with a goatee and a tuque