Probably. I have not looked into that. But be aware, the Quran appears to have more in common with apocryphal traditions rather than canonical. So it’s not really a question of which “of the four” Gospels because the answer then is zero.

There is a case I believe that it resembles most Syriac Christianity, and likely primitive heretical forms. Indeed Islam may be a survivor of the original Nazorian (Torah-observant) sect of Christians in the general area of northwest Arabia.

As for the rest, I don’t know. Medieval Arabic studies is not my area.

I already just refuted all those points. So, he’s basically just not listening to me. It looks like we’re just never going to get a good takedown of any of the best ahistoricity models. The most we get is that (as I said) the second half of Dr. Little’s article remains the best argument there is against them. But I don’t think it’s enough. I don’t see him taking the alternative models seriously. And until he does, he will never really be able to dispatch them. Their advocates will just always be able to claim their models are being ignored and not taken seriously and not really rebutted. And that will be the story of things. Until someone steps up and really does a comparison model-by-model.

Dr. Little responded with the following:

Dear Islam,

I disagree with a lot of these responses, unsurprisingly!

Regarding Carrier’s latest response, I did not move the goalposts. In Section 3 of my blog, I outlined the inter-regional and trans-sectarian consensus of all Islamic sources and then outlined three distinct arguments on that basis:

A genuine memory of a real founder that was shared by all early Muslims as they spread to many regions and diversified into conflicting factions is a far easier or simpler scenario than some kind of later false creation that arose, after this diffusion and diversification, in one region with one faction and then somehow spread to the other already-established regions and factions across the early Arab empire. It would have been difficult for any given faction to spread a new concept of a historical founder named Muhammad who lived in the Hijaz c. 600 CE and had a wife named Khadijah, etc., to all other regions, let alone to all rival factions. In short, a centralized diffusion of a common memory from the people who met and knew Muhammad, people who then spread out via the conquests and then split into rival tendencies in multiple regions, is more parsimonious.

The founder of a religion is a fundamental belief, not some interesting trivial information. It is extremely difficult to believe that one faction, at a secondary point, convinced rival factions to adopt a new belief of such fundamental importance. The idea of the Umayyads convincing the Kharijites, or vice versa, for example, is ludicrous.

Certain factions would not have accepted basic elements about Muhammad if these elements had been created by their enemies; and they would not have created these elements themselves; therefore, their accepting of these elements can only be reasonably explained by all factions having co-inherited the same basic and public background knowledge about the historical Muhammad.

At no point in this section (Section 3), when appealing to the later consensus, did I simply assume that the sources are independent and then conclude from this that there are independent memories of Muhammad and that Muhammad thus existed. The independence of the sources was not a baseless premise of my argument (pace Carrier), but in fact, the conclusion of the argument. It is the best explanation for the evidence, not a mere assumption. For example, after outlining the appeal to parsimony on my blog, I conclude:

“Already, the simplest explanation for this unanimous agreement—for the shared conviction of countless individuals living in Arabia, Syria, Iraq, and beyond, from Sunni, Shiʿi, and other backgrounds, in the 9th and 10th Centuries CE, that the founder of their religion was a man who lived in Western Arabia in the early 7th Century CE—is that Muhammad was indeed a man who lived in Western Arabia in the early 7th Century CE. In other words, the simplest explanation for this unanimous agreement is a broadly accurate collective memory: each early Muslim community independently inherited the same basic memory of the founder of their religion. Naturally, each community remembered things differently, and reports that arose in one community certainly spread to others; but overall, they shared, through common inheritance, the same basic memory of a historical Muhammad. Other explanations for this universal consensus are of course possible, but would require a more complicated process, involving not just the full-scale creation of Muhammad’s basic biography, but also, its spread across the Arab Empire and its universal acceptance amongst all of the Muslims therein. Thus, all else being equal, the simplest hypothesis—that Muhammad existed and was broadly remembered by his diverse and dispersed followers more than a century and a half later—is to be preferred for the data in question.”

This is the argument to which Carrier was responding (i.e., “His ‘9th century consensus’ argument”), and which Carrier has inadvertently misdescribed twice:

> “His “9th century consensus” argument is another faulty deployment of the multiple attestation argument. Sources a century late cannot be assumed independent of a singular inventor in the 8th century, so his argument there fails.”

And:

> “I pointed out that another, completely different argument fails: that multiple sources a century later allows the assumption they are independent. It does not. Instead, you need evidence of independence, not just their existence (e.g. all the variant versions of the Labors of Hercules, all the variant tales of the Roswell Saucer, etc.).”

Nowhere did I deploy a “multiple attestation argument” along the lines described by Carrier. Instead, I appealed to parsimony, then the difficulty of inter-regional and trans-sectarian spread of fundamental origins narratives, then embarrassment, to argue for the conclusion that what we have in this consensus is most likely a reflection of independently inherited common memories. In short, I provided considerations in favour of independence, just as Carrier asks.

Regarding this comment:

> “Little’s response is that he has a completely different argument, about theory complexity. That’s not answering my point. I was not calling out that argument but the other one. The other one remains fallacious exactly as I noted, and that knocks one pillar out of his case. His case still stands without it. But it can’t stand on it.”

There was no “other” argument, i.e., there was no multiple attestation argument of the type described by Carrier. The argument that Carrier was referring to at the outset (“His ‘9th century consensus’ argument”) was never an appeal to multiple attestation. It was an appeal to parsimony, followed by appeals to other considerations, resulting in the conclusion of independent regional and sectarian co-memories.

As for Carrier’s response to the appeal to parsimony, see above. I explained in a bit more detail there the way in which a common early memory is easier given the circumstances of early Islamic history.

As for the Syriac point, yes, I am familiar with the relevant literature. There is no doubt that the Quran draws heavily upon and reworks a bevy of Late Antique Christian and Jewish texts and traditions. Dhu al-Qarnayn is the obvious example, but there are many others; Reynolds, The Qurʾān and Its Biblical Subtext, is particularly illuminating. However, all of this is complicated by Marijn van Putten’s recent observations that the Quran’s so-called Syriac vocabulary is actually drawn from older forms of Aramaic, not directly from Syriac (see here); and by Nicolai Sinai’s observation (“Meccan Elephant”, p. 67) that the Quran lacks technical Christian jargon, which seems odd given that most of the Quran’s putative Syriac sources are Christian. All of this complicates the picture – hence, my cautious formulation and hedged language. Again, there is no doubt that the Quran is heavily influenced by numerous Late Antique Christian and Jewish texts and traditions, but “the Syriac Christian origin of the contents of the Quran” (as Carrier put it) requires some caveats. However, in my initial response, I set all of that aside, for the sake of argument, because even if the Quran is just a rehash of Syriac Christian material, it has no real bearing on the historicity vs. mythicism debate. It is compatible with either view.

I’m running short of time now, so here are some brief comments on some of the other points.

Carrier reiterates his claim that I argued “possibly, therefore probably”, but again, this was already dealt with. I outlined specific reasons to accept the basic historicist thesis over the basic mythicist thesis: parsimony, the difficulty of convincing rival sects of fundamental changes to origins narrative, and embarrassment. It is fine if Carrier rejects these, but he cannot say that I argued “possibly, therefore probably”.

Carrier’s appeals to early Christianity, Cargo Cults, Ned Ludd, etc., don’t really address my Section 3 arguments. Yes, it is possible to imagine a scenario whereby Muhammad was made up by one faction and then spread to others at some secondary point. However, the division of early Muslims into multiple regions (Arabia, Egypt, Syria, Iraq, Iran, North Africa, etc.) and conflicting factions and tendencies (Umayyads, Zubayrids, proto-Shi’ah, Kharijites, etc.) already within (putative!) living memory of Muhammad would have made this difficult; and it would have been even more difficult to convince hostile factions (e.g., the Umayyads convincing the Kharijites or vice versa); and it would have been even more difficult to convince factions of inexpedient facts (like convincing the Umayyads that Muhammad was a Hashimid and that the early Umayyads had persecuted him). I don’t see how appealing to the situation with early Christianity, Cargo Cults, Ned Ludd, etc., actually gets around all of this. I don’t see how those situations are actually comparable with the situation and considerations that I outlined for early Islam.

This is my last response like this, I’m afraid. It’s eating up way too much time.
(—– Islam editing here in the parenthesis: The next sentence is directed to me. Dr. Little expressed interest in a private conversation with you Dr. Carrier —–)
You’ll have to take it from here; otherwise, post on the AcademicQuran subreddit; they’ll be helpful with this kind of thing.

Please give my best to Carrier!

Kind regards,

– Joshua

There are issues here that reflect my continuing critique of historians for not attending to formal logic (I am very much with Fischer on this now), so do send this link to Dr. Little so he can see what I mean:

I agree that a late consensus could be wrong, but my point was that assuming a founder named Muhammad in the early 7th Century CE is simpler or easier than…

Note that that is a different argument.

This is a fallacy called Moving the Goalposts.

I pointed out that another, completely different argument fails: that multiple sources a century later allows the assumption they are independent. It does not. Instead, you need evidence of independence, not just their existence (e.g. all the variant versions of the Labors of Hercules, all the variant tales of the Roswell Saucer, etc.).

Little’s response is that he has a completely different argument, about theory complexity. That’s not answering my point. I was not calling out that argument but the other one. The other one remains fallacious exactly as I noted, and that knocks one pillar out of his case. His case still stands without it. But it can’t stand on it.

Now, if we want to play the game of moving the goal posts, we have to move to that other goal post: this claim of simplicity.

That has not actually been established. The problem with any historicist for any figure claiming “simplicity” is that that claim is often false. Tons of ad hoc assumptions are needed to get any hypothesis to explain all the evidence (Little’s essay is full of just-so stories about how things could have happened, which is not equivalent to evidence that they did happen). Comparative simplicity thus requires an honest comparison of the actual relative simplicity of the competing hypotheses. Which requires describing the best competing hypothesis.

Little does not effect any simplicity or evidential comparison with any specific alternative theory, least of all the best competing theory. That remains exactly the problem I said it was.

For example:

One source causing multiple sources is simpler than multiple independent sources. One guy invented the Labors of Hercules and a hundred later guys thus inspired composed their own redactions of it within two average lifetimes is far simpler than “there were a hundred different witnesses who all wrote stuff down that we now have conveniently lost even any mention of and this explains the hundred later legendary sources we have.”

That does not mean the latter theory is false or even less probable. What it means is that you can’t defend it with an argument from simplicity. The competing theory is at least as simple if not simpler.

Now, maybe we can tweak or build out things to get an argument from simplicity to work here. But that has to actually be done. Not having done it is a valid critique of the current state of play. Which was indeed my original point: you can’t just assume things. You have to prove them.

Because if anyone’s historicity is being defended on assumptions and not proofs, then you are handing the win to the doubters, not rescuing historicity from them. Historians need to take this point more seriously.

/// … there is no engagement with the Syriac Christian origin … ///

I don’t really understand this line of argumentation.

It sounds like Little has not read the theories and thus doesn’t know what the competing theory is that he needs to be addressing (I reference some in my article). That only proves the point I was making.

Let’s suppose that the Quran actually is just a rehash of Christian Syriac texts and traditions.

It’s not a supposition. Theorists have built elaborate linguistic and literary arguments with evidence for this conclusion. They might be shit (as I wrote in my own article, I have no skills to vet this so I have no idea if it carries or works). But you have to actually prove that. Claiming you don’t even know the arguments exist immediately loses the argument. To the contrary, you have to refute what these opponents are saying (their elaborate linguistic and literary arguments, their specific models for the development of the legend over time, and their stated reasons for its success—in other words, their actual arguments).

Little needs to do this. Not handwave it. That’s what I said. And he is only confirming what I said here.

I don’t see where I engage in this kind of reasoning in Section 3 (or anywhere for that matter).

Note I said immediately after that “the second half of Little’s article is much stronger.”

So I was not claiming he only had possibiliter fallacies. Rather, I said he had some (every time he builds out a just-so story to explain the evidence, he often gives us no evidence that story is true) and then he mistakes those as arguments for historicity, when they aren’t.

Both sides agree we can come up with stories to explain the evidence. The question is which story explains it better. And that can only be answered with evidence, not more stories.

But my point here was that when we are at that impasse (some item of evidence is equally explicable on both theories), only the priors can adjudicate. And that has to come from external not internal evidence. For example, is there anything about the Quran or the biographies of Mohammed that makes them look more or less like faked documents for driving a movement? That requires looking at other (non-Islamic) literature and seeing what markers indicate a text or figure to be more or less often real vs. fake. As with Jesus: he is heroized in such a way that he resembles classes of people who typically did not exist, rather than classes of people who typically did. That required looking at other people, like and unlike Jesus, and other literatures, like and unlike Christian.

So my point here is that “we can explain that” is not evidence for your theory. It just removes an argument against your theory. Which is not the same thing.

And in the first half of Little’s case there is a lot of this confusion (less so in his second half).

The question that is left that needs to be resolved is the priors. How typical would the Quran and a Mohammed biography being fake be? And that can only be answered by looking at analogous literatures or movements (not at Islam itself).

I don’t know how this would go. As I said, I don’t know how Mohammed mythicism would fare when that side of the equation is empirically completed. I am saying only that that needs to be done—if it can (and if it can’t, then the Principle of Indifference leaves the priors at 50/50 and we’re back to needing evidence for one side’s story over the other side’s story).

This was essentially already addressed in the article.

It is not. I don’t think Little understands the thesis and case these theorists have brought out. Does the Quran look a lot like Syriac literature, complete with linguistic genealogies and so on? Little never discusses this. It needs to be refuted before it can be dismissed. Likewise, do the Mohammed biographies look like aetiological myths, or actual memoirs? What was the actual timeline? When exactly was anything being claimed about Mohammed, beyond an author byline which could simply be pseudonymous? And how do we know that? And so on. There are many questions yet to be answered before the historicity of Mohammed can be secured against the doubters. That’s my point.

The Muslim community fractured into rival tendencies in different regions at the end of ‘Uthman’s reign

So did nearly every myth-based religion in history (look at all the rival factions that broke away from the Mormon’s central church; look at the mass of heresies that broke out in Christianity in just a hundred years, all claiming fake pedigrees to eyewitnesses). This therefore is no more or less likely on either model. It therefore can’t evince anything here. It’s useless data.

Indeed, in the Noll thesis (explaining the Cargo Cult and Luddite and other movements) is that inventing historicity for Mohammed would be precisely how factions would try to establish themselves over each other. The tendency is to invent a history and then control it, to win control over an institution or movement.

So, for example, Mohammed could have begun as just a made-up byline, with a mere brief report that he wrote a revelation down (like with the Book of Daniel or the Gospel of Peter), authorities then marketed this through standard power-and-influence levers to sell their clients and soldiers and subjects on it; then decades later, as people wanted to change doctrines or get their interpretations to win out, they invented ever more elaborate stories about this man, resulting in the detritus of materials that survive. And so a made-up man becomes a “real” man complete with biographies and within mere decades.

We know they would steal from and redact each other (see all the divergent legends about Roswell now). This is just like the Gospels: Matthew stole most of Mark, claimed its content as his own, and fixed it up to argue the opposite position; Luke stole both and fixed it up to take over control of both communities; then pseudo-Peter stole that and fixed it up to push new doctrinal views in the Gospel of Peter, and so on. Thus explaining their shared content.

This is so typically what happens that one cannot claim it is unlikely. It is in fact the most likely thing and must have happened even if Mohammed existed—because all the fake stuff about him came from somewhere and it wasn’t him. The only difference between historicism and mythicism is that the latter say more things were made up. They otherwise agree a ton was made up. So “they can’t have made it up” is not a good argument here.

it would have been very difficult for one faction to convince the rest that, actually, within living memory, just a few decades ago, there was a founder named Muhammad who nobody remembered; let alone a founder who was the member of a specific family that other factions hated, etc.

This is precisely what I have disproved with other examples, which show no such difficulty existed. It’s the other way around: it’s super easy to do this, wholly regardless of what the truth was. Truth (and rational evidence-based reasoning) have next to nothing to do with how religions fracture and evolve.

Rapid Legendary Development resulting in absolute immovable faith in competing factions takes only a few years, much less decades (Paul’s letters document this happening to Christianity in real time). And in fact it typically results in faking up histories. There is no religion on Earth that isn’t built on made-up heroes and made-up narratives.

But the question is:

Do we have evidence that, say, someone was arguing Mohammed was a real person (rather than just assuming it or saying he was just a symbol or metaphor) within 10 years of the Quran?

And do we even have evidence that was happening within 40 years?

As in, do we have documents from that time saying that? Or are we just theorizing or supposing that, based on later legends?

And if we have evidence and not just suppositions, can we move that back to 10 years, or no further than 40 years?

Because at 40 years we already know that’s the typical timeline for invented people and things to become historicized (Ned Ludd, John Frum, Roswell) as by then no believer has the means to question or disprove it or often even reason to, while the unbelievers and apostates we don’t get to hear from because no one preserved their options on the matter.

So you can’t “just so” a story about there being specific things preached at some specific time that would be a hard sell at that time, unless you actually have evidence that was actually happening (and that’s evidence, not supposition), and that the precedents in other religions or movements does show that it would be a hard sell (and not, as we see typically at the 40-year mark, not a hard sell at all).

You need both of those things.

Do we have both of those things?

—

Now, all that said, I’ll reiterate: Dr. Little’s second half leaves a much stronger case against doubt than the first half. So I’m not at all siding with the doubters here. I’m skeptical of their case. I just need them to be more decisively refuted before I can be fully confident in dismissing them. Hence the above critique relates only to what there is yet still to do here.

Dear Dr. Little,

I hope this email finds you well.

I came across your excellent article about the existence of Muhammad. I then discussed it with Dr. Richard Carrier on an article of his that was dealing with the same question 10 years ago:
https://www.richardcarrier.info/archives/8574

He has some feedback about the 9th century consensus argument and about addressing the Syriac origin hypothesis that you can find in our exchange in this comment thread:
https://www.richardcarrier.info/archives/8574#comment-41036

Dr. Carrier (and I) thinks that you are doing great work and would be happy to discuss these points privately with you in case you wish to. These are the ways to contact him if you are interested:
https://www.richardcarrier.info/archives/14439

Yours sincerely,
Islam Hassan

To which he replied:

===================

Dear Islam,

Thank you kindly for your email.

I am familiar with Richard’s work and have cited it at points, although you won’t be surprised to learn that I disagree with him on key points like this!

These are my immediate thoughts on some of his comments:

“His “9th century consensus” argument is another faulty deployment of the multiple attestation argument. Sources a century late cannot be assumed independent of a singular inventor in the 8th century, so his argument there fails. They could all agree because they are all trusting the same original fabricating source, which simply diffused over three continents in 100 years just as Christianity did.

Case in point: everyone in 200 AD across three continents and languages and dozens of divergent sects agreed Jesus flew into space before eyewitnesses. But that was made-up 80 years before; and that was 80 years after the religion even began. It’s also obviously ahistorical. So “late consensus” arguments cannot evince historicity. Those three continents were simply awash with the Gospel of Luke (and references thereto and legends therefrom) by then.”

I agree that a late consensus could be wrong, but my point was that assuming a founder named Muhammad in the early 7th Century CE is simpler or easier than assuming some kind of secondary spreading of a false narrative amongst all of the various geographically diffuse and religiously divided Arab/Muslim factions at some later point. I don’t think that this is “faulty”, although I agree that it could be overcome by stronger forms of evidence, i.e., this is a relatively weak consideration.

“Another defect I see is there is no engagement with the Syriac Christian origin of the contents of the Qumran, which is actually the most compelling argument against historicity. And yet that’s the argument that requires the most professional attention from someone expert in Medieval semitic languages and history.”

I don’t really understand this line of argumentation. Let’s suppose that the Quran actually is just a rehash of Christian Syriac texts and traditions. What would follow from that? This is compatible with any number of hypotheses: we could posit that such texts and traditions filtered down into the Hijaz prior to Muhammad and he draw upon them there; or that Muhammad acquired such material in Syria during his visits there (reported by both Muslim and Christian sources); or that a later Muslim ruler replaced Muhammad’s true teachings with Christian material; or that Muhammad never existed and a later Muslim ruler cobbled together such material to create a holy book that was then attributed to the mythical Muhammad; etc. In short, the Quran’s containing Syriac Christian material seems completely equivocal to me, i.e., irrelevant to the historicity vs. mythical debate.

“Little also falls into the trap of assuming “we can explain the state of evidence” with “therefore our explanation is correct.” This “possibly, therefore probably” fallacy is standard in Jesus historicity debates and is not logically sound.

That comes from mistaking an argument that P(e|h) = P(e|~h) as an argument that P(e|h) > P(e|~h), justly refuting the latter, and then acting like you’ve refuted the former. But the former still removes evidence for historicity. Which reduces P(h) to its prior probability. And that has to derive from comparable cases (comparands other than Islam), not the internal evidence of Islam. I don’t know how that would go (as I explain in my own article, I lack the skills needed to develop that prior here). But the procedure cannot be ignored.”

I don’t see where I engage in this kind of reasoning in Section 3 (or anywhere for that matter). On the contrary, in Section 3, I variously appeal, in light of the specific conditions of early Islamic history, to (1) parsimony, (2) the difficulty of one faction convincing others of a such a fundamental false narrative, and (3) the criterion of embarrassment, all of which favor the historicist explanation over the mythicist explanation. Thereafter, I note that the alleged silence of the earliest sources on Muhammad (which I accept for the sake of argument, despite the fact that he is mentioned in early sources) is equivocal and easily explained on the historicist view; then I reiterate that any such historicist interpretation should be adopted because it coheres with the other points just mentioned. None of this is “possibly, therefore probably”.

Regarding the last point:

“This is where I suspect a better model can be developed out of the Syriac theory and thus why that needs more attention than it is getting.
For example, suppose that the Quran was published exactly in the year believed, but by, say, Uthman (or some Muslim leader of that same generation), and it claimed on its title page a fake name meant to market it (Muhammad, “Praised One,” kind of like Theophilus, “Lover of God” as the fake patron named in Luke-Acts; or the fake Gospel of Peter being published as if written by Peter, or the fake Letter to Abgar published in the name of Jesus), and consisted of just an edited-up collection of prior Syriac Christian works or essays or ideas. And then within forty years a fake biography was written that became the standard go-to for every subsequent sect (just as happened to Christainity).

How would we disprove that theory?”

This was essentially already addressed in the article. The Muslim community fractured into rival tendencies in different regions at the end of ‘Uthman’s reign, so it would have been very difficult for one faction to convince the rest that, actually, within living memory, just a few decades ago, there was a founder named Muhammad who nobody remembered; let alone a founder who was the member of a specific family that other factions hated, etc.

I am happy to get in touch with Richard – he and I have a mutual friend, so I will do so via that avenue.

Kind regards,
– Joshua

Excellent. And good fortune!

I have a qualification exam for a 9 month graduate program that’s due in 10 days. Once I am done with it (hopefully successfully), I will definitely reach out to Dr. Little with your feedback framing it exactly as you suggested.

Thanks a lot for all your work and for your patience and generousity in answering my numerous questions.

Indeed, I am very much overwhelmed with a lot of behind-deadline work. It would be a welcome kindness if you carried that water for me!

I fully endorse your reaching out to him about it.

Show him this article. And our comments thread (here is a hyperlink for the latter and of course for just the article it’s this). Ask him a brief respectful question about it. And let him know he is welcome to reach out to me if he would like. Even if I am slow on comms I get to them eventually (here is A Guide to My Social Media if he is interested in choosing the best venue for getting in touch).

He’s definitely doing good work and has the skillset for this. So he might get another interesting article out of it!

Thanks a lot for your detailed comments. Very educational as always.

I agree that the argument from the 9th century consensus is very weak.

I unfortunately lack any knowledge of Syriac to be able to do my personal layman’s assessment.

I know you are very busy, but I think it would be great if you can reach out to Dr. Little with your feedback as he looks like someone who is willing to engage in real scholarship about the topic which is rare as in my experience.