A few years ago Strange Notions published a strange editorial by statistics professor William Briggs, called Bayes Theorem Proves Jesus Existed (And That He Didn’t). I say strange, because it’s weirdly dishonest, incompetent, and irrational coming from someone who teaches probability theory for a living, and has a whole Ph.D. in the subject. It’s especially perplexing because it seems to exhibit a nightmarish fear of probability theory. “Our beliefs cannot be subject to valid reasoning about probability! That cannot be allowed!” seems to be the subtext here. And that’s weird. Yet it seems to be a common Christian failing.

Being able to catch and dissect things like this is a skill I’ll be teaching tools and tricks for in a course online, starting tomorrow, so register now!

Why Are Christians Suckers for Fallacies?

It’s doubly weird that Briggs’ argument for this conclusion is that probability theory can be used to prove anything. Unless Briggs is a postmodernist who thinks all conclusions in probability and statistics are false, surely he cannot believe that’s a logically sound argument. He of all people surely knows that the only way “statistics can be used to prove anything” is if someone is lying (or making an identifiable mistake). Obviously, statistical arguments to a conclusion are only to be accepted when they are valid and sound, and they are only sound when the premises (the facts and probabilities) are factually true. And when they are, statistics cannot prove “anything.” In fact it can only ever prove one and the same thing. Same premises and same logic, always entails the same conclusion. Not multiple contradictory conclusions. Only logically invalid methods would do that. Other than discrepancies caused by subjective differences in observers (on which see Proving History, pp. 88-93).

Briggs thus commits the fallacy of affirming the consequent:

• If P, then Q.
• Q.
• Therefore, P.

Ergo:

• If Bayesian methods in history are logically invalid, then Bayesian methods in history will get contradictory results.
• Bayesian methods in history have gotten contradictory results.
• Therefore, “Bayesian methods in history are logically invalid.”

Nope.

That’s a fallacy. The conclusion does not follow. Because there is another, better explanation for why contradictory results are observed: and it’s not because Bayes’ Theorem is not a logically valid model of historical reasoning. It’s because someone is putting bogus data into the model. And that also does not mean all of them are: one of them might be doing it honestly. So it is the responsibility of anyone worried about this, to check. Why a statistics professor would be so desperate and confused he’d need to deny the applicability of Bayes’ Theorem is astonishing. Why he would do so without checking any relevant facts, is even more astonishing. Why he would lie and falsely claim he did is all the more astonishing. All three behaviors demonstrate how desperate and delusional the victims of Christian faith become.

Instead of using the fallacy of affirming the consequent, Briggs needs to show that Bayesian methods in history are actually invalid—which requires showing that the observation of conflicting results is not due to dishonesty or error in developing the inputs. Yet he never does this. He doesn’t even seem to know what Bayesian arguments about historical facts look like, or why or how they work. Much less that they don’t work because they are invalid—rather than because they are sometimes unsound, because some people put false premises into them.

I’ve discussed this whole point before, when showing the total sham of Swinburne and Unwin in their bogus and dishonest deployment of Bayesian reasoning in defense of Christian faith. See my article on Crank Bayesians. Just as Christians dishonestly or incompetently (or delusionally) abuse logic to the same end. Yet that does not mean logic is invalid. It means Christians are not competently or honestly employing it. Otherwise we’d have to conclude they are so consistently making mistakes in it as to defy explanation. Like Briggs just did: relying contentedly on a fallacy of affirming the consequent, oblivious to the irrational mistake he just made.

Why Are Christians Liars?

It’s bad enough that Briggs’ entire piece is a fallacy of Affirming the Consequent and he, a professor of mathematics, doesn’t even notice. What’s worse that it’s also a big bag of lies.

The title says Briggs is talking about examples of Bayes’ Theorem being used to prove “Jesus Existed (And That He Didn’t).” But he gives no example anywhere in his piece of Bayes’ Theorem ever being used to prove Jesus existed! Lie number one.

In fact, Briggs gives no actual example of Bayes’ Theorem being used to prove Jesus didn’t exist, either. He cites only my book Proving History. In which I never argue any conclusion about the historicity of Jesus. Much less mathematically. Lie number two.

Note this means no book he mentions in the piece, argues either thesis his title references. So his whole premise, of contradictory results, on which Briggs bases his fallacy of affirming the consequent, is not even established by anything Briggs presents in his essay! This is crazy town.

On the “for” side, Briggs only mentions Unwin’s book The Probability of God. Which never addresses the historicity of Jesus. At all. Much less mathematically. It simply presumes Jesus existed, and then muses about the resurrection. But it is all assumption, not demonstration. And no math. Unwin doesn’t even apply any math to the resurrection. Or that Jesus was the Son of God. He only calculates a probability for the existence of a god. Not even the Christian God specifically. Yet Briggs claims Unwin “uses Bayes’s theorem to demonstrate, with probability one minus epsilon, that the Christian God exists.” Lie number three.

Moreover, Unwin’s conclusion is that the probability of God’s existence based on his examination of the evidence is only 67%. Yet Briggs claims Unwin got the result of “probability one minus epsilon,” epsilon being a mathematician’s term for a very small number (in fact, usually infinitesimally small). In other words, Briggs lied. He said Unwin found the probability to be arbitrarily close to 100%. In fact, Unwin found it was far more ambiguously around 67%. Strange lie for Briggs to tell. But alas. Lie number four.

Briggs then says Proving History “uses Bayes’s theorem to prove, with probability one minus epsilon, that the Christian God does not exist because Jesus himself never did.” No such argument appears anywhere in Proving History. As I already noted, that’s lie number one. But even my book On the Historicity of Jesus, which was also out at the time (published in 2014), the one that does argue a low-ish probability of Jesus existing, did not get a result of “probability one minus epsilon.” It got a result of “between 1 in 12,500 and 1 in 3” (p. 601). Which is nowhere near any epsilon. Lie number five.

In fact the upper bound there is roughly 33%, which is my actual result—since the point of designating a margin of error is that we don’t know where the actual probability lies within the boundaries of that margin. Therefore it may lie at 1 in 3. Which entails only a roughly 67% chance Jesus didn’t exist. Nowhere near “one minus epsilon.” To the contrary, 33% is a respectably high probability Jesus existed. So where on earth did Briggs get one minus epsilon? His dishonest brain. That’s where.

And yet Briggs claims that Proving History “attacks” the probability that Jesus existed “with great gusto.” In fact, it never attacks it at all. It’s never discussed in that book. Yet somehow Briggs wants you to believe it’s not only dealt with there, but “with great gusto,” clearly implying he read the book and knows this is true. When he knew for a fact he knew no such thing. And it’s false. Lie number six.

Worse, Briggs says Proving History “uses Bayes’s theorem to prove, with probability one minus epsilon, that the Christian God does not exist because Jesus himself never did.” WTF? At no point anywhere in Proving History, nor anywhere in On the Historicity of Jesus, nor in fact anywhere ever in the entire history of everything I have ever written, do I argue “that the Christian God does not exist because Jesus himself never did.” That’s slanderously absurd. Lie number seven.

“So here we have,” Briggs then says, after telling seven shameless lies, “probability proving two diametrically opposite conclusions.” Nope. None of my actual mathematical conclusions in PH or OHJ align at all to any conclusion in Unwin, nor any mathematical conclusion in Unwin to any conclusion in PH or OHJ. So no contradiction can even in principle exist between them.

I have of course made probabilistic arguments against the existence of God (e.g. in The End of Christianity). But Briggs seems to have no knowledge of that. And in any case, no such arguments appear in PH or OHJ. And insofar as they contradict Unwin’s conclusions, it’s because I don’t cheat: I put back in all the evidence Unwin leaves out, completely altering all the input probabilities, and thus completely reversing Unwin’s conclusion. So our results differ not because the math is invalid. Our results differ because Unwin is a conman who rigged the inputs by hiding evidence; and I alone used the math correctly by putting all the evidence in. If Briggs would stop being hung up on fallacious consequent affirming, he’d know he needed to check if that was the case—if one of us was hiding evidence and thus rigging the results, and which one of us that is. But Briggs is evidently terrified of checking facts. He’s more comfortable living in a bubble of indignant fallacies and lies.

Briggs doesn’t stop lying there. He says I’m “convinced Jesus was a first-century creation, invented whole cloth, likely born of a conspiracy to create a new religion.” Total lie. I have not only never argued that, I have explicitly argued that I reject that explanation. I reject actual conspiracy theories and explain how none are needed to explain the evidence in OHJ, pp. 276, 291, 303, 305, 609. I do allow for the possibility that when the religion was “created,” it was based on a fabricated celestial (but not yet earthly) Jesus, but I explain in detail why no such assumption is required (see Element 15, Ch. 4, OHJ, esp. p. 131 with note 176).

And that isn’t how the Gospel Jesus was invented. The religion had already been created by then, in fact by then it had been around an average human lifetime, preached across three continents. And far from a conspiracy, the fabricators of the many historical Jesuses that ended up being invented and paraded by the faithful (in some forty Gospels in all) were all trying to battle and defeat each other in the arena of ideas. Dissenters ultimately were erased from history and their documents destroyed or coopted, even doctored (and this is not supposition but a demonstrated fact in mainstream scholarship). But that, too, required no conspiracy. Just a ruling sect whose individual members all acted independently of each other with the same goals and values. Which we know for a fact happened. So it requires no mere theory.

So what Briggs says about my thesis is another falsehood, aimed apparently to disparage it. Lie number eight.

Briggs then says he “won’t dig into the details of Carrier’s points,” but it’s clear Briggs doesn’t even know what those points are—he doesn’t even know where they are. He falsely believes they are in Proving History! So he again falsely represents himself as having surveyed and assessed all my reasons for doubting historicity. Lie number nine.

Then Briggs says this:

On the other hand, an early review of Unwin’s work, which I have read and which is mercifully brief (and in large font with small pages), asks just the right question: “Can you imagine anyone arguing that the existence of evil in the world, given that God exists, is 23% as opposed to 24%, for instance?” Indeed. Too bad this kind of question is not asked in science.

Those are the words of an Amazon customer Kevin Iga. Who is making a relevant point about Unwin’s failure to represent margins of error in his calculations. Which is indeed a sign of Unwin not doing an honest job of this. Note I do provide margins of error, thus meeting Iga’s objection. What’s weird here is that Briggs isn’t reading Unwin’s book (he falsely claims he did, but as he clearly doesn’t know what’s in Unwin’s book, that’s, as we already noted, a lie), he’s just reading cherry-picked Amazon customer critiques. WTF? But what’s wrong here is that after doing this, Briggs fails to note (because he is lying about having read our books) that in fact my treatment of the methodology does not meet with this objection. That in fact I did correctly, what Iga points out Unwin screwed up. And this is so directly relevant to Briggs’ entire critique it is damning. Because I have a whole page in Proving History dealing with this, and then a whole section explaining what it takes to get a valid and sound analysis by avoiding exactly this kind of mistake (pp. 66-67 & 85-88).

After all this, all his fallacies and lies, Briggs concludes: “focusing on probability is wrong.” Huh? Why is focusing on probability correctly, wrong? And if we aren’t to discuss probabilities, what then? Briggs seems to have no coherent grasp of how he can say it’s likely Jesus existed, if he is forbidden to discuss probabilities. Pro tip: “likely” means a probability. Nor does Briggs seem to have any coherent grasp of how he can say it’s likely Jesus existed, if he can present no valid argument for that conclusion. And such an argument must argue to a probability. So what form of argument does he think he can use? Try as he might, he will always end up finding out there is only one: Bayes’ Theorem.

Briggs’ concluding line that “These authors would help themselves better, and contribute to a more fruitful discussion about Jesus, by explicating the evidence and eschewing unnecessary quantification” is a brilliant combination of his fallacies and lies, capturing the entire mode of his argument throughout. It’s fallacious to say we can make claims about what’s probable without some kind of quantification. Probability is by definition quantification. And he never shows any of our quantifications are “unnecessary.” Of course he couldn’t in my case, as there are none for historicity in Proving History. He lied about what that book contains and argues. But more importantly, the book that does discuss that question, On the Historicity of Jesus, is 600+ pages full of “explicating the evidence.” For Briggs to claim otherwise? That I didn’t do this? Well. That’s lie number ten.

His terror of probability theory comes clear when Briggs says this:

The real question is this: how can probability prove a thing and its opposite simultaneously? The answer is simple: the same way logic can prove a thing and its opposite. This does not prove that logic should be lumped with pseudoscience, however. You can’t blame the tool for its misuse.

Here Briggs almost gets it right, that the problem is not the method as he’d been claiming throughout this whole essay. It’s the abuse of the method. But he never actually shows any examples of such abuse—not even in Unwin, despite so many juicy examples to pull from there. Instead he says we should abandon the method. That’s bizarre. Notice how bizarre it is. He just here said “logic can prove a thing and its opposite” but only because of misuse, so you shouldn’t reject logic; yet he then says we should abandon probability reasoning when discussing the probability of things being true: “focusing on probability is wrong,” he says. Analogously, he would have to say the same, then, of logic: “focusing on logic is wrong.” Likewise he says we should ‘eschew quantification’. Analogously, he would have to say the same of logic: that whenever we want to determine what’s true, we should “eschew logical argument.” The inconsistency is striking.

Just as Briggs, if he were honest and coherent, would admit that abuses of logic do not mean we should abandon logic—to the contrary, the very possibility of abuse means we should take even greater care to make sure our conclusions adhere to logic—so he would also admit that abuses of Bayes’ Theorem do not mean we should abandon Bayes’ Theorem. To the contrary, the very possibility of abuse means we should take even greater care to make sure our conclusions adhere to Bayes’ Theorem.

But Briggs is neither honest nor coherent. He is inexplicably terrified of ever asking what he really means when he says the existence of Jesus is probable. As if the moment he starts admitting that he cannot honestly mean 100%, that that will reveal he could be wrong, and that he doesn’t actually know how to justify the probability he does mean. The house of cards will crumble. “Don’t dare quantify any probabilities!” As then we might discover they have no basis. And thus neither do our corresponding beliefs. Scary indeed. No wonder Christians flee from the very idea. And no wonder Briggs is content to thus flee on a boatload of fallacies and lies. He must surely have convinced himself Jesus will forgive him for being an irrational liar. “The Ten Commandments? Fuck those. I need to prop Jesus up on a bed of lies!”

No wonder, then, that the only article of mine Briggs links to in which I even discuss the historicity of Jesus (and yet which still does not argue for any probability of it) is my exposure of the lies and errors in Bart Ehrman’s Huffington Post article (which Briggs bizarrely calls “a minor blog post” by some random “historian”…Briggs just can’t be honest about anything, apparently). Yet he never explains what in that critique is incorrect. Or how it helps his case to cite an exposé of shoddy arguments for historicity.

Taking probability theory seriously, entails exposing assumptions to the light of day, that once exposed, destroy the Christian faith. The resulting cognitive dissonance is so powerful only two options are available to the believer: make shit up (like Unwin and Swinburne, they fabricate fantastical probabilities that have no plausible basis in logic or reality) or declare probability itself the enemy. Briggs picks option B. Meanwhile, all peer reviewed work on the question finds the opposite: that history is in fact Bayesian.

Conclusion

William Briggs is a liar. He told ten glaring lies in one short essay. And rested his entire case on those lies. And yet he is so incompetent, that already with ten whole lies to work with, he still couldn’t get his conclusion by any logically valid route, but had to rely on a fallacy as well! A competent conman could at least develop a valid argument using false premises. Nor would they need ten lies to do it. All this would be laughable if it wasn’t so disturbingly sad. This is what Christian apologetics inevitably becomes.

It is almost never the case that any critic of my work, actually reads my work. They just lie about what my books say instead. That tells you pretty much all you need to know about the legitimacy of their position. They are terrified of what’s in my books. They are terrified that anyone might actually find out. So they lie. They tell lie after lie, all to try and dissuade themselves and anyone else from ever looking.

And they don’t just tell lies. They engage in wildly irrational reasoning and incoherent thought. A rational human being who thought Bayes’ Theorem was not the logical formula that validates your conclusion that historicity (of Jesus or anything else) is probable, would endeavor immediately in the same essay to explain what the correct logical formula for doing that is. Yet never does anyone do this. Instead, like Briggs, they say we should just abandon all reasoning about probability entirely, and never mention probability ever again. In other words, they want us all to avoid ever justifying or validating any belief we have, about anything. Which is, conveniently, the only way false beliefs can survive scrutiny. Which tells you what this is really all about.

If you don’t know by what logic your claim that something is probable is valid, then you don’t know your conclusion is valid. And that should worry you. Until you’ve figured out why and how any conclusion of probability you ever reach is valid at all. And if you think it’s by some other formula that turns out to be invalid, then all your conclusions about the world have been invalid all this time. So what formula do you think it is? If it’s not Bayes’ Theorem, what is it? It’s therefore telling that Briggs criticizes using this formula to check our conclusions as to what’s probable, but offers no replacement for it. How does he know the historicity of Jesus is probable? He has no idea. And if you’re a Christian, that should worry you most of all.

Dear Christians:

The facts are not what you think. Your leaders, your apologists, your websites, lie to you. Really. Check. You’ll find so many lies, you will have every reason to stop listening to them. You can no longer trust them. They do not believe in telling you the truth. They are just like the Russian hackers trying to manipulate your beliefs with lies. Stop letting them. Take back ownership of your own mind. Stop being a dupe. For the sake of us all.