Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Jade comments on Crisis of Faith - Less Wrong

57 Post author: Eliezer_Yudkowsky 10 October 2008 10:08PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (243)

Sort By: Old

You are viewing a single comment's thread. Show more comments above.

Comment author: Jade 11 January 2017 02:53:48AM 1 point [-]

Would you say the origins of other religions become more mysterious if there never were whatever magical beings those religions posit? Would you think it likely that Guanyin was real human of unknown gender? Do the origins of fictional stories become more mysterious if there never were the fictitious characters in the flesh? Did Paul Bunyan exist, as there were similar lumberjacks?

You're not supposed to tie yourself to any hypothesis, even if mainstream, but rather update your probability distributions. Bits of the NT weren't written until long enough after the supposed death of Jesus that people wouldn't have been like, 'Who you talkin' about?' And I doubt they would've cared whether the character existed, like no one cares whether Harry Potter existed, because it's the stories that matter.

Comment author: Salemicus 12 January 2017 11:36:08AM 1 point [-]

Would you say the origins of other religions become more mysterious if there never were whatever magical beings those religions posit?

Yes, of course.

The least mysterious explanation of Paul Bunyan stories is that there really was a Paul Bunyan. And the closer the real Paul Bunyan hews to the Bunyan of the stories, the smaller the mystery. P(stories about Bunyan | Bunyan) > P(stories about Bunyan | !Bunyan).

But just because a story is simple, doesn't necessarily make it likely. We can't conclude from the above that P(Bunyan | stories about Bunyan) > P(!Bunyan | stories about Bunyan).

Comment author: CCC 12 January 2017 05:19:42PM *  1 point [-]

Hmmm. To mess around with equations a bit... what can we say about P(Bunyan | stories about Bunyan) and P(!Bunyan | stories about Bunyan), given P(stories about Bunyan | Bunyan) > P(stories about Bunyan | !Bunyan)?

Let's genaralise it a bit (and reduce typing). What can we say about P(A|B) and P(!A|B) when P(B|A) > P(B|!A)?

Consider Bayes' Theorem: P(A|B) = [(P(B|A)*P(A)]/P(B). Thus, P(B) = [(P(B|A)*P(A)]/P(A|B)

Therefore, P(!A|B) = [(P(B|!A)*P(!A)]/P(B)

Now, P(!A) = 1-P(A). So:

P(!A|B) = [(P(B|!A)*{1-P(A)}]/P(B)

Solve for P(B):

P(B) = [(P(B|!A)*{1-P(A)}]/P(!A|B)

Since P(B) = [(P(B|A)*P(A)]/P(A|B):

[(P(B|A)*P(A)]/P(A|B) = [(P(B|!A)*{1-P(A)}]/P(!A|B)

Since P(B|A) > P(B|!A)

[(P(B|A)*P(A)]/P(A|B) > [(P(B|!A)*P(A)]/P(A|B)


[(P(B|!A)*{1-P(A)}]/P(!A|B) > [(P(B|!A)*P(A)]/P(A|B)

Since probabilities cannot be negative:

[{1-P(A)}]/P(!A|B) > [P(A)]/P(A|B)

.[1-P(A)]*P(A|B) > [P(A)]*P(!A|B)

...which means that either (1-P(A)) > P(A) or P(A|B) > P(!A|B), and quite possibly both; and whichever of these two inequalities is false (if either) the ratio between the two sides is closer than the inequality that is true.

To return to the original example; either P(Bunyan | stories about Bunyan) > P(!Bunyan | stories about Bunyan) OR P(!Bunyan) > P(Bunyan).

Also, if P(Bunyan | stories about Bunyan) > P(!Bunyan | stories about Bunyan) is false, then it must be true that P(Bunyan|stories about Bunyan) > P(Bunyan).

Comment author: Jade 16 January 2017 03:00:52AM *  1 point [-]

You left out the 'magical' part of my question. If magical beings exist(ed), then everything becomes more mysterious. That's partly why we don't pester JK Rowling about what extra-special boy Harry Potter was based on. We don't even suspect comic superheros like Batman, who has no magic, to have been based on a real-life billionaire. We certainly don't have scholars wasting time looking for evidence of 'the real Batman.' Modern stories of unlikely events are easily taken as imaginings, yet when people bucket a story as 'old/traditonal', for some people, that bucket includes 'characters must've been real persons', as if humans must've been too stupid to have imagination. https://en.wikipedia.org/wiki/Fakelore

Comment author: Salemicus 19 January 2017 12:08:27PM *  1 point [-]

No, I didn't leave that part out.

the closer the real Paul Bunyan hews to the Bunyan of the stories, the smaller the mystery

Of course magic makes everything else more mysterious i.e. P(magical Jesus) is infinitesimal. But P(non magical Jesus) is not low. We do ask JK Rowling what non magical boy inspired Harry Potter.

Comment author: CCC 19 January 2017 12:22:03PM 1 point [-]

Interestingly, after looking over Wikipedia a bit, apparently there may have been a Paul Bon Jean on whom the earliest Paul Bunyan tales could have been based... a big lumberjack, but with "big" being more like six to seven foot and less like sixty to seventy foot.

Comment author: gjm 20 January 2017 02:35:36PM 1 point [-]

We do ask J K Rowling what non-magical boy inspired Harry Potter.

I guess you mean that we could and it wouldn't be obviously silly, with which I agree. But, for what it's worth, it never crossed my mind to assume that Harry Potter was based on any specific non-magical boy. The characteristics he has that aren't essentially dependent on story-specific things (magic, being the prime target of a supervillain, etc.) seem pretty ordinary and not in any particular need of explanation.

I wouldn't be astonished if it turned out that there was some kid Rowling knew once whom she used as a sort of basis for the character of Harry Potter, but I'd be a bit surprised. And if it did, I wouldn't expect particular incidents in the books to be derived from particular things that happened to that child.

In particular, I wouldn't say that the simplest (still less the most likely) explanation for the Harry Potter stories involves there being some non-magical child on whom they are based.

I don't think any of this has much bearing on whether the simplest explanation for stories about Jesus, Muhammad, the Buddha, Zeus, etc., involves actual historical characters on which they're based. The answer to that surely varies a lot from case to case. (FWIW I'd say: historical Jesus of some sort likely but not certain; historical Muhammad almost certain; historical Buddha likely but not certain; historical Zeus-predecessor very unlikely. But I am not expert enough for my guesses to be worth anything.)

Comment author: Jade 23 January 2017 02:57:37AM *  0 points [-]

Historical Muhammad not certain: http://www.wsj.com/articles/SB122669909279629451 . Of course, people have set about trying to protect minds from a 'fringe' Bayesian view: "Prof. Kalisch was told he could keep his professorship but must stop teaching Islam to future school teachers." In case anyone missed it, Richard Carrier explicitly used Bayes on question of historical Jesus. I don't know if Kalisch used Bayes, but his language conveys intuitive Bayesian update.

The bearing of fictional stories is simple: calculate probabilities of historical X based on practically 100% probability that human imagination was a factor (given that the stories contain highly unlikely magic like in known-to-be fiction stories, plus were written long after X supposedly lived). Note that that still leaves out probabilities of motivations for passing fiction as nonfiction like Joseph Smith or L. Ron Hubbard did. Once you figure probabilities including motivations and iterations of previous religious memes, it becomes increasingly unlikely that X existed. Paul Bunyan, AFAIK, wasn't based on previous memes for controlling people, nor were the stories used to control people, so I wouldn't be suspicious if someone believed the stories started based on someone real. When people insist religious characters were real, OTOH, I become suspicious of their motivations, given unlikelihood that they examined evidence and updated Bayesian-like.

@Salemicus: Citation for "We do ask JK Rowling what non magical boy inspired Harry Potter"?

Comment author: gjm 23 January 2017 10:28:47AM 0 points [-]

Historical Muhammad not certain

I did say almost certain. My impression -- which, as I said above, is no more than that and could easily be very wrong -- is that the Jesus-myth theories require less "conspiracy" than the Muhammad-myth ones.

Comment author: Lumifer 23 January 2017 05:47:52PM 2 points [-]

Historical Muhammad not certain

What's your comparison baseline? Compared to the screen in front of your face, he's not certain. Compared to pretty much anyone born in the VI century, he is quite certain.

Comment author: Jade 24 January 2017 01:56:00AM *  0 points [-]

Then why don't you just point to evidence of his existence being more likely than others'? We have bodily remains, intact own writings, or historical records made during the lives of many born in 6th century, e.g. Columbanus, Pope Gregory I, founding emperor of Tang Dynasty, Radegund, Venantius Fortunatus, Theodora). So why don't we have any one of those types of evidence about Muhammad?

Comment author: Lumifer 24 January 2017 03:11:56AM *  1 point [-]

You don't count the Koran as "intact own writings"? :-) Yes, I am well aware that it was compiled quite some time after his death from a collection of records and that, by tradition, Muhammad was illiterate.

The Arab society around VII century wasn't big on writing -- the cultural transmission was mostly oral. However external sources mention Muhammad already in 636 AD.

Comment author: Jade 25 January 2017 12:43:05PM *  0 points [-]

You're referring to the phrase "many villages were ravaged by the killing of the Arabs of Muhammad", written after Muhammad's supposed death, "Arabs of Muhammad" meaning 'Muslims' the way "people of Christ" means 'Christians'. That Muslims and Christians existed doesn't mean the characters they invoked to justify violence, supremacism, etc. existed as actual humans.

Criteria for considering Muhammad and Jesus near certain are so lax, we'd have to consider some Greek/Roman gods near certain.

Comment deleted 31 January 2017 02:29:35AM [-]
Comment author: gjm 31 January 2017 02:50:55AM 1 point [-]

no one questions the historicity of say Pythagoras


Here's a newspaper review whose author says Pythagoras "may well be a mythical amalgam of various forgotten sages". The book under review itself says "Sadly, it is now almost universally assumed by classical scholars that Pythagoras never existed". I suspect this is partly tongue in cheek, since the other information I can find doesn't seem consistent with what it says on its face, but if it's a joke I think it's the sort that depends on not being too far from the truth. Here's the History Channel suggesting Pythagoras may not have existed. Everything I can find about Pythagoras, scholarly or popular, emphasizes that our sources of information about him are late and untrustworthy and that scarcely anything is known about him.

It looks to me as if the usual belief is "probably real but essentially nothing is actually known about his life", and a few people, mostly not actual scholars, say "actually, the evidence is so thin he may well not have been real". Which is not so different from the situation with Jesus, except that most people who aren't out-and-out mythicists about Jesus are willing to concede that some things are known about his life.