My review of Richard Carrier's "On the Historicity of Jesus"

Discussion about the New Testament, apocrypha, gnostics, church fathers, Christian origins, historical Jesus or otherwise, etc.
andrewcriddle
Posts: 1425
Joined: Sat Oct 05, 2013 12:36 am

Re: My review of Richard Carrier's "On the Historicity of Je

Post by andrewcriddle » Mon May 02, 2016 5:44 am

timhendrix wrote: ...................................
I am honestly surprised at how much of the historical information in OHJ which can be questioned. For instance I thought Carriers interpretation of the Ascension of Isiah was pretty rock solid, but your discussion and thread:

viewtopic.php?f=3&t=2346

just makes a lot of sense. I can see how one can interpret this nearly any way with some levity, but at the moment the text seems to make more sense as a description of how a heavenly being ended up dying on earth than anything else.
One issus about the Ascension of Isaiah is whether the Ethiopic or Latin/Slavonic version of the descent is original.

The recent (mainly Italian) scholarship tends to favour the originality of the Ethiopic text.
There is a long thread here

Andrew Criddle

User avatar
GakuseiDon
Posts: 639
Joined: Sat Oct 12, 2013 5:10 pm

Re: My review of Richard Carrier's "On the Historicity of Je

Post by GakuseiDon » Mon May 02, 2016 5:44 am

timhendrix wrote:I have been reading your review over the weekend and I think I understand most of it now :-). One thing that surprised me was that if I didn't know any better I would have thought I had simply structured my review around section 2 of yours... as far as I can tell we point out exactly the same things:

1) The problem of inflation of errors in BT (section 2 beginning)
2) That h and ~h are defined as specific hypothesis and not simply as "existence or not" and that they are not (as would be implied by the notation and subsequent computations) each others negation
3) That the Rank-Raglan prior computation appears to ignore this
4) ...and we are both puzzled by the quote on p245 that it does not matter how Jesus came to be a RR hero because it never happened to historical figures
5) we both immediately think that the principles can be applied to other reference classes to give inconsistent results
Hahaha! I actually thought the same thing initially: that you were working from my review. But my comments on Bayes Theorem were also based on others with BT experience whom had looked at Dr Carrier's usage and had come to the same conclusions as you, so it isn't really surprising we both reported the same problems.

I think I agree with Carrier that we tend to argue in Bayesian terms anyway, so BT can be used for questions of history. But one important comment I included in my review was by a poster called arcseconds, whom said that Carrier should have shown BT working on non-controversial examples first, to show some kind of calibration had been done, before using it on something as controversial as the historicity of Jesus. Carrier has "Proving History", but he barely covers any actual situations, to demonstrate "proof of concept" when it comes to questions of history.
timhendrix wrote:A point about the Gospels you highlight:
Since the Gospels, Paul and other early epistles talk about a crucified Jesus Christ whom was a descendent of David, I believe that even if we discount nearly all of the Gospels' content as unverifiable, then the mere existence of the same character appearing in the epistles of Paul and in the Gospels needs to be evaluated. But this, again, is an argument for another day.
There are two things that worries me. First that it seems like Carrier ends up discarding everything in the gospels as historically irrelevant, just with the exception of the RR-related information which turns out to be hugely historically relevant.. That seems a bit too convenient.

Secondly that the gospels is an excellent illustration of how the specific form of ~h helps explaining the evidence. Since we assume in ~h that Christians believed or taught a historical Jesus (at some point) it would seem that given ~h we can explain *any* amount of historical details about Jesus found in christian writings because ~h already assumes that was what Christians actually believed.
Yes, it is a bit like defining h to include "Paul is not interested in the historical Jesus, so won't include details about him", then pointing to Paul as evidence for h. 100% expected on minimal historicity! But I'm not sure what the alternative is. Shouldn't h and ~h be tailored to the evidence?
timhendrix wrote:I think it is a bit odd Carrier haven't responded directly to your review considering how many other reviews of much worse quality he has discussed, but perhaps it is a time issue.
I don't expect Carrier to respond to my review. I really am an amateur on the topic, so I don't think he even needs to (though I'd be interested to know whether I am a liar or whether I am insane as well! But I guess since I don't have a PhD, I'd only be a liar :( ). But I expect that when experienced scholars look at Carrier's work, they will see the same problems as I did, much like you and I saw many of the same problems with Carrier's use of BT. Carrier is going to be hammered for his ill-use of BT, his weird definition of 'euhemerism', his misreading of 'the Ascension of Isaiah' and Plutarch's 'Isis and Osiris', etc. And rightfully so.
It is really important, in life, to concentrate our minds on our enthusiasms, not on our dislikes. -- Roger Pearse

timhendrix
Posts: 62
Joined: Thu Mar 10, 2016 3:56 am
Contact:

Re: My review of Richard Carrier's "On the Historicity of Je

Post by timhendrix » Mon May 02, 2016 10:49 am

Andrew:
Just to summarize, if the original text is closer to the Eheopian version that would support Carriers hypothesis compared to the case the original text is closer to the Slavonic?

GD:
I think I agree with Carrier that we tend to argue in Bayesian terms anyway, so BT can be used for questions of history. But one important comment I included in my review was by a poster called arcseconds, whom said that Carrier should have shown BT working on non-controversial examples first, to show some kind of calibration had been done, before using it on something as controversial as the historicity of Jesus. Carrier has "Proving History", but he barely covers any actual situations, to demonstrate "proof of concept" when it comes to questions of history.
That's also exactly what I thought after reading PH. I think the truth is that if Bayes is applied to "prove" historical propositions that are known the results are simply going to seem very obvious in hindsight. Something I see as more promising if one wish to apply Bayes to history is Aviezer Tuckers book (see a review here https://ndpr.nd.edu/news/22998-our-know ... riography/; it was published in 2004 and easily found by googling "Bayes" and "historiography" so I got no idea why Carrier is not citing it).
Tuckers focus is different than Carrier since he does not appear to be interested in using numbers to answer historical questions but in examining historical arguments in Bayesian terms such as the importance of multiple attestations. This is in my opinion much more likely to be useful (but it is really up to historians to judge) since you don't have to rely on the exact numbers. Tucker also has a review of PH if you are interested.

I think a lot of the discussion on applying Bayes to history misses the boat because it focuses on whether probabilities *in principle* can describe a historical argument (I think the answer is yes) and not if it is practical.. Carriers proof in PH only considers the first case and is a tautology if spelled out. It's a bit like asking if Newtons laws can tell us about traffic safety. In a way it can because what happens in traffic obeys Newtons laws, but it might not be practical.
Yes, it is a bit like defining h to include "Paul is not interested in the historical Jesus, so won't include details about him", then pointing to Paul as evidence for h. 100% expected on minimal historicity! But I'm not sure what the alternative is. Shouldn't h and ~h be tailored to the evidence?
I think the most obvious idea is to come up with a truly "minimal" historicity and non-historicity (lets call it H and ~H and assume they are defined to actually be each others negations) and then try to assign these a prior. The prior probability of for instance ~h would then by p(~h) = p(~h.~H) = p(~h|~H)p(~H) and so the argument would be on how likely the extra elements in ~h over ~H are.
I think an obstacle here is that the concept of minimal historicity might not be easy to define. For instance we should properly include the general area where "minimal jesus" lived (0-50CE) and region (Judea) but we got to be careful not to include too much because then the prior becomes very complicated. For instance just including the time where he lived and that subsequent christians believed he was real later (for instance 50-100CE) that would properly already tip the prior in favor of historicy as you pointed out in the discussion of the Gospels: Most people who are believed to have lived 50 years ago actually lived.

outhouse
Posts: 3525
Joined: Fri Oct 04, 2013 6:48 pm

Re: My review of Richard Carrier's "On the Historicity of Je

Post by outhouse » Mon May 02, 2016 2:18 pm

MrMacSon wrote:Consensus can mean anything above 50% --as in election results.
Ill call bs on that one

There is no real vote, just people that all agree on historicity, less the oddballs. And FREAKS do not count as 49%

User avatar
Kapyong
Posts: 489
Joined: Mon Oct 07, 2013 4:51 pm
Contact:

Re: My review of Richard Carrier's "On the Historicity of Je

Post by Kapyong » Mon May 02, 2016 4:58 pm

Gday all,
outhouse wrote:There is no real vote, just people that all agree on historicity, less the oddballs. And FREAKS do not count as 49%
Everyone agrees !
(Because anyone who doesn't agree is just a freak or an oddball.)


Kapyong

User avatar
MrMacSon
Posts: 5162
Joined: Sat Oct 05, 2013 3:45 pm

Re: My review of Richard Carrier's "On the Historicity of Je

Post by MrMacSon » Mon May 02, 2016 7:05 pm

timhendrix wrote:
Something I see as more promising if one wish to apply Bayes to history is Aviezer Tuckers book (see a review here https://ndpr.nd.edu/news/22998-our-know ... riography/ )* - it was published in 2004 and easily found by googling "Bayes" and "historiography" so I got no idea why Carrier is not citing it).
  • * the 'link' in your post picks up the semi-colon as part of the url, so it doesn't work (replacing the ; with a ')' should help)
Carrier has said -
"After I published Proving History a reader said I should check out Aviezer Tucker’s book 'Our Knowledge of the Past: A Philosophy of Historiography', since it appeared to back up the entire core thesis of my book. I am amazed and ashamed that I did not discover this book sooner. It must not have been indexed well in databases, since my searches for Bayesian historiography did not discover it ...

"It’s a great shame that Proving History does not cite it, and I am writing this review now to redress that gap ..."

http://freethoughtblogs.com/carrier/archives/3923
timhendrix wrote:
  • Tucker also has a review of PH if you are interested.
Aviezer Tucker (2016) 'The Reverend Bayes v Jesus Christ' History and Theory Vol 55, Issue 1, pp.129–140.

Abstract
The Bayesian perspective on historiography is commonsensical: If historiography is not certain like a priori knowledge or sense data, and it is not fiction, historiography is probable. Richard Carrier's book argues for a Bayesian, probabilistic interpretation of historiography in general and of the debate about the historicity of Jesus in particular. Jesus can be interpreted as a historically transmitted reference of “Jesus,” as a bundle of properties, or literally. Carrier devotes too much energy to debating literalism that confuses evidence with hypotheses. But evidence preserves information to different degrees; it is true or not. Carrier proposes to apply objective, frequentist Bayesianism in historiography despite the difficulties in assigning values. He argues that ranges of values can determine historiographical hypotheses. Carrier does not analyze in Bayesian terms the main method for Bayesian determination of posterior probabilities in historiography: inference from multiple independent sources. When the prior probability of a hypothesis is low, but at least two independent evidential sources, such as testimonies, support it, however unreliable each of the testimonies is, the posterior probability leaps.

The problem with the Synoptic Gospels as evidence for a historical Jesus from a Bayesian perspective is that the evidence that coheres does not seem to be independent, whereas the evidence that is independent does not seem to cohere. Carrier's explanation of some the evidence in the Gospels is fascinating as the first Bayesian reconstruction of structuralism and mimesis. Historians attempted to use theories about the transmission and preservation of information to find more reliable parts of the Gospels, parts that are more likely to have preserved older information. Carrier is too dismissive of such methods because he is focused on hypotheses about the historical Jesus rather than on the best explanations of the evidence. I leave open questions about the degree of scholarly consensus and the possible reasons for it. http://onlinelibrary.wiley.com/doi/10.1 ... 1/abstract
Carrier has responded in a blog-post - http://freethoughtblogs.com/carrier/archives/9619
.

User avatar
GakuseiDon
Posts: 639
Joined: Sat Oct 12, 2013 5:10 pm

Re: My review of Richard Carrier's "On the Historicity of Je

Post by GakuseiDon » Tue May 03, 2016 1:24 am

timhendrix wrote:Andrew:
Just to summarize, if the original text is closer to the Eheopian version that would support Carriers hypothesis compared to the case the original text is closer to the Slavonic?
I'll field this one. The answer is "no". The Ethiopic text (written in Greek) is presumed to be a later text, containing a 'pocket Gospel' with Jesus being born on earth to a virgin Mary, etc. So it would go against Carrier's hypothesis if it were earlier. The Slavonic and Latin2 texts (written in Latin, and largely consistent with each other) appear to be more primitive versions, missing out on the 'pocket gospel' and other passages. Carrier uses the S/L2 texts to advance his hypothesis.
timhendrix wrote:
Yes, it is a bit like defining h to include "Paul is not interested in the historical Jesus, so won't include details about him", then pointing to Paul as evidence for h. 100% expected on minimal historicity! But I'm not sure what the alternative is. Shouldn't h and ~h be tailored to the evidence?
I think the most obvious idea is to come up with a truly "minimal" historicity and non-historicity (lets call it H and ~H and assume they are defined to actually be each others negations) and then try to assign these a prior. The prior probability of for instance ~h would then by p(~h) = p(~h.~H) = p(~h|~H)p(~H) and so the argument would be on how likely the extra elements in ~h over ~H are.
I think an obstacle here is that the concept of minimal historicity might not be easy to define. For instance we should properly include the general area where "minimal jesus" lived (0-50CE) and region (Judea) but we got to be careful not to include too much because then the prior becomes very complicated. For instance just including the time where he lived and that subsequent christians believed he was real later (for instance 50-100CE) that would properly already tip the prior in favor of historicy as you pointed out in the discussion of the Gospels: Most people who are believed to have lived 50 years ago actually lived.
Out of interest, if h was defined as "lived in first half of the First Century" and "lived in Judea", what would ~h be? "Lived in first half of the First Century" and "NOT lived in Judea"?
It is really important, in life, to concentrate our minds on our enthusiasms, not on our dislikes. -- Roger Pearse

maryhelena
Posts: 1488
Joined: Tue Oct 08, 2013 11:22 pm
Location: England

Re: My review of Richard Carrier's "On the Historicity of Je

Post by maryhelena » Tue May 03, 2016 2:11 am

GakuseiDon wrote: Out of interest, if h was defined as "lived in first half of the First Century" and "lived in Judea", what would ~h be? "Lived in first half of the First Century" and "NOT lived in Judea"?
Now, that is an interesting question! Not simply in regard to Bayes theorem (on which I can't comment....) but in the fact that it 1) opens up the debate for historicity to include a wider area than Judea and 2) has removed crucifixion from being a definition of a messiah figure. (note Wells on his non-crucified Galilean preacher figure)

1) could include the figure of Philip the Tetrarch; ruler of the region of Ituraea and Trachonitis, including Bethsaida - the place where some disciples came from. It was in the area of his capital, Caesarea Philippi, that the question of the messiahship of Jesus arose. Philip, re Josephus, died in 34 c.e. in the time of Pontius Pilate.

2) could include the figure of Agrippa I. Josephus making allusions of messiahship towards this ruler; rebuilding the walls of Jerusalem; the silver of his garments shine from the rays of the sun - the star prophecy; his Joseph type imprisonment and his subsequent elevation - and gift of a gold chain. Agrippa died in the reign of Claudius in 44 c.e.

Methinks, lots of historical possibilities open up when the research area is widened and the crucifixion is removed from the time of Pilate or Claudius. (i.e. the Roman execution of a King of the Jews took place earlier, in 37 b.c.e.)

A composite Jesus figure could well be reflecting the lives of historical figures - historical figures the gospel writers found to be relevant for their story.
Tread softly because you tread on my dreams.
W.B. Yeats

timhendrix
Posts: 62
Joined: Thu Mar 10, 2016 3:56 am
Contact:

Re: My review of Richard Carrier's "On the Historicity of Je

Post by timhendrix » Tue May 03, 2016 3:10 am

GakuseiDon wrote: I'll field this one. The answer is "no". The Ethiopic text (written in Greek) is presumed to be a later text, containing a 'pocket Gospel' with Jesus being born on earth to a virgin Mary, etc. So it would go against Carrier's hypothesis if it were earlier. The Slavonic and Latin2 texts (written in Latin, and largely consistent with each other) appear to be more primitive versions, missing out on the 'pocket gospel' and other passages. Carrier uses the S/L2 texts to advance his hypothesis.
Thanks, I am amazed anyone can keep all this information straight. According to wikipedia, the ascension of Isaiah was composed around late 1st century to early 3rd century AD. I suppose that would place it in around the time where John (where Jesus/the word is also a heavenly pre-existing creature?) was composed and at a stage where Carrier (and many scholars) argue that Christians was making up details about Jesus left and right. I guess what I am wondering is why this particular heavenly journey would be so important when other details from around that time are not? (e.g. details in the Gospels which Carrier sees as nearly fully made up).
GakuseiDon wrote: Out of interest, if h was defined as "lived in first half of the First Century" and "lived in Judea", what would ~h be? "Lived in first half of the First Century" and "NOT lived in Judea"?
Just logically, if

h = (lived in first half of first century) and (lived in judea)

then

~h = (did not live in first half of first century) or (did not live in judea)

so defined in this way ~h would not match John the Baptist but it would match Josepheus and Budda.. that doesn't quite seem right. If you are interested, Tucker briefly discusses these issues in his review of proving history.

User avatar
Ben C. Smith
Posts: 4479
Joined: Wed Apr 08, 2015 2:18 pm
Location: USA
Contact:

Re: My review of Richard Carrier's "On the Historicity of Je

Post by Ben C. Smith » Tue May 03, 2016 4:38 am

GakuseiDon wrote:The Ethiopic text (written in Greek)....
Written in Greek? Or do you mean originally written in Greek (before being translated into Ethiopic)?

https://archive.org/stream/cu3192401459 ... 1/mode/2up.
ΤΙ ΕΣΤΙΝ ΑΛΕΘΕΙΑ

Post Reply