Carrier's numbers and math in OHJ

Discussion about the New Testament, apocrypha, gnostics, church fathers, Christian origins, historical Jesus or otherwise, etc.
User avatar
GakuseiDon
Posts: 2296
Joined: Sat Oct 12, 2013 5:10 pm

Re: Carrier's numbers and math in OHJ

Post by GakuseiDon »

Bernard, unsurprisingly perhaps, I agree with your numbers more than with Carrier's! But do you think that Bayes Theorem is in fact a valid approach towards deciding historicity after all?

I've read that something like Bayes Theorem is really being applied unknowingly anyway for these types of questions, so formally applying BT makes sense. I'm still not sure. I'm wondering that if we can't use BT here, then it may be because the question being examined -- historicity of Jesus -- cannot be answered confidently at all. Perhaps I should start a poll on this.
It is really important, in life, to concentrate our minds on our enthusiasms, not on our dislikes. -- Roger Pearse
User avatar
Peter Kirby
Site Admin
Posts: 8048
Joined: Fri Oct 04, 2013 2:13 pm
Location: Santa Clara
Contact:

Re: Carrier's numbers and math in OHJ

Post by Peter Kirby »

Bernard Muller wrote:
I arrived to an astounding 99.99% probability for historicity. And yes, I kept Carrier's math intact, with its multiplications of odds.
The conclusion follows from the premises. You can expect contention, such as it is, to be in regard to the premises.

It takes a special kind of obstinacy to fuss overmuch with the "math" and "multiplications" here. You shouldn't expect it from other people, who I am sure will be much more charitable (and much less special) than you have been with Carrier in this regard.
I did not fuss here about Carrier's multiplications, just telling I did not change Carrier's math.
And I explained thoroughly on my blog post how I arrived to these 99.99% and 7200/1: http://historical-jesus.info/110.html
Yes the contention is based on premises. I never meant it was a critique of Carrier's math. Actually, Carrier's math is to my advantage because, with the same numbers (including my "corrections") but calculating the consequent odds by additions then averaging, the result would be only a probability of 82%.
Okay, right. No problem. But the math is just a formal shell for what either of you are expressing. The underlying point here is not the math, nor the numbers. The real point here (apparently) is that Richard Carrier and Bernard Muller disagree in their assessment of the evidence regarding 1 Corinthians 9:5/Galatians 1:19, Galatians 4:4, and Romans 1:3. We might have saved some trouble focusing on this point of really significant disagreement.

Incidentally, a problem with Carrier's "naive" application of Bayes' theorem--a genuine problem, which I've discussed, unlike some of the pseudo-problems that have been proposed in this thread--becomes even more apparent when we use the extreme 50:1, 20:1, and 20:1 numbers that are proposed.

That problem is the assumption of conditional independence. I don't attempt to explain it in a way that you understand (not sure how I'd do that, actually), but it is a problem. Basically speaking, what you really want isn't P(e1|h)P(e2|h), when computing P(e1, e2 | h), as that assumes conditional independence. What you really want is P(e1|h)P(e2|h,e1). And that does make a difference.

What is the difference? We're not really supposed to be asking just "what is the probability that Galatians 4:4 would have 'born of a woman', on h and on ~h" but rather "what is the probability that Galatians 4:4 would have 'born of a woman', on h and on ~h and with all previous items combined," including the previous item that Romans 1:3 has the 'seed of David' bit. I'm pretty sure your argument is that the seed of David bit would also imply the born of woman bit. Therefore you'd have to place P(e2|h,e1) and P(e2|~h,e1) as both being very high, because it's a conditional probability under the 'seed of David' bit as well and not just under the hypothesis or its negation.

Likewise, you would have to look at the 'brother(s) of the Lord' references similarly, to determine whether the other information should change your assessment of conditional probability. Perhaps it doesn't here because 'seed of David' and 'born of woman' doesn't imply a reference to brother(s) of the Lord. But it'd have to be considered in any case.

With this in mind, and even accepting the full strength of your exegetical argument, the substitution of 10:1, 1:1 (yes, this implies that Carrier is wrong here, in that he doesn't consider the problems with the assumption of conditional independence), and 5:1, respectively, for the arguments regarding Romans 1:3, Galatians 4:4, and 1 Corinthians 9:5/Galatians 1:19, would appear to be a more-than-reasonable numerical summary of your own hermeneutical argument and the great weight that you attach to it. (I.e., if one were to agree with all your words and only disagree with your attempt to express it in numbers.)

Of course that'd still tilt the total towards historicity, just not in quite such an "astounding" fashion. With only these substitutions, one would arrive at P(h|e,b) of 97.9% and not 99.996% ... and, in fact, don't you think a reasonable person, even a reasonable historicist (since we are using historicist-argument numbers), would consider the former's 2.1% chance of the non-existence of Jesus a somewhat better approximation than the less than 1 in 20,000 implied by the latter?
"... almost every critical biblical position was earlier advanced by skeptics." - Raymond Brown
User avatar
Peter Kirby
Site Admin
Posts: 8048
Joined: Fri Oct 04, 2013 2:13 pm
Location: Santa Clara
Contact:

Re: Carrier's numbers and math in OHJ

Post by Peter Kirby »

GakuseiDon wrote:Bernard, unsurprisingly perhaps, I agree with your numbers more than with Carrier's! But do you think that Bayes Theorem is in fact a valid approach towards deciding historicity after all?

I've read that something like Bayes Theorem is really being applied unknowingly anyway for these types of questions, so formally applying BT makes sense. I'm still not sure. I'm wondering that if we can't use BT here, then it may be because the question being examined -- historicity of Jesus -- cannot be answered confidently at all. Perhaps I should start a poll on this.
The historicity of Jesus might not be answerable (IDK), but, if you accept some of the arguments being made in this thread (regarding the extremely low likelihood of Carrier's interpretation of Pauline texts combined with an implicit acceptance of the authenticity of the same), the falsehood of Carrier's "minimal mythicism" might indeed be answerable, even if the historicity of Jesus might not be.

And this is of course a problem with Carrier's bifurcation of the probability space into two specific hypotheses. Because even if the rest of the non-historicity probability space has low prior probability somehow (and I'm not really sure why it should, actually), there's the possibility that the rest of the non-historicity probability space might be far more compatible with the evidence (much better consequent probabilities), leading the rest of the non-historicity probability space to fare relatively well when considering the final, posterior probabilities. Thus, Carrier's bifurcation could lead to a gross underestimate of the probability of the non-historicity of Jesus (a problem of method that Carrier too quickly neglects due to his emphasis on giving his opponents as much rope as possible to hang him with). You could indeed argue that this has happened when looking at Bernard's result of P(~h|e,b) of ~0.004%, which is arrived at by the rejection of Carrier's particular "miniminal mythicism" (especially because of assigning it low consequent probability due to some particular evidence, chiefly some verses in the letters of Paul) and not a general consideration of all possible non-historicity hypotheses.
"... almost every critical biblical position was earlier advanced by skeptics." - Raymond Brown
User avatar
Peter Kirby
Site Admin
Posts: 8048
Joined: Fri Oct 04, 2013 2:13 pm
Location: Santa Clara
Contact:

Re: Carrier's numbers and math in OHJ

Post by Peter Kirby »

Earl Doherty's November 3, 2000 "inaugural address" to the Jesus Mysteries group ("A Welcome Forum") actually touches on both issues of the uncertainty involved in historical assessment and the multiplicity of non-historicity hypotheses, which are now exercising us here in this thread fifteen years later.
Earl Doherty wrote:I for one will be willing to consider and acknowledge any good argument for
the existence of Jesus. A given argument by itself does not necessarily
prove the case either way, and one does not need to score all the points
for one's own side. Nothing in historical research can achieve that kind of
blanket or mathematical certainty. We can only aim for the balance of
probability, based on the evidence. I sometimes say that my viewpoint needs
only to persuade, not prove; to commend to the reasonable, unprejudiced
person not locked into rigid confessional interests, that the evidence of
the early Christian record, together with elements of the wider ancient
picture, strongly indicates that there was no historical Jesus. To produce
that 'conviction of probability' is all one can hope to achieve, or all one
*needs* to.

Even the mythicist position is not monolithic. Our bottom line may be that
the figure portrayed in the Gospels is fictional, that early Christian
prophets like Paul were not preaching a recent man who was a teacher and
miracle worker in Galilee. Beyond that, there is room for maneuver. Most
people interested in this field know that researchers like G. A. Wells and
Alvar Ellegard have concluded that early Christians believed their god-man
Jesus had lived on earth at a more distant time in the past (Ellegard that
he was an actual man known to the Essenes as the Teacher of Righteousness,
Wells that he was only an imagined historical figure who had lived
obscurely). Wells has also recently allowed for the possible existence of a
teaching sage at the beginnings of the first-century Kingdom of God
movement in Galilee. I and other mythicists disagree with such
interpretations. But these variations are relatively minor. They don't
change the fact that the Gospel presentation and its central character are
basically fictional and symbolic (its spirit and details derived from
scripture and wider sources), and that Paul's faith, with its heavenly
Christ and system of salvation very similar to the Greek mystery cult
ethos, has nothing to do with a recent or actual historical man.
"... almost every critical biblical position was earlier advanced by skeptics." - Raymond Brown
Bernard Muller
Posts: 3964
Joined: Tue Oct 15, 2013 6:02 pm
Contact:

Re: Carrier's numbers and math in OHJ

Post by Bernard Muller »

to Gakuseidon,
Bernard, unsurprisingly perhaps, I agree with your numbers more than with Carrier's! But do you think that Bayes Theorem is in fact a valid approach towards deciding historicity after all?
Thanks, finally something encouraging (I got that from Peter also, which came as a surprise: but thanks for both of you).
I think not only the Bayes theorem for this application is wrong (except for the final one, but with no valid prior odd, as I proposed (with justifications), it is useless), because too much dependent on suggestive input data, as I demonstrated on my blog post. But no need to blame that on the Bayes theorem: any other mathematical apparatus would do the same. However these multiplications of consequent odds & their components are suspiciously amplifying greatly any deviation from 1/1, as I have shown.
I've read that something like Bayes Theorem is really being applied unknowingly anyway for these types of questions, so formally applying BT makes sense. I'm still not sure. I'm wondering that if we can't use BT here, then it may be because the question being examined -- historicity of Jesus -- cannot be answered confidently at all. Perhaps I should start a poll on this.
A poll will not solve any problem, but you still can do it. That might be interesting.
I wish expert mathematicians on odds and probabilities could be consulted. Actually I remember you mentioned one or two of them on the Amazon review of Carrier book who were not favorable on Carrier's math: viewtopic.php?f=3&t=1317&p=31298&hilit= ... ian#p31298

Cordially, Bernard
I believe freedom of expression should not be curtailed
Bernard Muller
Posts: 3964
Joined: Tue Oct 15, 2013 6:02 pm
Contact:

Re: Carrier's numbers and math in OHJ

Post by Bernard Muller »

to Peter,
Okay, right. No problem. But the math is just a formal shell for what either of you are expressing. The underlying point here is not the math, nor the numbers. The real point here (apparently) is that Richard Carrier and Bernard Muller disagree in their assessment of the evidence regarding 1 Corinthians 9:5/Galatians 1:19, Galatians 4:4, and Romans 1:3. We might have saved some trouble focusing on this point of really significant disagreement.
I focused greatly on 1 Corinthians 9:5/Galatians 1:19, Galatians 4:4, and Romans 1:3 in my blog posts, as I indicated on my latest blog post with its links (http://historical-jesus.info/110.html) and even already on this forum (more so on this thread for "Made from sperm").
Incidentally, a problem with Carrier's "naive" application of Bayes' theorem--a genuine problem, which I've discussed, unlike some of the pseudo-problems that have been proposed in this thread--becomes even more apparent when we use the extreme 50:1, 20:1, and 20:1 numbers that are proposed.
Yes, you said it.
That problem is the assumption of conditional independence. I don't attempt to explain it in a way that you understand (not sure how I'd do that, actually), but it is a problem. Basically speaking, what you really want isn't P(e1|h)P(e2|h), when computing P(e1, e2 | h), as that assumes conditional independence. What you really want is P(e1|h)P(e2|h,e1). And that does make a difference.
I want to believe you, but it is very fuzzy in my mind. Why don't you do use your equation for at least one of Carrier's table such as for EPISTLES. BTW, I do not think my numbers are extreme after considering Carrier's counter arguments. I refrained to write 1000/1 for "made from sperm", which is what I would prefer as a maximum. However, I read much better arguments against "seed of David" but my blog post is only about Carrier's OHJ, not about other mythicist's books.
What is the difference? We're not really supposed to be asking just "what is the probability that Galatians 4:4 would have 'born of a woman', on h and on ~h" but rather "what is the probability that Galatians 4:4 would have 'born of a woman', on h and on ~h and with all previous items combined," including the previous item that Romans 1:3 has the 'seed of David' bit. I'm pretty sure your argument is that the seed of David bit would also imply the born of woman bit. Therefore you'd have to place P(e2|h,e1) and P(e2|~h,e1) as both being very high, because it's a conditional probability under the 'seed of David' bit as well and not just under the hypothesis or its negation.
I explored that ("but rather "what is the probability that Galatians 4:4 would have 'born of a woman'") thoroughly here: http://historical-jesus.info/18.html. Compare that to my comments about Carrier's position on that topic: http://historical-jesus.info/72.html
and
http://historical-jesus.info/95.html
Yes. I admit I used corroborations (which can be called conditional probability) as Carrier did, with "sperm of David", AoI and the 'Romans' doxology, among other bits. But I think that backfires for Carrier:
- "sperm of David" is the weakest counter argument of Carrier, based on something unevidenced, a product of Carrier's imagination: "It would not be unimaginable that God could maintain a cosmic sperm bank." and "it is not an improbable assumption" http://historical-jesus.info/70.html
- AoI does not say Jesus was arrested and crucified in the firmament or in the air below it, nor that was done by demons spirits and that in every known versions. More details here: http://historical-jesus.info/100.html
- the 'Romans' doxology is most likely an interpolation and, anyway, does not say what Carrier want it to say: http://historical-jesus.info/60.html

It's a bit like using a picture of earth from the moon as corroborating evidence for earth having corners :lol: .

One definition for "corroboration": Corroborating evidence (in "corroboration") is evidence that tends to support a proposition that is already supported by some initial evidence, therefore confirming the proposition. from http://en.wikipedia.org/wiki/Corroborating_evidence
Likewise, you would have to look at the 'brother(s) of the Lord' references similarly, to determine whether the other information should change your assessment of conditional probability. Perhaps it doesn't here because 'seed of David' and 'born of woman' doesn't imply a reference to brother(s) of the Lord. But it'd have to be considered in any case.
Yes, yes, yes. But corroborations come from Josephus' Antiquities 20, 9, 1 and Mark's gospel.
With this in mind, and even accepting the full strength of your exegetical argument, the substitution of 10:1, 1:1 (yes, this implies that Carrier is wrong here, in that he doesn't consider the problems with the assumption of conditional independence), and 5:1, respectively, for the arguments regarding Romans 1:3, Galatians 4:4, and 1 Corinthians 9:5/Galatians 1:19, would appear to be a more-than-reasonable numerical summary of your own hermeneutical argument and the great weight that you attach to it. (I.e., if one were to agree with all your words and only disagree with your attempt to express it in numbers.)
I do not want to haggle with your numbers, but I think 1/1 (for "Made from a woman") is unjustified (even Carrier had 2/1 maximum and 1/1 minimum). Maybe it is a typo and you meant 10/1.
Of course that'd still tilt the total towards historicity, just not in quite such an "astounding" fashion. With only these substitutions, one would arrive at P(h|e,b) of 97.9% and not 99.996% ... and, in fact, don't you think a reasonable person, even a reasonable historicist (since we are using historicist-argument numbers), would consider the former's 2.1% chance of the non-existence of Jesus a somewhat better approximation than the less than 1 in 20,000 implied by the latter?
This 97.9% for average (between maximum and minimum), I could tolerate. But my maximum is still 99.999999%, but because of the nature of the evidence, and the fact the events happened almost 20 centuries ago, there are some doubts (maybe because I have been partly brainwashed by this avalanche of mythicist arguments (even if I find them weak, spotty, indirect, remote, far-fetched and many times ridiculous), I would allow (grudgingly) a minimum of 95% (mostly because of the very small chance, in my mind, Paul & his epistles might be a 2nd century clever fabrication. But I have strong arguments against that also).

Cordially, Bernard
Last edited by Bernard Muller on Fri Mar 20, 2015 8:47 am, edited 1 time in total.
I believe freedom of expression should not be curtailed
User avatar
Peter Kirby
Site Admin
Posts: 8048
Joined: Fri Oct 04, 2013 2:13 pm
Location: Santa Clara
Contact:

Re: Carrier's numbers and math in OHJ

Post by Peter Kirby »

Bernard Muller wrote:
Peter Kirby wrote:Incidentally, a problem with Carrier's "naive" application of Bayes' theorem--a genuine problem, which I've discussed, unlike some of the pseudo-problems that have been proposed in this thread--becomes even more apparent when we use the extreme 50:1, 20:1, and 20:1 numbers that are proposed.
Yes, you said it.
Bernard Muller wrote:
That problem is the assumption of conditional independence. I don't attempt to explain it in a way that you understand (not sure how I'd do that, actually), but it is a problem. Basically speaking, what you really want isn't P(e1|h)P(e2|h), when computing P(e1, e2 | h), as that assumes conditional independence. What you really want is P(e1|h)P(e2|h,e1). And that does make a difference.
I want to believe you, but it is very fuzzy in my mind. Why don't you do use your equation for at least one of Carrier's table Such as for EPISTLES. BTW, I do not think my numbers are extreme after considering Carrier's counter arguments. I refrained to write 1000/1 for "made from sperm", which is what I would prefer as a maximum. However, I read much better arguments against "seed of David" but my blog post is only about Carrier's OHJ, not about other mythicist's books.
I think it's better if I start with an example from any other domain.

Actually I have one ready already. Posted to this forum in September of 2014.

http://www.earlywritings.com/forum/view ... 407#p19407
There is a serious wrinkle in the use of Bayes' theorem as applied by Carrier. It is the same wrinkle someone once pointed out as an objection to my essay linked above. It is as follows:

Let's stipulate some made-up numbers:

In a year a country had domestic and foreign vehicle sales. All sales were truck or sedan, and had either a 4-cylinder or 6-cylinder engine. Breakdown:

15% were 6-cylinder domestic trucks.
5% were 4-cylinder domestic trucks.
5% were 6-cylinder foreign trucks.
10% were 4-cylinder foreign trucks.
25% were 6-cylinder domestic sedans.
5% were 4-cylinder domestic sedans.
5% were 6-cylinder foreign sedans.
30% were 4-cylinder foreign sedans.

Now let's use Carrier's Bayesian method and tell him that a vehicle bought this year had 6 cylinders and is a truck. What is the probability that it is domestic?

After assigning a 50/50 prior probability (half of sales are domestic), Carrier starts by looking at 6-cylinder vehicles. 80% of 6-cylinder vehicles were domestic, while 20% were foreign. The updated probability becomes an 80% chance it is domestic. Also, 62.5% of trucks were domestic, while 37.5% of trucks were foreign. Carrier thus arrives at an 87% chance that a 6-cylinder truck is domestic.

However, if you look at 6-cylinder truck sales, 75% were domestic and 25% were foreign. Thus the actual probability is 75% that it is domestic. What went wrong?

Carrier's application of Bayes' theorem doesn't take into account whether the terms/variables he uses are independent. When we speak of independence, we mean (for example) that a vehicle in general that is 6-cylinder is just as likely to be domestic as a truck that is 6-cylinder is likely to be domestic... but that isn't true. These aren't independent variables. And treating them as independent variables leads you to a greater confidence level than is actually justified.

I'm not sure how one would go about ensuring that the variables used by Carrier are independent, and I doubt that it can be done. This would be fine, perhaps, if Carrier admitted that his application of probability theory was merely heuristic and that he doesn't view it as anything other than a blunt tool for making guesstimates by stringing together intuitions regarding various bits of evidence. It's not completely clear that this is his stance on it (even though it seems to me the most plausible - and certainly the most defensible - interpretation of Carrier).
In this particular case (or so the argument goes--one can disagree with the particular application here), every Paul who writes about the "seed of David" is also completely likely (or at least much more likely) to write about "born of a woman." Perhaps one colloquial way of understanding this is to consider it a form of cheating, or double-counting the evidence. Sort of like it would be a form of double-counting if you assigned a 20:1 to the 1 Cor 9:5 passage and than another 20:1 to the Gal 1:19 passage. Either assignment could be correct, but when using them together, you are forced to use one of them as 20:1 and the other as 1:1 (or something like that), because of the fact that conditional independence is false.

Any number of sources will talk about the relationship between conditional independence and the product rule (or "chain rule"):

http://en.wikipedia.org/wiki/Conditional_independence
http://en.wikipedia.org/wiki/Chain_rule ... ability%29
http://www.eecs.qmul.ac.uk/~norman/BBNs/Chain_rule.htm
http://www.cs.uni.edu/~campbell/stat/prob4.html

Here is an instance of the "chain rule" (the general form of the product rule, with multiple terms). Notice how the terms after the "|" symbol pile up as the chain grows. This is what I'm talking about. Carrier assumes (or seems to assume) that he can ignore this general form and use the assumption of conditional independence to simplify things down to the very simple P(e1|h) times P(e2|h), etc., but that assumption (at least for some of his numbers) would have to be considered false.

Image

Image
Peter Kirby wrote:What is the difference? We're not really supposed to be asking just "what is the probability that Galatians 4:4 would have 'born of a woman', on h and on ~h" but rather "what is the probability that Galatians 4:4 would have 'born of a woman', on h and on ~h and with all previous items combined," including the previous item that Romans 1:3 has the 'seed of David' bit. I'm pretty sure your argument is that the seed of David bit would also imply the born of woman bit. Therefore you'd have to place P(e2|h,e1) and P(e2|~h,e1) as both being very high, because it's a conditional probability under the 'seed of David' bit as well and not just under the hypothesis or its negation.
Bernard Muller wrote:
Peter Kirby wrote:With this in mind, and even accepting the full strength of your exegetical argument, the substitution of 10:1, 1:1 (yes, this implies that Carrier is wrong here, in that he doesn't consider the problems with the assumption of conditional independence), and 5:1, respectively, for the arguments regarding Romans 1:3, Galatians 4:4, and 1 Corinthians 9:5/Galatians 1:19, would appear to be a more-than-reasonable numerical summary of your own hermeneutical argument and the great weight that you attach to it. (I.e., if one were to agree with all your words and only disagree with your attempt to express it in numbers.)
I do not want to haggle with your numbers, but I think 1/1 (for "Made from a woman") is unjustified (even Carrier had 2/1 maximum and 1/1 minimum). Maybe it is a typo and you meant 10/1.
It is completely intentional and is basically the point of most of my previous post, criticizing the "naive" application of Bayes by Carrier and its assumption of conditional independence (which means, like I said, that Carrier's numbers are in a real sense wrong, also).

And it really depends on whether you come to the evidence of the 'born of woman' first or the evidence of the 'seed of David' first. If you come to the evidence of 'seed of David' first, then you should assign something like 10:1 to the 'seed of David' and then 1:1 to the 'born of woman.' But if you come to the evidence of the 'born of woman' first, then you should assign something like 5:1 to 'born of woman' and then 2:1 to the 'seed of David.' In each case, the result should be the same.

And, in each case, you need to be careful to account for the falsehood of the assumption of conditional independence by considering the likelihood of each additional piece of evidence, not only in the light of the hypothesis and its negation, but also in the light of all the evidence previously considered.

I apologize that I don't know of a very simple, clear, and convincing way to express this in non-mathematical terms.

Carrier neglects the issue in On the Historicity of Jesus. I don't know if he brings it up in Proving History. But it certainly throws a spanner in the works, not least because the math becomes a bit harder (mathematicians call it the " curse of dimensionality" in the context of Bayesian probability when they have to keep adding additional things to the part after the "|" thus creating more "dimensions" to the analysis) but also because our confidence in our intuitions becomes a bit weaker (due to the twistiness of thinking about things exactly this way). Certainly some people who were ready to sign up for the "naive" form of the application of Bayes might at this point find it a "hard saying" that they should further have to consider the interactions caused be the falsehood of conditional independence, and thus part ways with Carrier at that point.
"... almost every critical biblical position was earlier advanced by skeptics." - Raymond Brown
User avatar
maryhelena
Posts: 2902
Joined: Tue Oct 08, 2013 11:22 pm
Location: England

Re: Carrier's numbers and math in OHJ

Post by maryhelena »

Peter Kirby wrote:Earl Doherty's November 3, 2000 "inaugural address" to the Jesus Mysteries group ("A Welcome Forum") actually touches on both issues of the uncertainty involved in historical assessment and the multiplicity of non-historicity hypotheses, which are now exercising us here in this thread fifteen years later.
Earl Doherty wrote:I for one will be willing to consider and acknowledge any good argument for
the existence of Jesus. A given argument by itself does not necessarily
prove the case either way, and one does not need to score all the points
for one's own side. Nothing in historical research can achieve that kind of
blanket or mathematical certainty. We can only aim for the balance of
probability, based on the evidence. I sometimes say that my viewpoint needs
only to persuade, not prove; to commend to the reasonable, unprejudiced
person not locked into rigid confessional interests, that the evidence of
the early Christian record, together with elements of the wider ancient
picture, strongly indicates that there was no historical Jesus. To produce
that 'conviction of probability' is all one can hope to achieve, or all one
*needs* to.

Even the mythicist position is not monolithic. Our bottom line may be that
the figure portrayed in the Gospels is fictional, that early Christian
prophets like Paul were not preaching a recent man who was a teacher and
miracle worker in Galilee. Beyond that, there is room for maneuver. Most
people interested in this field know that researchers like G. A. Wells and
Alvar Ellegard have concluded that early Christians believed their god-man
Jesus had lived on earth at a more distant time in the past (Ellegard that
he was an actual man known to the Essenes as the Teacher of Righteousness,
Wells that he was only an imagined historical figure who had lived
obscurely). Wells has also recently allowed for the possible existence of a
teaching sage at the beginnings of the first-century Kingdom of God
movement in Galilee. I and other mythicists disagree with such
interpretations. But these variations are relatively minor. They don't
change the fact that the Gospel presentation and its central character are
basically fictional and symbolic (its spirit and details derived from
scripture and wider sources),
and that Paul's faith, with its heavenly
Christ and system of salvation very similar to the Greek mystery cult
ethos, has nothing to do with a recent or actual historical man.
my formatting

1) The gospel 'central character' being 'basically fictional and symbolic' does not rule out historical figures as having been relevant to the creation of that gospel figure.

2) That Paul's faith might have nothing to do with a historical man does not rule out historical figures being relevant to the gospel figure.

Yes indeed, as Doherty said - ''there is room for maneuver''. Sadly, Carrier, with his theory of what minimal mythicism is - that Jesus began as a celestial being - has allowed himself no such 'room for maneuver'. Rather than advance the ahistoricist/mythicist position, Carrier's MM and his BT approach has placed his theory, on the historicity question, into an intellectual straitjacket.

Probability theory might have its uses - but luck and the idiosyncrasies of life have as big a role to play in history as they do in our daily lives. The outlier or random effect can throw the mightiest of mathmatical probabilities - as anyone working in economic markets knows only too well. Thus, whatever the odds are, that are stacked against a historical Jesus, these odds could not rule out the possibility that such a figure existed historically. A historical Jesus just could be that outlier.

Carrier, it seems, wants us to place our bets on the question of the historicity of Jesus, on his number system. Sure, playing the numbers has put man on the moon - but playing the numbers game also dropped an Atom bomb. i.e. there are some things that numbers are not good for. Turning the Jesus historicity question into a mathmathical question could just be the equivalent of dropping an intellectual bomb on the ahistoricist/mythicist position. :eek:

Oh, well - maybe it just had to happen....

But maybe, as Hiroshima arose from the ashes of the Atom bomb - the ahistoricist/mythicst position will arise from the wreckage that Carrier' intellectual Atom bomb has inflicted upon it. :)
Tread softly because you tread on my dreams.
W.B. Yeats
Bernard Muller
Posts: 3964
Joined: Tue Oct 15, 2013 6:02 pm
Contact:

Re: Carrier's numbers and math in OHJ

Post by Bernard Muller »

to Peter,
Sort of like it would be a form of double-counting if you assigned a 20:1 to the 1 Cor 9:5 passage and than another 20:1 to the Gal 1:19 passage. Either assignment could be correct, but when using them together, you are forced to use one of them as 20:1 and the other as 1:1 (or something like that), because of the fact that conditional independence is false.
First, Carrier combined 1 Cor 9:5 with Gal 1:19, in his component for EPISTLES "brothers of the Lord" (page 594). I followed him on that. I certainly do not have 20/1 for 1 Cor 9:5 AND 20/1 for Gal 1:19. I would not do that. But I do not see any connection between your example ("the brothers of the Lord" & "the brother of the Lord") and "seed of David" & "born of woman", which are not related in my view. Why would a 20/1 odd for the first one require a 1/1 odd for the second one? that's beyond my understanding.
And it really depends on whether you come to the evidence of the 'born of woman' first or the evidence of the 'seed of David' first. If you come to the evidence of 'seed of David' first, then you should assign something like 10:1 to the 'seed of David' and then 1:1 to the 'born of woman.' But if you come to the evidence of the 'born of woman' first, then you should assign something like 5:1 to 'born of woman' and then 2:1 to the 'seed of David.' In each case, the result should be the same.
I made my arguments about "seed of David" and "born of a woman" being in my favor totally independently of each other. I made arguments on these two components against the ones of Carrier totally independently of each other. Carrier made also his arguments independently of each other.
I used corroborations in my previous posts as just a bonus.
Neither Carrier nor I said "if seed of David" implies that (ahistoricity or historicity), therefore "born of a woman" implies that also (ahistoricity or historicity).
And, in each case, you need to be careful to account for the falsehood of the assumption of conditional independence by considering the likelihood of each additional piece of evidence, not only in the light of the hypothesis and its negation, but also in the light of all the evidence previously considered.
I do not know how that would work. Why not consider the evidence independently first for each components and then, if you think the results are in your favor, use them collectively as strengthening corroborating evidence for components already appraised independently? That's what I have been doing, and so Carrier at times.
Certainly some people who were ready to sign up for the "naive" form of the application of Bayes might at this point find it a "hard saying" that they should further have to consider the interactions caused be the falsehood of conditional independence, and thus part ways with Carrier at that point.


If you think Carrier's math requires correction, and you know the appropriate solution, why don't you explain it with an example based on one of Carrier's table?

Cordially, Bernard
Last edited by Bernard Muller on Fri Mar 20, 2015 8:38 am, edited 1 time in total.
I believe freedom of expression should not be curtailed
User avatar
Peter Kirby
Site Admin
Posts: 8048
Joined: Fri Oct 04, 2013 2:13 pm
Location: Santa Clara
Contact:

Re: Carrier's numbers and math in OHJ

Post by Peter Kirby »

I don't have time to make another go of it. Either you understand, someone else helps you understand, or you don't understand. Either way is fine with me.
"... almost every critical biblical position was earlier advanced by skeptics." - Raymond Brown
Post Reply