Back in 2007 there was a documentary and a great deal of subsequent discussion about the so-called Jesus Tomb at Talpiot. I wrote a few posts on the claims here, as well as a couple at ReJesus, but Mark Goodacre provided the most coverage.I was reminded of these discussions today when listening to the podcast for the BBC Radio 4 maths show "More or Less". Towards the end they discuss the probability of five people meeting whose fathers all had the first names John Charles. The initial calculation comes out to be one in several billion billions, but what's significant in this case is that this is not theoretical, "but" as the narrator says in Magnolia "it did happen".
The team however quickly whittle down this astounding statistic down to something much more reasonable, and you can hear their reasoning about 23 minutes into the podcast (actual file here).
A number of these reasons are also relevant to the Talpiot Tomb question. Firstly it was an actual discovery, so that changes the statistical calculations altogether, secondly the location appears significant, but various other locations would have given rise to a similarly apparent significance. Thirdly, the smaller and smaller the probabilities get the more likely it that a reality blip changes everything.
Coincidentally I was also musing on a related point again this weekend, how we tend to find names cluster together rather than occur at random. I once commented on Mark Goodacre's blog that a modern day example might be the cluster of names Seumas, Mary and Patrick. Individually the probability wouldn't be that high, purely on the basis of their popularity in the population as a whole. But in reality because they are all Catholic names the likelihood of finding such a cluster would be much higher than this simple basis for the calculation. If you searched for it in a Catholic part of Belfast you'd get a much higher number of families than if you searched in Kent, or a Protestant part of Belfast. Given how sectarian Judaism was at the time, it's reasonable to want to know about these names relate to each other before assuming the probabilities are all independent.
Which leads me onto another question. During the Radio 4 podcast the expert says that looking at the 40s and 50s he had difficulty finding "the distribution" even though they had the rankings. So if we're lacking this key piece of data for just 60 years ago, how accurate is the data that was used to calculate the probability regarding the Jesus Tomb? If I remember rightly, the overall figure was calculated my multiplying the assumed probability for each name individually. Now the probability for each name was drawn from other ossuaries found in the region from around the same period of time. The problem with this is that it's not representative of the whole, at best its representative of those rich enough to have a bone box. But Jesus and his family were not rich. Were this to be their tomb then it would only exist because Jesus' life had elected their status. We have no reliable information of the distribution and occurrence of names of people within Jesus' social class and so this is another flaw (amongst many) that the programme makes.
Let us not forget that anything with Pellegrino's name attached to it might as well be consigned to the dustbin, seeing as he has been shown to be nothing more than a liar and a fraud e.g.:
ReplyDeletePellegrino does not have PhD
Hello Matt!
ReplyDeleteYou wrote: "But Jesus and his family were not rich."
[To differentiate], In fact Rabbi Y'hoshua could very well have had much money. Success is promised in Torah as a blessing for Torah-observers. It is a mitzwah [directive] in Torah to give tzedaqah, which requires money. The 'gospels' are not reliable: Documentation
When it comes to statistics:
"“In a peer-reviewed article published last month in the prestigious Annals of Applied Statistics, Andrey Feuerverger places the odds of the 2,000-year-old tomb not belonging to the J*esusfamily at 1 in 1,600.”
“This figure is even more bullish than the 1-in-600 figure that Dr. Feuerverger calculated a year ago…” (Toronto Globe and Mail, 2008.04.22).
This is much more realistic than his first figure. Still, even this figure cannot have taken into account the Ya·aqov ossuary—which raised the original probability to about 1:30,000. Predicated on his newest calculation, that probability would correspondingly rise to 1:80,000 (1600÷600*30,000)—families. Yet, most scholars would put the total number of families in 1st-century Yәrushâlayim at less than 8,400 families. (The consensus of scholars is that Josephus and Tacitus exaggerated the numbers; that, based on water supplies, etc., the population was more likely around 50,000. Families were typically larger then. Assuming 6 persons/family yields a population slightly under 8,400 families.) Thus, it would still be unlikely to find another such tomb even in ten cities about the size of 1st-century Yәrushâlayim!!!
Beyond this, finding two families that each have these six names doesn’t entirely solve the problem. While the statisticians’ methods are correct, their premises are sometimes conservative to the point of inadequate. Even limiting possible names to Bauckham’s catalog is like claiming that if a name isn’t included in the membership list of a certain church then it’s impossible that a person by that name exists in that city at that time. Three generations of a typical family, at a typical rate of 6 children each and including wives, would have totaled around 100 members. What are the odds that, of those 100 family members (who could possibly each have had a unique name), the discovered inscribed ossuaries would be these six names instead of one of the other1,192,052,400 permutations of 6 names from the pool of names of 100 family members? Statisticians have factored in a very conservative set to represent perumutations that could have included (among the six ossuary inscriptions) some different family member or members (e.g., Daniel or a Tsiporah or a Devorah, etc.)—even though all six names were in the family. That name, or names—from a family having relatives with these six names—would have conflicted with the family of Ribi Yәhoshua." [Full article here: Link]
Anders Branderud
Thanks Anders. I don't have time to read that right now, but I'll hopefully add some comments here if I get a chance. Thanks again for an enlightening reply.
ReplyDeleteMatt
Thanks for the enjoyable post, Matt. I heard that on More or Less too, with a short follow-up the following week.
ReplyDeleteThe John Charleses would have been less interesting if in fact one was Jonathan Charles and another Jonathan Carlos. One of the things that was really troubling about the Talpiot Tomb case was the number of non-matches.