Response to Tabor’s Remarks on Statistics – March 26 (expanded March 26 and 28)

I now look at the Tabor entry on the issue of the statistics, which bears much of the weight of the hypothesis:

________________
Statistics: Although the names are “common” as is so often pointed out by so many, it does indeed seem to be the case that the statistical grouping of these particular names in this particular tomb is far from common. This is confirmed by the mathematical probabilities based on name frequencies, or more directly, by looking at the names in tomb after tomb of which we have record. Nothing like this occurs anywhere else. This is, after all, the only tomb known with a “Yeshua son of Joseph.” Even if the probabilities were 50/50 the tomb would be of great interest and worth examining in this regard. As it stands they are surely much higher than that. Statisticians often point out that “common sense” when it comes to probability theory, is often quite misleading. What we have to ask is what are the probabilities of these six names occurring together in a 1st century Jewish family tomb, namely: Mary, a second Mary, Jesus son of Joseph, Jude son of Jesus, Joseph, and Matthew. I have independently consulted with several statisticians who work with demographics and probability theory with the following results.

I now look at the Tabor entry on the issue of the statistics, which bears much of the weight of the hypothesis:

________________
Statistics: Although the names are “common” as is so often pointed out by so many, it does indeed seem to be the case that the statistical grouping of these particular names in this particular tomb is far from common. This is confirmed by the mathematical probabilities based on name frequencies, or more directly, by looking at the names in tomb after tomb of which we have record. Nothing like this occurs anywhere else. This is, after all, the only tomb known with a “Yeshua son of Joseph.” Even if the probabilities were 50/50 the tomb would be of great interest and worth examining in this regard. As it stands they are surely much higher than that. Statisticians often point out that “common sense” when it comes to probability theory, is often quite misleading. What we have to ask is what are the probabilities of these six names occurring together in a 1st century Jewish family tomb, namely: Mary, a second Mary, Jesus son of Joseph, Jude son of Jesus, Joseph, and Matthew. I have independently consulted with several statisticians who work with demographics and probability theory with the following results.

Assuming a family size of six the probability of these six names in these relationships occurring together in one family is 1/253,403.Therefore, out of 253,403 families (a population of 1,520,418), this particular combination of names would occur only once. Obviously the population of late 2nd Temple Jerusalem was nothing of that sort, but perhaps only 25,000 (Jeremias) to 75,000 (high estimate). These numbers are based on good data of name frequencies from inscriptional and literary sources of the period. They are also conservative, in that I asked my consultants to use the generic form of the names (Mary, Joseph, Jesus, etc.), not the much more rare specific forms on the ossuaries themselves: Mariamene, Maria, Jose, and Yeshua. To further illustrate the ways in which a cluster of “common” names ends up being rare or unique, here is an example. Imagine a football stadium filled with 50,000 people—men, women, and children. This is an average estimate of the population of ancient Jerusalem in the time of Jesus. If we ask all the males named Jesus to stand, based on the frequency of that name, we would expect 2,796 to rise. If we then ask all those with a father named Joseph to remain standing there would only be 351 left. If we further reduce this group by asking only those with a mother named Mary to remain standing we would get down to only 173. If we then ask those of this group with a brother named Joseph to remain standing only 23 are left. And finally, if we add the condition of just one brother named James, there’s less than a 3/4 chance that even 1 person remains standing.

_____________

My Comments: This argument, simply put, is stacked. Here is why.

(1) There is one name that has no evidence of being family, that is, Matthew. The claim is it "family related" only serves to show how the evidence is being handled in a very open ended way. Once we allow for the possibility of names outside the family to impact the site, we also have opened the door for non-family realities to impact the analysis. We, however, also undercut the claim this is really a family tomb. One cannot have it both ways. Once we ceaase to have a family tomb, then how do we analyze the names? If this is a family tomb, then what is Matthew doing in it. Either way, there is a problem with the identification.

(2) There has been extensive discussion about the Mariamne-Mary Magdalene name connection, with it being brought into great doubt by all the name experts, including François Bovon, who was the person the special interviewed to discuss the link. If this name goes out, we lose several features, namely, the idea of a marriage and then where does the son of Jesus comes from. This Mary, then becomes another may among many with tons of women in the period who have this name.

(3) Finally, there still is an appeal to James to get the figure down to only one in the example given above. This makes a double appeal that is unlikely. This identification requires that the James ossuary not only be authentic, which though possible is debated; but that it be identified with the now much discussed tenth ossuary of this tomb. This is an identification for which there is no clear evidence (as the dimensions do not match and the tenth ossuary of the tomb was described as "plain" [not inscribed] when it was catalogued and reported on (Something Amos Kloner has confirmed to me personally).

Any of these points raises severe questions about the hypothesis. Any two of them together, when each of them is likely, negate it entirely.

Now we are down to four names: Jesus, Joseph, Mary and Jose. All of a sudden our statistics do not look so compelling, given that Mary is the most popular of names among women (21 percent of the total) and Joseph and variations of it are the second most popular male name (just short of 10 percent) with Jesus being the sixth most popular name of the period. Given that we have at least thirteen people in the tomb (given the bones found outside the ossuaries) and possibly more (if ossuaries have more than one person in them — possible, but not hard evidence, see next post), the odds are good that all we may well have is a distribution of common names. What we really have are just three different names that match Jesus’ family list we possess: Mary, Joseph variants, and Jesus (given the repetition of Joseph and Jose, variations of the same name).

All of a sudden the statistical argument does not look so strong. I actually worked through this example argument in one of my earlier posts (see More stats for you all March 12 (Updated March 15)), but now I explain why we never get to the point of working through such a scenario.