Last week I published a critique of the rather sensationalized claims of the Discovery site about the dating of the crucifixion (see also here, here, here). Since then Jeff Williams, one of the authors of the study upon which the sensational article was written, has contacted me in an attempt to clarify some of the issues. Jeff was apologetic towards the sensationalism surrounding his paper and assured me that he been in touch with the author of the original piece and their editor and a corrected story will be published soon.
He also generously sent along his original paper so I could read and comment on it. Overall, the report is very technical. The first few pages have nothing to do with the crucifixion, instead the focus is on the methods and previous studies in the region. I won’t comment on these sections because I don’t know enough about them to comment, suffice to say that I have no direct reason to be skeptical of the information. However since most of their sources and reference material are behind pay walls, it makes it difficult to determine exactly what their source material stated. This is problematic, for when I see a statement like ’28 historically documented earthquakes’, I can not be certain if ‘historically documented’ means ‘from the written historical record’ or if they mean more broadly, that is, from the geological record. If they mean ‘from the written historical record’, then there may be implications to the discovery which may invalidate it or at least draw additional caution, since the written accounts we have from the past were not restricted to telling ‘what happened’. That said, there are some issues in the paper that do need to be addressed, primarily with the discussion of the earlier paper by Humphreys and Waddington.
First, I’m concerned that the use of various texts, both vital to the conclusions of the paper and to the dating process throughout, seem so uncritical without any sort of discussion with the problems of reception–problems which are so demonstrably important to their arguments. For example, the reference in Josephus, Wars 1.19.370 (also discussed in Antiquities 15.121), to an earthquake in 31 BCE (discussed on pp. 3-4 of the paper) is a great deal different than the two earthquakes discussed in Matthew at the crucifixion and resurrection. While Josephus exaggerates a real event (in one account he says 10,000 men perished, in another 30,000 perished) and uses it to make a political point (i.e., about the enemies of the Jews in the region), it is clear that within the context the earthquake has some significance. But the same cannot be said for Matthew’s account. And the way with which the accounts appear, sans any discussion on how Biblical authors wrote (imitating earlier texts to express theological messages in new ways) and how these might be different than, say, an ancient historian (without ever addressing the issues with ancient historians) seems a little irresponsible. One may argue that the purpose of the study was not about the value of ancient texts, but they would be wrong since the value of the “primary source” used (Matthew) is presumed in order to reach the conclusion (that the crucifixion happened on a specific date).
When the time comes to discuss the findings of Humphreys and Waddington, whatever textual evidence they might have had for the earthquake is abused. Aside form the the fact that the article goes into the criterion of multiple attestation (i.e., that multiple Gospels generally agree), the authors (and conversely, Humphreys and Waddington) don’t use the criterion correctly. They miss the vital fact that only Matthew recounts an earthquake (and I’ve already laid out my thoughts on one of its possible origins) and Mark, our earliest Gospel, is silent. Luke, who was aware of Matthew, doesn’t mention an earthquake, and neither does John. So in this instance the criterion of multiple attestation fails, since there aren’t multiple accounts–and even if there were multiple accounts, none of these would be considered ‘independent’ witnesses since the later Gospels utilized the earlier ones to formulate their narratives and narrative structures.
In addition to this, as Mark has already pointed out, Matthew–the Gospel which mentions earthquakes–does not date the crucifixion to the 14 Nisan, but the 15th (on Passover). John is the only canonical Gospel which suggests that Jesus is crucified on 14 Nissan, and he does not recount an earthquake. So how anyone has come to the proposed dates of Friday 7 April 30 CE (14 Nisan) or Friday 3 April 33 CE (14 Nisan) is beyond me (note here that these are not Jeff’s estimates, but those of Humphreys and Waddington), since that would mean that Matthew is not the source of the account by John, but he is not a primary source for an earthquake.
Finally, the use of Acts of the Apostles is problematic for the same reasons that Matthew’s gospel is a problem. Acts is possibly much later than Matthew (with a new date suggested around the early second century, a la Joe Tyson) and highly influenced by different theological agendas than Matthew would have been. Still, even if Acts could be authoratatively dated to the first century, the earthquakes in Acts like those in Matthew are theological representations and not historical ones.
These criticisms aside, I am interested in the general conclusions of the piece. Jeff has made it clear that there are three possibilities.
- the earthquake described in the Gospel of Matthew occurred more or less as reported;
- the earthquake described in the Gospel of Mathew was in effect ‘borrowed’ from an earthquake that occurred sometime before or after the crucifixion, but during the reign of Pontius Pilate;
- the earthquake described in the Gospel of Matthew is allegorical fiction and the 26–36 AD seismite was caused by an earthquake that is not reported in the currently extant historical record.
Of course I am more partial to the third option and I have to imagine that the consensus would fall into that third option as well. Even the second option would not be out of the question, though I would remove the caveat that it had to have been ‘during the reign of Pontius Pilate’ since, without the crucifixion, the earthquake would have no relevance: the two events are too interconnected in the narrative to remove one or the other.
Also, interestingly enough, this paper is generally contrary to the article written by Discovery, which is refreshing. I’m not convinced by the arguments overall. I am not sure why one even needs to include the crucifixion as it is neither relevant to this sort of study or valuable as a tool to date earthquakes. It just seems rather silly and contrived. Still, I thank Jeff for clarifying his position with me and sharing his paper. It was much less incautious than I was led to believe (via news media reporting on his story) and overall I found it interesting, even if I didn’t find it all that compelling.