Dear Chris,
You say:
"Recalculated how? Do you have access to the IntCal09 program?"It's not so difficult. You can freely download both Oxcal and Intcal calculators from the internet. The Oxcal program is not as intuïtive as the Intcal program. You can freely download the latest version of Calib (Calib601, exactly the one they or their references have been using and refering to as "Calib-Rev. 6.0.1", and from which my data also come from) which uses the latest IntCal09 data as database, from here:
http://calib.qub.ac.uk/calib/I have checked the data of table 1 of the article, and it is clear that they are close enough to the Intcal data, but they are all rounded to the nearest tenth. Rounding numbers is one thing one should NOT yet do in such an early stage. Anyhow, they have been selective in some cases in choosing the date range that suited their purpose best.
In their article they claim:
"The samples pooled in the matrix (Table 1) are statistically the same at the 95% confidence level using a Chi-square (χ2) test (sample key 7, where T = 6.09<14.1). The weighted average date (2962±14 14C yr BP) gives a 1 sigma (σ) calibrated age range of 1215–1190 BC with 34.3% relative probability and another age range of 1180–1160 BC with 26% relative probability, using Calib-Rev. 6.0.1 [22] and Oxcal 4.1 [23] with IntCal09. Calibrated age ranges in details at 60.3% of the 100% dating probability (Fig. 4). Hence, there are two chronological possibilities for the calibrated date of the destruction Level 7A, between the end of the 13th century and the beginning of the 12th century BC or the first half of the 12th century BC. By contrasting historical-archaeological and radiocarbon-based data sets, the best candidate for the destruction date of the harbour town is the Sea People invasion."I note the following:
1. Since they give us only rounded figures in their table 1, on which they say above that they have mixed Oxcal with Intcal considerations (not specified in their table 1), we cannot even check their so-called "weighted average" which is magically given as a precise (!) figure!!!! We are forced to recalculate all the RC data ourselves in both Oxcal 4.1 and in Intcal09 separately, in order to find out what they have been up to. Nobody is going to go into that much trouble in order to check on their data, especially not with the out-dated Oxcal, which also has various specific options in the way it is calculated, but with Intcal it's peanuts. So far the data actually follow the IntCal ranges, but there are some inconsistencies in their choice of date range, as I've said before. Anyhow, there is no way you can mix Oxcal with Intcal data in any scientific way, and if you do, you still have to give us both the Oxcal and Intcal data seperately, as well as your method of mixing/combining/averaging.
2. There is no certainty at all that the destruction layer was indeed caused by the Sea Peoples. There is also no certainty at all as to whether this destruction occurred before or after the battles with either Merenptah and/or Ramesses III. But since the authors already assume a low chronology, they also seem to assume à priori that the destruction occurred (shortly) before the battles with Ramesses III.
Their article further on says:
"The cuneiform tablet RS 86.2230 [29], sent by the Egyptian Bay (1194–1190 BC), the Great Chancellor of the Pharaoh Siptah-Queen Tawsret, to the last king of Ugarit provides the final evidence of a living kingdom at the dawn of the 12th century BC. A document from Egypt relating the execution of Bay as a traitor in Siptah's regnal year 5, states that the cuneiform tablet RS 86.2230 must have been written before an historical date of 1190 BC [30]."They really assume too much here as to the minimum date of Bay's letters. It does not prove their dating scheme to be correct.
You say:
You are reading into the article something that isn't there. They do not describe the calibrated dates as weighted. And here is where they say they have "refined" their weighted "radiocarbon" results with both "archaeological", historical and astronomical considerations:
The radiocarbon results, in relation to archaeological and historical data, lead us to propose a date of 1194–1190 BC for the Sea People event in the northern Levant. This radiocarbon-based archaeological date can be refined with the astronomical observation related on the cuneiform tablet KTU 1.78 (RS 12.061) found among the ruins of Ugarit. The sun eclipse depicted on the cuneiform tablet KTU 1.78 was dated to the 21 January 1192 BC [31], suggesting that the destruction of the city and the fall of the kingdom occurred after this observation. Before you know it they make you believe that this combined and "refined" date is the radiocarbon date-range itself, and they consistently present it themselves as such:
The date of 1192–1190 BC for the Sea People invasions in the northern Levant, and the end of the ancient Eastern Mediterranean world fits the radiocarbon, historical, archaeological and astronomical data.They do not fit at all! The radio-cabon date ranges point to a higher date instead. Their "refinement" is already a farce since there is no such thing as an "archaeological" date, because all archaeological dates are based on certain presumed historical dates. Astronomical dates too not so certain but in any case cannot be used to "refine" an RC-date. Yet the authors claim to have done just that.
You say:
To accuse the authors of weighting their data by non-RC factors when all they did was to apply ordinary arithmetic to RC dates is excessive and undermines the credibility of your critique.As you can see, the authors overrule their own RC data by adding into the mix the historical and astronomical data they believe in and then go on from there, still calling their dates "radio-carbon" data, just as proponents of the Bayesian formula do. That is what I cally "trickery". It undermines the credibility of the article. Of course we can still call such dates "radio-carbon
based", but only until so far, and not make them sound like radioin-carbon dates themselves, only to make them sound more solid than they are. It is certainly tricky not to fall for the trap.
...smacks of Luddismmmm, Luddism seems to have been an honest massive uprise to preserve the jobs of professional textile handworkers and to oppose to automated machinery that could be run by any uneducated person:
http://en.wikipedia.org/wiki/Luddite. :)
Getting back to the paper, my impressions when I read it was that the application of KTU 1.78 and a low Egyptian chronology are at least naive, and that seems to be the consensus here. I agree.
But on the RC side I think it would be more productive to try to arrive at a clear understanding of how the authors actually calibrated the dates than to accuse them of deliberately fudging the results.How the authors calibrated the dates, is already clear. The dates in the table are in fact Intcal dates, rounded to the nearest 0, and in one case the nearest 5, and in a few cases I detected selectiveness. Apparently they meant with Oxcal that they had compared it with the usually slightly higher dates of Oxcal, and since higher dates than the intcal dates would not have served their low-dated conviction they have simply dropped Oxcal altogether.
It would be more productive though, to check out the origin of those selectively chosen RC dates, and which dates they have discarded, since the whole table of RC dates is unnaturally smooth in a way you would never see in any other table of RC dates and that is why I suspect them of giving us a plain false total picture of RC dates. You are right that I cannot accuse them directly of falsification, since I don't have access to their background notes and lab-results, but their table 1 I cannot call trustworthy to either to say the least.
Regards,
Ian