> So I am curious as to why, on what grounds you reject the C-14
> dating projects being valid evidence for the pyramids being
> built during the third and fourth dynasties?
I have no problem with C-14 dating technology, but I have a problem with how it's used, the presumptions that are made, and the conclusions drawn from it.
The authors make a strange point in the Introduction. They state that early C-14 studies arrived at a timeline that significantly contradicted the traditionally held "historic chronology". Oddly, the authors believe the fault was in the C-14 studies and not the historic chronology. No explanation was offered to explain why the authors did not allow for the possibility that the historic chronology may be at fault and not the C-14 studies.
Note that this study was funded by David Koch, of the Koch Brothers; He provides major funding for every one of Lehner's AERA project. In such a situation the authors need to make a statement of full disclosure announcing whether they have any conflicting extracurricular interests.
I am not as interested in the mudbrick samples. It's when we get to the stone work that things get murky. The investigators acquired 353 samples in the 1995 study, and they divided those samples into three groups. More than half (183) of those samples were held "in a reserve pool" and were not analyzed. No selection criteria for these "reserve" samples were stated in the paper which I think is an odd omission.
As I feared, they took surface samples, not deep core samples:
There is no other statement regarding how each sample was extracted. Were those surface joints remortared in more modern times thanks to the repair work of, e.g., one or more Dynastic king, or perhaps contaminated by modern hydrocarbon pollution, or finger oil and other bodily fluids deposited there by untold thousands of climbers and creatures over the centuries? The article gives us no information that would give us a reason to prefer or doubt one possibility over the other. The authors simply state that some dates were not included in the final dating conclusions and that...Quote
"While searching the monuments, we examined seams between stone blocks for mortar filling and for black specks of charcoal inside the mortar."
...without stating what those other contexts are, how they were able to control for such contexts within the samples that were included in the final result, or what their criteria were for assessing whether samples are "probably from another context" other than stating the mathematical algorithm used to omit outliers without stating the rationale for that degree of filtration:Quote
"Screening was used in an attempt to remove dates from samples which are probably from another context."
It's interesting that they claim to have outliers in their data without running Bartlett's test for population homogeneity. In other words, they did not test to see if all mortar samples are from the same homogeneous population; they simply assumed homogeneity and omitted "outliers" according to that presumption.Quote
"The difference between the weighted mean of all dates and the individual dates, divided by the product of ␣2 and the error of the date, was used to flag outliers. Consistently eliminated were all dates where the computed number exceeded 5.0."
The black bars indicating the confidence band of dating the samples (Fig. 1) all indicate date ranges that are significantly earlier than the historic chronology for almost every structure that was sampled. For example, the beginning of G1's construction is generally believed to be around 2580BC. However, Fig 1 indicates C-14 dating of G1 mortar as early as 2850BC (ie, 270 years earlier). And note that the confidence band is only "1 sigma" in that figure, which is only 68% probability. When you open it up to a 95% confidence band (2 sigma, which is a more common standard in the quantitative sciences), the margin of error goes back to about 3000 BC. And that's the surface mortar which has a significant probability of being contaminated with contemporary carbon, e.g, from local fires, vehicle exhaust - which has become epidemic there - and the other contaminations I mentioned above), etc., which artifactually would introduce a bias into the dating methodology. It could very likely appear to be a more recent date than it really is, due to such contamination form exposure to the environment (no such contamination would be possible in deep core samples). Such contamination might result in the variability observed:
I admit I'm no expert on C-14 or AMS, but it does seem like a wide spread which could indicate a non-homogeneous population of mortar (e.g., original old vs. newer restoration(s)). In addition to that spread, there isn't much difference in the C-14 dating of the pyramids of Djoser, Dashur, Meidum, and the three at Giza; they all test for the same general date and do not at all show the temporal spread in chronology that is historically accepted. That is, C-14 data did not confirm the sequence of pyramid construction accepted in the historic chronology. As a result, although Menkaure has traditionally been placed later than G1, or around 2500 BC, the C-14 dating placed it in the same date range as G1, with a start date possibly as far back as 3000 BC (with 95% confidence). That's a 500 year disparity for Menkaure. And that's not taking into account the possible bias due to modern carbon contamination.Quote
"Some monuments include sample dates which are much older or younger than the established mean."
The authors clearly state that they only took samples on which they could visibly see a 1-2 mm of ash particle (i.e., surface samples). They didn't comment on whether the mortar was homogeneous in organic materials all over the sampled structures, vs. whether different types of mortar were used (e.g., original vs. restoration/repair) which might suggest a different timeline for each different type of mortar. It certainly can account for the wide spread in the C-14 results. I'm tending to doubt that all AE mortar is of the same homogeneous composition.
Note that the ARCE samples (form an earlier study) were included in this report, and they resulted in average dates for G1 that are almost 250 years older than the samples taken in '95. No explanation was offered for this significant discrepancy.
Meanwhile, Hawass apparently learned a few lessons about carbon dating during this time which led him to decry the use of such methodology as recently as 2010:
"However, Zahi Hawass, Egyptian archeologist and secretary-general of the Egyptian Supreme Council for Antiquities, strongly disagrees with the use of carbon dating in archeology. 'Carbon-14 dating has a margin of error of 100 years. In order to date Egyptian dynasties, we need to have specific dates; you cannot use carbon dating,' Hawass explained to Al-Masry Al-Youm. 'This technique shouldn’t be used at all in making changes to the chronology of the ancient Egypt, not even as a helpful addition.'"
“ 'Not even in five thousand years could carbon dating help archeology. We can use other kinds of methods like geoarcheology, which is very important, or DNA, or laser scanning, but carbon dating is useless. This science will never develop. In archeology, we consider carbon dating results imaginary.'"
The instructions to authors describe a relatively weak "peer review" process which allows authors to recommend a reviewer and which apparently does not require a full disclosure statement regarding potential conflicts-of-interest.
There is enough quirkiness in this study that makes me want to see another study with less potential for conflict-of-interest, less involvement with Zahi, and a more scientific consideration toward sample selection. It's hard to believe a skilled geologist would accept such raw surface samples of mortar as being able to provide accurate C-14 dating after being exposed to the environment over several millennia.
In summary, my concerns about this C-14 report include:
- Authors are biased against previous C-14 studies which conflict with the historic chronology and completely neglected to consider reassessing the historic chronology
- Authors did not address possible COI with Koch funding of AERA
- Inexplicable omission of more than half the C-14 data points
- All samples were scraped from surface mortar, no deep samples
- No description of surface cleaning to remove outer layer of modern carbon contamination before taking the surface sample
- All samples were biased toward mortar with visible charcoal specks and were not randomly selected
- No rationale given for the process of eliminating "outlier" data points
- AMS shows wide spread in dating, suggesting multimodal, non-homogenous samples; authors only ran Chi square for fit, but not Bartlett's for (unimodal) normal distribution homogeneity
- No control for long-term environmental contamination
- No discussion of the discrepancy between the ARCE and the '95 data sets
- Involvement of Zahi Hawass in the study, especially vis a vis his expressed opinion completely decrying the use of any C-14 dating in archeology afterwards
- The authors do not address the systematically older C-14 dates for almost every structure sampled vs. the traditional historic chronology
- The authors do not address the lack of C-14 confirmation of the historically accepted sequence of pyramid construction.
- Questionable manuscript review process by which the authors may recommend a reviewer; the apparent lack of a 'full disclosure' requirement
Post Edited (10-Mar-15 15:08)
How can any of us ever know, when all we can do is think?