Dale Glover has produced a new podcast, called “Shroud Wars: Breaking the Breaking News (Fan Requests on WAXS vs. C-14),” of which the first half seems to return to his old style of listening to everybody and summing up with his own opinion, and concerns Giulio Fanti’s WAXS experiments – two years old now but achieving sudden prominence in the worldwide popular media – but the second half attempts justify a criticism I made of his assessment that the AMS technique was too unreliable to produce a reliable result.
This seems a curious way to begin a criticism of the radiocarbon results, in view of the subsequent continuous and persistent attempt to explain them away. If the AMS technique was unreliable, then nobody would have been… um… relying on it, but of course it was both well-established and well-recognised by 1988, and the main reason why the Vatican decided that the Shroud was, at last, able to be reliably dated. Carbon dating itself was about 40 years old, and the AMS technique about 10. Is that “new”? I think that’s debatable, but it’s surely not “brand new” which was the phrase I specifically objected to, and still object to. Glover, however, says: “At the time it dated the Shroud it was only about ten years old. […] Hugh says this is a lie.” Of course I don’t. Glover seems not to have read my post – or has conveniently forgotten what I said. He continues: “It was around the late 1970s that Harry Gove and his other co-inventor invented AMS carbon-14 dating so I have zero clue why Hugh in his blog says that it wasn’t; that this is wrong.” I said no such thing. If this is not a lie on Glover’s part then it is surely culpable ignorance.
It gets worse. Up on his screen there is a quotation:
Of this quote, Glover says: “Notice that Hugh is actually admitting here that I was correct in the quotes that I gave. […] New Scientist is where I’m getting it from and I’m quoting it correctly.” That’s exactly what I mean by ‘Qui tacet consentire videtur.’ Just because I didn’t say it was wrong in my critique, therefore, according to Glover, I am admitting he was correct. Huh. Of course he wasn’t. This is a garbled version of the opening sentence of the New Scientist article by Andy Coghlan, dated 30 September 1989, which actually reads:
“The margin of error with radiocarbon dating, an analytical method for finding out the age of ancient artefacts, may be two to three times as great as practitioners of the technique have claimed. […] …the accuracy with which 38 laboratories around the world dated artefacts of known age. Of the 38, only seven produced results that the organisers of the trial considered to be satisfactory.”
No singling out of AMS at all. But astonishingly, Glover actually shows the article from the New Scientist as part of his denunciation of me, and claims that both versions are the same! “That’s exactly as I’ve quoted in my blog,” he says, “so I’m correct.” No, it’s not exactly what he quoted, and he’s very incorrect to claim that it is.
Glover then quotes from Murdoch Baxter, whom I described as disgruntled because his laboratory (not AMS) did not fare particularly well. Glover misquotes him as well, but according to the New Scientist article, with respect to the International Collaborative Program, he said that “accelerator mass spectrometry, used last year by a laboratory at the University of Oxford to date the Turin shroud, allegedly the burial shroud of Jesus Christ, came out of the survey badly. Five of the 38 participating laboratories used this technique, for which samples weighing a few milligrams are acceptable. The other techniques require grams of the sample. Baxter says that some of the accelerator laboratories were way out when dating samples as little as 200 years old.”
Now, apparently I was lying to you all when I mentioned the relevant paper in which the AMS labs didn’t do particularly badly, as Glover is delighted to tell us now that “there’s actually two peer-reviewed papers.” Actually there are several,1 but let’s go on. “Hugh only mentions the last stage. I could accuse him of lying to you.” He could; indeed he does, and has done, repeatedly, in this post and the previous one. But is he justified? Let’s see. After admitting that the conclusions refer to all the types of dating technique involved, Glover says: “But the problem is the results. Even Hugh admits, from the peer-reviewed papers, the carbon-14 labs sucked. They performed horribly, and that’s including the AMS labs.” Hugh does not admit any such thing. In fact all the labs, especially when treated as a group rather than individually, bracketed the known dates. Here is a graph of the actual dates achieved (from my October 2022 post, ‘Interlaboratory Radiocarbon Comparisons’).
Glover’s quote from Baxter looks a bit sour in this context. Did the AMS labs come out particularly badly? In every case above, the AMS results are more closely grouped about the known dates than either of the other two methods, and if it is true to say that “some of the accelerator laboratories were way out when dating samples as little as 200 years old,” then it is quite clear that the others, including Baxter’s Liquid Scintillator method, were considerably worse. Never mind, Glover believes Baxter – or at least what Baxter was reported to have said in the New Scientist. “I believe him, not you, Hugh.” Fair enough, whatever floats your boat. I was actually relying on a peer-reviewed paper published in a secular scientific journal (one of Glover’s favourite phrases) rather than a popular magazine.
Eventually we get to some attempt to understand the results of the International Collaborative Program, but Glover doesn’t really understand them and flounders with every sentence. He fails to understand “systematic bias” and thinks Table 2 in the ‘Third Stage’ paper refers to ‘failing’ labs when in fact it refers to ‘participating’ labs, and although he seems to understand that the program clearly shows that the AMS labs performed better than the others, he wanders off into a distraction about the expected error for the Shroud, and thinks that because 40% of the Collaboration Program labs showed a consistent (but obviously small: see their results) systematic bias, then “statistically we would expect two out of the three labs to screw up and have the wrong date. […] Frankly, because they only had three AMS labs carbon-14 dating the Shroud it is very, very likely that all three labs had a systematic bias.” No, we wouldn’t, and no it isn’t.
Then Glover goes on to make a hopeless mess of an earlier paper, ‘An Interim Progress Report on Stages 1 and 2 of the International Collaborative Program,’ which he thinks shows that the AMS process “was pitiful. It was worse than everything else.” It’s true that I didn’t mention it in ‘Absence of Evidence? Make it up!,’ but not because I thought it weakened my case, simply because I thought the case was strong enough without it. Yup, Stage Two looked at cellulose, which is what linen is largely made of; let’s see how our labs got on.
Figure 2, in ‘An Interim Progress Report on Stages 1 and 2’ shows the dates all the labs achieved from twelve different samples, from the top: one peat, one shell, three wood, one cellulose, one humid, one algal, two benzene and two carbonate. It is divided into three sections, from left to right, Gas Counter, AMS and Liquid Scintillation. The cellulose results have been highlighted, by me. The greater accuracy and narrower spread of the AMS labs are obvious to see.
But Glover is still clinging to the New Scientist report and its interview with Murdoch Baxter. “So three out of the five AMS labs fulfilled the criteria, but qualitatively, how well did the AMS labs fulfil the criteria? Murdoch Baxter is not talking in terms of a simple pass/fail here. His quotes, saying that the AMS labs sucked, worse than everything else, and were horrible, were probably based on qualitative assessment. Remember, I quoted him as saying that the AMS labs were horrible. They were so bad they were way out. When they make a mistake, they really make a mistake, even with dating samples as young as two hundred years old. The conventional labs are way better; at least when they make a mistake they’re not as bad as mistake as the AMS labs are.”
That’s ludicrous. For start Glover seems to be saying that it can be better to fail than to pass, and secondly, he doesn’t realise (see my graphics above) that however poor the AMS labs, the others were considerably worse.
Then Glover returns to Table 3 and again, completely misreads it, in an increasingly desperate attempt to malign the clearly superior AMS. Here it is:
He totals the three numbers on the LSC and the GC lines, and decides there were 16 of the former and 15 of the latter, but fails to notice that the AMS line totals to 2, when there were 5 AMS participants. In fact, as he finally realises, there were 13 LSC and 20 GC. The reason for the higher LSC total is that some of the LSC labs failed on more than one count, so their “FAILING” total adds to 16. In fact we know that eight of the labs failed on “two or three counts.” None of them were AMS labs.
But suppose, Glover tries, desperately, there had been more AMS labs participating. Would the obvious supremacy of the method be mitigated? Statistically, as he well, knows, the answer is no.
Some three or four times during this harangue, Glover invites his audience to choose between a world-famous expert in radiocarbon dating and ‘a high-school science teacher’ – what a feeble argument. I don’t want to be part of an argumentum ad auctoritatem. It shouldn’t matter if I’m a teacher or a plumber or a check-out assistant at a supermarket. My evidence is set out here, just as Glover’s is set out in his podcasts. Judge on the evidence, not on the person.
==================================
Now we move on to the second of the Interlaboratory Comparisons mentioned in Refuting the Skeptics, ‘An intercomparison of some AMS and Small Gas Counter Laboratories,’ by Richard Burleigh et al.’ Glover gets this hopelessly wrong, but puts all the blame on Mark Antonacci, from whom, rather than the paper itself, he seems to have acquired his information. Having done that,
Glover gets around, presumably for the first time, to looking at the paper itself, and completely misunderstands it. His argument centres around Table 1 of the paper, so here it is:
Glover’s point is that “All five labs got the ages wrong by half the known age or more in a total of 7/17 experiments.” The dates from the 18 experiments are listed under ‘Date,’ and in order to be wrong by half the known age, perhaps we must halve the dates given under the sample titles themselves, i.e. 1-Egyptian: 3000BC / 2 = 1500, 2-Peruvian: AD1200 / 2 = 600, and 3-Peruvian AD1000-1400 / 2 = 500-700. Now all we have to do is find any experimental dates outside these parameters, 1-Egyptian: outside 4500BC-1500BC, 2-Peruvian: outside AD600-1800, and 3-Peruvian: outside… difficult to say; let’s take medians… outside AD600-1800. By my count, that’s one 1-Egyptian, all five 2-Peruvian and one 3-Peruvian. That’s 7. Hooray, hooray, Glover wins the day!
How absurd. This simplistic gibberish is so wrong it’s difficult to know where to start. What, for example, does “half a date” mean? A calendar date is not just a number. That’s probably the silliest idea, but more serious is the fact that neither Antonacci nor, now, Glover, have read what Table 1 actually reports. Let’s cast our eyes up to the paragraph under the heading Results:
The Dates from the laboratories are not Calendar Dates, and thus cannot be compared to the Calendar Dates of the samples listed above. Either the BP Dates must be calibrated into Calendar Dates, or the Calendar Dates converted into BP Dates, before any comparison can be made. I choose the latter. From OxCal, the online calibration program, we get: 1-Egyptian: 4400BP, 2-Peruvian: 870BP, and 3-Peruvian: 1050-580BP. Matching these to the measurements made by the laboratories, we get:
There is an outlier for the 3-Peruvian results, at 1550BP, and all five labs dated 2-Peruvian as more modern than it had been thought before. Even then, three of them were within Glover’s “half the known age.”
For the next five minutes Glover dances on what he hopes is my grave, without noticing that it is empty, before moving onto another attempt at defending his errors, this time with respect to Dr Lloyd Currie’s article in the Journal of Research of the National Institute of Standards and Technology. All I mentioned here was that Glover misquoted the article, in a way that I thought was misleading. Rather than accept this, Glover spends undue time explaining to his audience that he is entitled to misquote anybody he feels like for the sake of clarity. I’m afraid I don’t think that’s true.
Here is his quote:
His insertions and omissions are not indicated by square brackets, as is conventional, but included in the same colour, within the direct quotation marks. Not to understand that this is academically wrong is, I’m afraid, culpable ignorance. But it’s OK, because this time it turns out that is Ian Wilson’s fault, because Glover hasn’t read the primary source, just copied it from somebody else.
=================================================
This podcast peters out in more self-justification and abuse, which need not concern us here. The bottom line is, and it’s repeated over and over again, that I’m a liar, and surely the essence of a lie, be it malicious, inadvertent or whatever, is that it is an untrue statement. So, if any of my readers have got this far, return to Absence of evidence? Make it up! and then come back to this Lasciate ogni speranza, and tell me if I’ve said anything untrue.
1 — ‘An Interim Progress Report on Stages 1 and 2 of the International Collaborative Program,’ E.M. Scott et al, Radiocarbon, Volume 31, 1989
— ‘Report on Stage 3 of the International Collaborative Program,’ T.C. Aitchison et al., Radiocarbon, Volume 32, 1990
— ‘An Overview of All Three Stages of the International Radiocarbon Intercomparison,’ E.M. Scott et al, Radiocarbon, Volume 32, 1990
— and numerous discussions and further papers looking into particular aspects of the study in the ‘Proceedings of the International Workshop on Intercomparison of Radiocarbon Laboratories,’ published in Radiocarbon, Volume 32, 1990