The very definition of pseudo-science

[Edited to take account of the comment by Michael Kowalski, below]

On 18 April 2024, two months ago as I write this, an online petition appeared on change.org, calling for “the retraction or amendment of the article published February 16th, 1989 by Nature Magazine titled ‘Radiocarbon Dating of the Shroud of Turin.'” The grounds for this claim are quoted as “substantial doubts cast on the validity of the 1989 article’s conclusions,” which are enumerated under three headings:
– Sample Contamination Concerns: Subsequent analyses and critiques have raised significant concerns about the possibility of sample contamination. The area from which the 1988 radiocarbon dating samples were taken may have been subject to repairs in medieval times, thus not representative of the entire shroud.
– Methodological and Analytical Flaws: Questions have been raised regarding the methodology employed in the 1988 study, including the statistical analysis of the dating results. These concerns suggest that the conclusions drawn do not reflect the true age of the Shroud of Turin.
– Impact on Interdisciplinary Research: The 1989 Nature article has had profound implications for subsequent research on the Shroud of Turin. A retraction and or amendment would acknowledge these concerns and promote a more nuanced and careful approach to the study of this and similar historical artifacts.

No further information is given, although perhaps, should the petition ever be sent to the Editor of Nature, a covering letter would include more detail. Given the petition as it stands, however, the rationale for retraction is pretty weak. The claims that the area tested “may have been been subject to repairs in medieval times” or that “questions have been raised” about the methodology of the tests and the analysis of the results are speculative in the extreme, and do not amount to any kind of challenge, and the idea that the paper has had “profound implications for subsequent research on the Shroud of Turin” is more of a cause for retention than retraction.

Before signing the petition, it would be better for the potential signatory to try to elucidate these reasons, and where better than to listen to three podcasts involving the promoter himself, Guy Powell? They can be found at
— 30 April 2024. Guy Powell’s own site (youtube.com/watch?v=cshgsXbOJfc)
— 30 April 2024. Real Seekers (youtube.com/watch?v=Q44h9Cl77zQ&lc=UgyeoMnZwFdMVbm0NWN4AaABAg.A2u_GWc7GaYA2xYTmttkRt)
— 4 May 2024. The Gracious Guest (youtube.com/watch?v=qkROGzEzeQY&t=2124s)

The petition has also been discussed by Bob Rucker and Dale Glover on Stephanie Thomason’s podcast, and briefly reiterated by Guy Powell in a discussion with Michael Lofton.
— 11 May 2024. Truth Matters w/SJ Thomason (youtube.com/watch?v=Oen0VAX_Quk&t=5346s)
— 21 May. Reason & Theology (youtube.com/watch?v=l43eDe6tNCY)

The Real Seekers podcast has been duplicated on Faith Unaltered, and there have been two repeat pleas on Guy Powell’s website, one a thirty second booster, and one a discussion with Matt Collins of ‘Third Day,’ and the petition has been publicised on the “Shroud of Turin” and “British Society for the Turin Shroud” Facebook pages.
— 30 April 2024. Faith Unaltered. (.youtube.com/watch?v=6OEKA9_Tljw&t=1682s)
— 28 May 2024. Guy Powell. (.youtube.com/watch?v=KlIVLB4wdi8)
— 31 May 2024. Guy Powell. (youtube.com/watch?v=tokv7uEHju4)

After all this, and about 7000 views, the petition has gathered just over 400 signatories. Signing costs nothing, takes almost no time, and signatories do not have to give their interests, qualifications, or any indication whether they know or care anything about the Shroud. Few of them will have read the Nature paper they would like retracted, and few of those who have read it will have understood the statistics anyway. Those who know anything about scientific research and publication, even if they are convinced authenticists, will simply be embarrassed by it.

Perhaps it would be better to let sleeping dogs lie, as the petition seems to have quietly evaporated and no doubt a podcast such as this could inspire a few more signatories from die-hard authenticists with little or no scientific background, but since the whole affair encapsulates a substantial proportion of authenticist sentiment, and since I have been taken to task for supposing that the petition was in fact never intended to be taken seriously, and since, just possibly, it may end up being sent to the editor of Nature, it may be worthwhile explaining why I find it disappointing, and why I strongly encourage anybody truly interested in the Shroud to have nothing to do with it.

====================

The best place to start will be the detailed discussion Guy Powell had with Dale Glover on Real Seekers on 30 April. Bearing in mind that this is meant to explain in detail why the Nature paper should be retracted, and therefore, we might assume, aimed as squarely at the people with the power to do so as much as interested sindonlogists, it seems bizarre that the introductory text below the YouTube window refers to the paper as containing “deliberate lies, fudged data and proven false conclusions.” This is strong stuff, which, following Carl Sagan’s aphorism: “extraordinary claims require extraordinary evidence,” will need substantial verification in the 70 minutes of the video itself.

Unfortunately it soon becomes clear that the principal point of the podcast is not about the science, but about the character of the participants, ranging from the Cardinal Archbishop of Turin, Anastasio Ballestrero, through Professor Michael Tite, of the British Museum, to Sir John Maddox, Editor of Nature. Almost immediately, Guy Powell demands a retraction, rather than calls for an amendment, of the Nature paper, and then states his case succinctly:

“That radiocarbon dating was flawed in a huge number of ways. It’s just hard to believe that you could have scientists that believed in facts, and leading from facts and theories to then come up with a date that was just totally devoid of proper scientific rigour and everything like that.” I agree. It’s very hard to believe. There had better be very good evidence to mitigate our incredulity, although our confidence that this will be forthcoming is weakened immediately by two verifiable errors, namely that the publication of the paper “stopped research for quite a while,” and that the paper was “supposedly peer-reviewed by a little over 20 different scientists.” A glance at indexes of scholarly research shows that research on the Shroud scarcely wavered, let alone stopped, and as for the peer-reviewed scientists, I think Powell is confusing authors (21) with reviewers (2). It would also be more usual to refer to a scientific paper in a scholarly journal rather than an article in a magazine, but let that be.

Well, never mind, how was the Nature paper flawed? “We have uncovered the flaws. We know that the radiocarbon dating has been totally debunked, and we want Nature magazine to re-write their article.” Yes, but how was the Nature paper flawed?

The next thing is an interjection by Dale Glover, saying that I and other medievalists should sign the petition. Why? “Because you should be supporting the truth. Look, it’s making your side look bad. All you people have this flawed data. This is the best the skeptics can come up with?” OK, so it’s flawed, but how?

At last, Powell mentions “nine or ten bullet points” describing the flaws, in a video by Michael Kowalski. Here we go:

First: “The agreed-on process was that there were going to be seven different carbon-dating laboratories…” And, as we all know, there were in fact only three laboratories. But wait a minute, what does Powell mean by “agreed-on”? Agreed-on by whom? Certainly not agreed-on by the custodian of the Shroud, his scientific advisor, or the owner, Pope John Paul II. Guy’s information comes from the so-called Turin protocol, which was drawn up by Harry Gove after the Turin conference in 1986 and published in November 1987. By that time it had been overtaken by events and was more of a forlorn aspiration than a formal proposal. It was never adopted by any laboratory at all. It is sometimes confused with the transcript of the conference drawn up by the Pontifical Academy of Science, which was 417 pages long, convoluted and contradictory, and in no sense a protocol of any kind. The actual choice of laboratories was made by the Vatican Secretary of State in May 1987, and the formal protocol adopted by the people involved in the dating was communicated to the laboratories in a letter from Luigi Gonella in February 1988. The Nature paper recognises these events as follows: “… a meeting was held in Turin in September-October 1986 at which seven radiocarbon laboratories (five AMS and two small gas-counter) recommended a protocol for dating the shroud. In October 1987, the offers from three AMS laboratories (Arizona, Oxford and Zurich) were selected by the Archbishop of Turin, Pontifical Custodian of the shroud, acting on instructions from the Holy See, owner of the shroud.”

No cause for retraction there.

Next: “They were going to use two different carbon-dating methods…” No. Powell is still talking about the Turin protocol, which was never accepted by the Church authorities.

Next: “They were originally going to take samples from three different areas.” Turin again. Not relevant to the Nature paper.

Next: “The STuRP scientists, “the best and brightest in the world,” […] were also going to be the ones who were going to cut the samples from these three different areas.” Turin again, and not relevant, but wait a minute! Who were these “best and brightest scientists in the world” who were going to be involved? The 1984 StuRP Test Proposal lists them: Dinegar, Harbottle, Oeschger, Damon, Tite, Stuiver, Otlet, Hall, Gove, Coleman and Donahue. Of these, Damon, Donahue and Tite were authors of the Nature paper, and Stuiver acted as a consultant. More than a third of the team proposed by STuRP were closely involved, as was Giovanni Riggi di Numana, who organised and supervised the whole sampling procedure, and firmly listed himself as a member of STuRP. And where did STuRP want to take their samples from? The 1984 StuRP Test Proposal mentions 200 mg samples from: each of the burned areas under the patches, the side strip, the Raes corner, one of the patches and the backing cloth. The only ‘clean’ area of undoubted Shroud required was from the Raes corner, where, in fact, the sample was finally taken.

Next: “The samples that they took were supposed to be then disguised…” At last! A real deviation from the real protocol. The idea was that the labs would be given three samples, two controls and the Shroud, and they would not know which was which. This was only partially successful as the chevron weave of the Shroud is almost unique, and sufficiently well known as to be easily recognisable. Nevertheless, the samples were placed, without being videoed, in unlabelled (except with a number) containers, so that until they were opened they were unidentifiable; and after they had been burnt and reduced to carbon dioxide, at least two labs changed staff, so that the actual measurers did not know which was which. The Nature paper reports:
“Because the distinctive three-to-one herringbone twill weave of the shroud could not be matched in the controls, however, it was possible for a laboratory to identify the shroud sample. If the samples had been unravelled or shredded rather than being given to the laboratories as whole pieces of cloth, then it would have been much more difficult, but not impossible, to distinguish the shroud sample from the controls. (With unravelled or shredded samples, pretreatment cleaning would have been more difficult and wasteful.) Because the shroud had been exposed to a wide range of potential sources of contamination and because of the uniqueness of the samples available, it was decided to abandon blind-test procedures in the interests of effective sample pretreatment. But the three laboratories undertook not to compare results until after they had been transmitted to the British Museum. Also, at two laboratories (Oxford and Zurich), after combustion to gas, the samples were recoded so that the staff making the measurements did not know the identity of the samples.”

Does this constitute deliberate lies and fudged data? I think not, but the next thing we hear really is a deliberate lie and fudged data, only it’s by Powell, not by the Nature paper! “The results were shared across the labs, shared across other folks, and so it was not kept confidential. That also then means there could have been collaboration, or other kinds of things that were just wrong.” There is absolutely no evidence for that at all, and to imply that it might have occurred is simply an unjustified slur on the characters of the scientists involved.

But Powell’s got into his stride now, and we reach the statistical analysis of the data which was “either fraudulently done, or incorrectly done at best.” Really? First flawed, now fraud. I hope the evidence stacks up. But we’ll come to that later. Beforehand, we hear that: “at the end of all of that, the institutions that were the statistics guys, had to then have a kind of a joint agreement with the labs that they had properly assimilated those results, and properly studied them, and then put them into the right format so that they all agreed and it really made sense. That was not done.” How does Powell think he knows this? The British Museum archives are stuffed with letters back and forth between the British Museum and the laboratories doing exactly that.

We are about a third of the way through, and Powell sums up that so far, that he has listed nine or ten bullet points, all of which should have been followed, but none of which were. His account, his reasoning and his accusations are all, however, so transparently incorrect that we find absolutely no cause to retract or amend the Nature paper in any way. Maybe the next two thirds will change our minds? (Spoiler alert: don’t hold your breath).

In an interlude, Dale Glover decides that because the tests were not as blind as he wanted, there “was an active – for lack of a better word – conspiracy,” to which Powell replies, “personally, I think there was a conspiracy,” although he immediately admits, “it’s hard to prove it so maybe let’s call it an alleged conspiracy.” This is fudging. They both think that some presently unspecified people conspired fraudulently to achieve some presently unspecified object, without any evidence. And this is the basis for their appeal to retract the Nature paper? I hope it gets better…

At the end of the day, says Powell, the labs made a ton of money, because of the notoriety they gained from being able to prove the Shroud to be medieval, and “basically to thumb their noses at Christianity.” This is just scurrilous. The dating of the Shroud was a justification of the AMS process, and such sponsorship as the labs acquired was to further its development. The idea that either the scientists or the sponsors wanted to “thumb their noses at Christianity,” is not even an allegation, it is a whine. But even if there was some substantiation for it, is it a sensible way to present a petition for the retraction of their paper?

For the next ten minutes there’s a general attack on the integrity of the radiocarbon scientists and statisticians; they were a “conspiracy,” perpetrating a “crime,” by “hiding” information that refuted their results, and Nature “just went ahead and published” a paper that “carbon-14 scientists” had told them contained “certain things that should not be published.” However, amongst all the bitterness there is a kernel of information that has informed modern detractors, but which was not known to the authors of the paper. This is hardly evidence of dishonesty, but could be a reason for re-evaluating the conclusions of the paper if the new information significantly affected them. If Powell knew or Dale cared, enough about statistics, they might validly focus on this aspect, but unfortunately it easily gets lost among the overall emphasis on dishonesty and malice.

The information not considered by the Nature paper is simply the order in which the samples were cut from the Shroud. Of the six ways that three letters, say A, Z and O can be arranged, only two have Z in the middle, but careful study of the video taken of the cutting of the samples suggests that, indeed, the Zurich sample was originally between Oxford and Arizona. This being so, then the possibility of a gradient in age along the sample strip becomes statistically more probable, and a reason for that gradient becomes worth investigating, and if found significant, could merit a re-analysis of the radiocarbon data.

This doesn’t bother Powell much, of course, who ploughs on with his accusations of fraud. A typical authenticist moan is that “the British Museum did not publish the raw data. Normally in any kind of a scientific study you publish the raw data. […] They hid the data.” This is not true at several levels. First the British Museum originally submitted a much more detailed analysis of the results to Nature, who replied that since their Journal was more of a general interest than a specifically statistical publication, a more compact description would be more appropriate. Secondly, even in statistical journals, full details are often not published, but available as an online supplement (see, for example Tristan Casabianca’s paper in Archaeometry), and in the 1980s, were not published at all, but could be requested from the authors. Thirdly, when further details were requested, such as by statistician Remi Van Haelst in the 1990s, they were freely given and discussed in detail. Fourthly, when Tristan Casabianca visited the British Museum to examine their archives, he discovered nothing that added significantly to what had been released to Van Haelst twenty years previously. Thus it is completely untrue that the British Museum was “forced” “through a Freedom of Information act” to release their archives; they had been freely open for inspection for thirty years.

“When he [Casabianca] got the raw data, what came out of it was absolutely stunning.” Was it? Has Powell read the bit which says: “Our statistical results do not imply that the medieval hypothesis of the age of the tested sample should be ruled out”? Hardly a damning indictment. Three subsequent papers (not “a handful of other studies,” unless you include some of my own blogposts), one by di Lazzaro et al. and two by Schwalbe and Walsh, have not changed that position. Di Lazzaro specifically says, “In other words, in order to avoid misunderstandings, we do not claim that the radiocarbon results are incorrect, and it is not our purpose to try to shift the time range,” and the other two calculate reasons for the discrepancy that hardly shift the medieval date at all.

We are halfway through trying to establish some reason for retracting the Nature paper, with absolutely no success at all. I do hope the next half is a bit more convincing.

I forgive Powell his next faux pas, as it is a common one among statistical extrapolators, which is to suppose that a gradient established by measuring three points is a pointer to information way beyond those points. If I set off up a slope climbing two metres in ten steps I do not suppose that after a thousand steps I must climb two hundred metres. It may only be fifty metres to the top. Similarly, the fact that the Shroud appears to decrease in age by thirty years per centimetre in one corner does not imply that after a metre it decreases by three thousand years.

But I do not forgive his following claim. “It’s something that they hid. Not only did they ‘fake it up and flog it,’ but they hid it,” or Glover’s “They just dismissed it, literally, mindlessly, as random measurement error.” No, they didn’t; they were unaware of it. But here, as often happens with conspiracy theories, Powell trips over his own logic. If there was a plot to produce false results, and collaboration between the labs and the British Museum, then the results would have been unequivocally consistent. Discovering these anomalies demonstrates far more clearly that there was no conspiracy than that there was.

But Powell wades further in. “If you were a true scientist, and you saw that systematic error… if the British Museum was being run by true scientists at that time, and they saw that systematic error they would have said, you know, ‘OMG, what happened here? What’s going on here? We need to investigate further. We need to understand further.’ And instead, like you said, they swept it under the rug.” How unfortunate that the data was unable to be examined by “true scientists,” but by Michael Tite, Morven Leese and Anthos Bray.

And then suddenly, at about 36 minutes in, we get a massive retraction. “I think though,” says Powell, “that the dating of the material that they received was probably done correctly. […] I think all of the labs, based on what they did […] they did the radiocarbon part of it correctly. It was the work that was done before that, and the work that was done after that that was then incorrect.” So, let’s get this clear. In spite of all the collaboration they were accused of previously, the three labs, measuring their individual samples, each cut into several smaller pieces, measured, recorded, and reported the radiocarbon present correctly, namely that every test showed a minimum of 90% of modern radiocarbon present in their sample.

So it was the sample itself that was unrepresentative of the cloth as a whole, and, if there was indeed a conspiracy to provide a medieval date at the the end, the Church authorities, who only permitted one sample site, and the two independent textile specialists who chose that specific area, were aware that it was not representative, and in fact consisted mostly of new material from the 1600s. Without any prejudice to the scientists, the British Museum or Nature, the paper should be retracted because of a conspiracy to provide a faulty sample by the custodians of the Shroud in Turin. It’s a fairly extreme claim, and, as such, requires some fairly extreme evidence, whereas in fact there is no evidence at all.

But no, wait! Maybe there wasn’t any conspiracy at all! Paddling backwards hard, Powell now suggests that maybe, all unknown to anybody concerned, and undetected by the labs, the sample actually consisted mostly of surface grime accumulated over the years from the constant handling of that area by sweaty clerics. There seems to be a bit of backtracking here, from ‘Fraud’ back to ‘Flawed,’ or from ‘Crime’ to ‘Grime.’

Then there is an interjection from Glover, presumably referring to the papers by Schwalbe and Walsh, who attribute the difference between the Oxford result and the other two to unremoved contamination. He claims that the skeptical position used to be that the radiocarbon samples were cleaned perfectly (“the carbon-14 scientists were perfect; they can never be wrong”), but that after the discovery of the chronological gradient, they now agree that the labs were “incompetent” and did leave contamination on the samples. This is a travesty of anybody’s opinion, and as usual, without any evidence to sustain it.

However, this is followed by a particularly generously-spirited interjection from Powell, who commends the skeptics (naming myself as one) for finding apparent flaws in the authenticist arguments, enabling the authenticists, by countering the flaws, to strengthen their case.

The main thrust of the argument is over, and we have nothing to go on.

At last, however, thanks to a beautifully made presentation by Michael Kowalski, we get to statistical objections to the Nature paper, which, if sustained, could merit re-examination or even retraction if sufficiently severe. Here are Powell’s comments: “For example, they didn’t disclose all the test results. They inflated the Arizona margin of error, […] so that it would correspond to the other two labs and their margin of error.” And finally they rounded 4.1 up to 5 instead of down to 4. And of course, “I don’t know if that’s a mistake […] but it sounds like they’re trying to hide something.” “How can you round up 4.176 to 5?”

If Powell wasn’t so committed to the conspiracy theory, he might have looked into some of these points, and realised that Kowalski didn’t claim that the Arizona margin of error was inflated to “correspond to the other two labs” (after all, as Kowalski points out, the error was reported before either of the other labs had even carried out their tests), and that “they” did not round “4.1 up to 5 instead of down to 4,” but neither Powell nor Glover want to know. Powell goes straight on to pretend he knows something about peer review, and to make a couple of glaring errors. The Nature paper was not reviewed by “21 different scientists;” It was reviewed by two, but Powell not only muddles authors and reviewers, he then blithely tells us that peer reviewers don’t make much effort to review the papers they’re assigned to anyway. “The peer review was probably not as rigorous as we would have liked.” Wishful thinking. No evidence.

Finally, the fact that the chair of Archaeological Science established by Edward Hall was subsequently sponsored by ITV and a consortium of rich friends is assumed by Powell to be evidence of malpractice. Considering he exonerated the labs of dishonesty and blamed everything on the British Museum a few minutes previously, this seems muddled thinking.

The rest of the podcast wanders off onto different topics and replies to online questioners, which need not concern us here, but ends with the following paean from Glover, which is guaranteed to bring the editor of Nature, Dr Magdalena Skipper, to her knees: “The carbon dating was deeply flawed. It’s garbage. Even if you want to still believe the Shroud’s a medieval fake, because you’re new to it, the carbon-14 is still flawed, and if you’re interested in truth, even as a Shroud skeptic, you should be backing us up, in getting this horrible paper from 1989 retracted because it is the very definition of pseudo-science.” Well, good luck with that! Anybody really interested in truth, even as a Shroud authenticist, must realise that this kind of rhetoric could not be better calculated to antagonise every one connected with Nature from the humblest type-setter to the Editor-in-Chief.

====================

What might have been a more sensible critique of the paper could have been offered by Bob Rucker, who actually is a scientist, and so focusses on the nub of the contentious points, namely the statistical analysis of the results, rather than any alleged jiggery-pokery in producing them in the first place. He explained his point of view at Truth Matters w/SJ Thomason, but in order to make things easier for “the layman” to understand, I believe he actually made them more confusing, and ended up ignoring the single value in the entire paper which is most in need of explanation.

If radiocarbon dating was reconsidered from scratch today, the process of converting a proportion of carbon-14 in a sample to a calendar date would be a simple matter of calibration. We would go straight from 95% radiocarbon to 1450 AD, without any intermediate steps. Unfortunately, by accident of history, proportions of radiocarbon are usually converted first into “Years Before Present,” which are simply a number produced by plugging the radiocarbon proportion into a simple mathematical formula, and then the “Years Before Present” are converted into calendar years by reference to the calibration charts. Rucker, as we shall see, introduces a third step, going from Radiocarbon Proportion to Years Before Present to what he calls Years AD which then have to be calibrated into Calendar Years. Whether this makes things easier for the layman is debatable, in my opinion; it certainly doesn’t make things easier for himself, as we shall see.

Here is Rucker’s first slide on the subject, and I find it misleading.

“All the black values,” says Rucker, “are from Damon [the lead author], published in Nature.” But they’re not, of course. Rucker’s values are what you get by subtracting the actual values published in Nature from a nominal 1950. He could have subtracted them from 1900, or 2000, or added them to a million. Rucker goes on: “When I calculate the data, I obtained different values from the values shown in Damon. And I shouldn’t. I should be able to calculate the values, because they give me this data. [..] What is going on here? That is why this paper should be retracted.”

But of course he obtains different values from the values shown in Damon. Damon didn’t subtract his data from 1950. If Damon had subtracted his data from 1950 then he would have got the same results as Rucker, except in the case of Arizona, for which, as carefully explained to Morven Leese by Doug Donahue, exceptional circumstances applied. It would be relevant and useful to know how the actual figure published for the Arizona sample, was achieved, but the idea that a scientific paper should be retracted simply because one of its readers doesn’t know “what is going on here,” is ludicrous.

====================

Roughly 8600 views, and about 600 signatories (September 2024). Maybe this podcast will stimulate a few more.

I have been taken to task elsewhere for supposing that the petition was never sincere, and should be thought of as a rallying cry from well behind the front line than a serious attempt to persuade a scientific journal to retract a scientific paper. The accusations of conspiracy amount to nothing more than personal abuse, the alleged evidence for it would be largely irrelevant even if true, the statistical recalculation is ill-considered, and the whole tone is so aggressive as to guarantee minimal consideration by any responsible authority. This is so obvious that it cannot be unintentional, and therefore the petition cannot be intended even to reach the desk of the editor, let alone merit consideration.

Perhaps that’s the point. Send a petition, and when it’s ignored smugly rejoice that once again the armies of Satan, represented by the conventional science establishment, have shown their evil hand. In fact, of course, if it has any effect at all, it will simply act as one more nail in the coffin of the authenticist cause. Serious scientific researchers, authenticist or not, should be appalled.