Abstract
In 1988, three laboratories dated a small section of the Turin Shroud using the relatively new Accelerator Mass Spectrometry (AMS) method, and returned three dates, all round about the end of the thirteenth century. As these were collated, it was noticed that, even allowing for the errors suggested by the laboratories, two of the dates did not overlap. Statistically, this could be said to indicate that the two fragments giving those dates were significantly different. They could have come from completely different artifacts, or represent material differences within the four centimeter strip from which they came. Alternatively, the material was homogeneous, but the errors quoted by one or other of the laboratories were underestimated. This was assumed the most likely, the errors re-estimated, and a cohererent medieval age for the Shroud calculated and published.
Almost immediately, however, there was a backlash from those who felt that other evidence, suggesting that the Shroud was the authentic burial cloth of Christ, was more compelling than the radiocarbon date. They set about accounting for the anomalous results in various different ways, mostly on the basis that the three dates showed a significant chrolonogical gradient along the sample strip, suggesting that it demonstrated progressively different contamination, thread replacement, or carbon-14 enrichment from natural or supenatural events. All these depend on a particular order of the three laboratories’ sub-samples along the strip, usually expressed as Oxford-Zurich-Arizona, from the closest to the corner of the cloth, with the Oxford sample the oldest and the Arizona sample the youngest.
In this paper I attempt to reconstruct the radiocarbon sample pictorially, from documentary evidence taken at the time and photos of the samples.
Evidence
In view of the importance and gravitas of the occasion, it is remarkable that a truly accurate and comprehensive account of the events of 21 April 1988, and the subsequent division and testing of the samples in the laboratories is by now almost impossible to recreate. Measurements were either not made or not recorded, photos not taken, and memories, even only a year later, became confused. What follows is chiefly derived from the videotape of the proceedings in Turin, of which only a tiny fragment has been published, and some still photographs taken there and at the laboratories. The few minutes of video I examine are part of an Italian film, La Notte della Sindone, made in 2012, so I cannot tell if its editing was carried out then, or by the instigator (and financer) of the recording of the events in Turin, Giovanni Riggi di Numana, although I do know that at least one “short version” was presented at a lecture in Melbourne in 1990.
A hour or so before noon on 21 April 1988, in the sacristy of the Royal Chapel behind the high altar of the cathedral of John the Baptist in Turin, Giovanni Riggi di Numana, a microanalyst who had been involved with the study of the Shroud since 1978, cut a small strip of cloth, about 8cm long and 2cm wide, off the corner of the Shroud adjacent to the ventral image of the legs. He passed it to Franco Testore, Professor of Textile Technology at the Polytechnic of Turin, who carried it to an electronic balance and weighed it. As it contained much of a rolled hem which had been stitched to the backing cloth of the Shroud, Testore placed the sample on a piece of metal foil, where Riggi trimmed off the hem, leaving a neater strip, which Testore again weighed, coincidentally achieving the remarkable mass of 300.0mg exactly. This trimmed strip was then folded in half and cut down the crease, and the two halves weighed, and then the smaller ‘half’ cut into three and the three little pieces weighed. Two were about 50mg, and one about 40mg, so a thin strip was cut off the larger ‘half’ to make up the mass.
Almost all the weighings, and the readings on the scale of the balance, were observed by the video camera, and there are also shots of the piece of paper beside the balance on which the masses were recorded. Significantly, the order of the weighings shown in the video (at least, the version included in Il Notte della Sindone) is different from the order shown by the numbers on the paper, and, as will be seen, cannot be correct. The voiceover (which I only have access to in the English dub) is also incorrect.
The table below shows the masses of the various samples and sub-samples as they appear on the video. The order is that in which they were written down. The abbreviations N/S and N/R mean Not Shown and Not Recorded. Sometimes a reading appears on the balance scale, but was not, apparently, written down, and sometimes a record appears on the paper which is not shown on the balance. Sometimes a written total may be calculation carried out on paper to cross-check total mass after a piece was subdivided, but in no case does the whole literally equal the sum of its parts.
The Radiocarbon Sample Masses – Raw Data
1) The trimmings may not have been weighed at all. 2) Presumably the balance flickered between 0.1548g and 0.1549g. 3) This addition may have been on paper only. It is 0.0004g less than its previous total of 0.3000g (B). 4) These pieces are labelled 1, 2 and 3 on the paper, but shown in the order H, I, L in the video. 5) Probably never weighed together, but the total is 0.0003g less than E. 6) It looks a bit like 0.0162, but 0.0141 is the commonly accepted value. 7) Adding G and K in fact comes to 0.0538g, which is 0.0001g heavier than L, which was the two pieces weighed together. 8) 0.1407 g. Maybe never weighed separately. 9) This comes to 0.0141g, not the recorded 0.0142g, but is the commonly accepted value.
We note that, over the shots of the three laboratories’ samples being weighed, the voiceover says: “The first sample weighed exactly 52mg, the second 52.8mg, the third, in two parts, 53.7mg”. However two shots of the paper on which the masses were recorded show, firstly:
1. 0.0396 2. 0.0520 3. 0.052.8
and secondly, with addition:
0.0537 1. 0.0396 + 0.0142 2. 0.0520 3. 0.052.8
It is thus clear that the mass of the small sliver (0.0142) and the mass of the first sample and the small sliver combined (0.0537), were added after the first three weighings. From the paper, it could be interpreted that the Oxford sample (0.0520g) was in the middle, but the video makes it clear that it was as one end, opposite the larger Arizona sample.
It will be as well to mention here that at no point during the portions of the video so far released is there any indication that any measurements of the dimensions of the samples were taken or recorded, apart from a rough estimation of the total length and width of the orginal strip cut from the Shroud. In my opinion all such ‘measurements’ on subsequent diagrams are little better than guesses.
Some time later, probably in 1989 but perhaps later still, Giovanni Riggi di Numana attempted to draw an accurate diagram of the various pieces, in their proper place on the sample. Not only did each piece have a weight, but some measurements in centimetres were also recorded.
He appears to be using the data from the sheet of paper written at the time of the weighings, with the exception of the 14.2mg for the ‘sliver’, which derives from a calculation (53.7mg-39.6mg), not the measurement (16.2mg)
Almost immediately he must have realised he was wrong. Either he realised that he had labelled the smallest of the three sections on the left hand side with the biggest mass, or that the shape of the piece kept back by Turin has a more pentangular than rectangular shape. For whatever reason, he redrew the diagram thus:
For more data, we will now turn to what photographic evidence there is. In Turin, lots of photographs were taken of the space left after the sample had been cut away, but very few of the area prior to being cut. Here is the best:
Note the very irregular shape of the upper edge of the Shroud, the ruler, the ‘spine’ of the herringbone pattern, and the ruck in both cloths.
And here is the area after the sample has been cut off, being held flat. The ruck has flattened out somewhat. The sample area is considerably larger than than that shown in the previous photo, showing that it was actually bigger than the area in the upper photo.
Here the first picture, trimmed of its frame but including the ruler, is overlaid on the second picture, correctness of scale and alignment being ensured by using a series of stitch-holes visible in both photos.
Next, let’s look at the available photos of some of the pieces the sample was cut into.
From left to right, as they were on the Shroud, there is the ‘riserva’ section that has been kept in Turin ever since; the Zurich sample, which was published quite soon after the sampling; the Oxford sample, which was released in 2014 after some determined foraging by Pam Moon; and the only photo of the Arizona sample, in two parts, together with the tinfoil in which they were wrapped and the metal capsule in which they were carried from Turin to the laboratories. This was, according to Bruno Bonnet Aymard “a ‘ souvenir-photo’ (!) taken at the Tucson laboratory, Arizona, on Sunday April 24, 1988, which Professor Donahue sent to us on January 3, 1991.” Although little information can be gained from this last, the first two have scales, and the third is clear enough to be superimposed accurately enough to establish its size. All we learn from the fourth is the approximate shape of the two pieces taken to Arizona.
Here they are, to the same scale as the pictures of the whole corner above, in situ (with the exception of the Arizona sample, which is too vague for sufficient precision). The ‘spines’ of the chevrons are in red, with other coloured stripes every ten threads away, although the ‘riserva’ section is not clear enough for the intermediates.
We notice the two blank areas on either side of the photographed pieces. The space on the right was occupied by the Arizona sample, but the space on the left is intriguing. Although the edge on that side, like the edge along the top, was trimmed away, the area seems a bit too large. It is not impossible that the extra sliver cut from the ‘riserva’ piece in order to make up the weight of the initially too light Arizona sample might have come from this end.
However, there are two other photos to fit into the picture, both fragments of the Arizona sample. One wasn’t used but retained by the laboratory and photographed in 2012 by Barrie Schwartz, and the other was subsequently tested, and destroyed, but its photograph was shown at the 1991 New York Shroud Symposium by Douglas Donahue.
It is not possible, simply by looking at these photos, to say which way is “up”, but they are sufficiently clear for a sensible attempt to fit them into the picture.
Finally, let us see how they match up to the Shroud as it is currently seen, as photographed by the Haltadefinizione team in 2008.
When Sue Benford and Joe Marino first promulgated their “invisible patch” hypothesis, they suggested – and were apparently supported by textile experts – that the area above the spine (the upper red line) was modern, and the area below it was original.
(From: Marino, Joseph and Benford, Sue; Evidence for the Skewing of the C-14 Dating of the Shroud of Turin Due to Repairs, 2002, on shroud .com. Text rewritten for clarity.)
Since 2002, the idea of a literal patch has morphed into ‘invisible mending’, and the proportion of modern to ancient material has been found to have to be much greater than illustrated above. Even from an authenticist point of view, the drawing above should better appear with a ‘bleed’ of modern into ancient material, like this:
Even this ‘improvement’ however, is unsatisfactory. Although such a scenario could explain the published radiocarbon dates if each laboratory tested its entire piece all at once, that’s not what they did. They sub-divided their samples into at least 3 (Oxford), 5 (Zurich) and 4 (Arizona), giving 12 individual dates in a chronological sequence. However, since some of the Zurich and Arizona dates overlap, the sub-divisions could not have been made like this:
[N.B. The left-hand sliver, labelled A1, was not used in the test at all, and, as mentioned above, the area labelled A was actually divided into at least five pieces, four of which were tested and one retained. The Oxford lab used up all their sample, and it is not known whether the Zurich retained any or not.]
Marco Riani and and Anthony Atkinson used the twelve individual dates to achieve a statistically most probable way in which the three labs may have divided their samples (Riani, Marco, Atkinson, Anthony et al.; ‘Regression Analysis with Partially Labelled Regressors: Carbon Dating of the Shroud of Turin’, Statistics and Computing, Vol 23, Issue 4, 2013). They analysed “387,072 plausible spatial allocations” to find the closest correlation to a chronological gradient, and this was their result:
As can be seen this does not correspond at all with Benford and Marino’s supposition. Half the subdivisions are entirely made of ‘modern’ material, and would have dated to the 16th or 17th centuries, while five others contain such a small proportion of interpolated material that it would have dated to only a few centuries AD. The Riani and Atkinson gradient looks like this:
Interestingly, this direction of gradient is at right angles to that proposed by Bob Rucker, who thinks that it could have been caused by neutron radiation from the dead body lying on the Shroud (Rucker, Robert, The Neutron Absorption Hypothesis, 2018, at shroudresearch.net).
If the Riani/Atkinson gradient is correct, then it could be the result of either of two possibilities, an increasing amount of more modern contamination towards the top right of the sample, or an increasing amount of more ancient contamination towards the bottom left.
In fact, neither of these options fits well into the larger context of this corner of the Shroud, and it may be that the gradient is entirely spurious. Bryan Walsh and Larry Schwalbe have suggested that the cleaning regime of the Oxford laboratory, which, alone of the three labs, involved the use of petroleum ether, may have resulted in removing more contamination, making it appear 88 years older than the others. (Walsh, Bryan and Schwalbe, Larry, ‘An Instructive Inter-Laboratory Comparison: The 1988 Radiocarbon Dating of the Shroud of Turin’, Journal of Archaeological Science, Vol 29, 2020). They therefore suppose that the Oxford date is more likely to be correct. Another possibility is that the petroleum ether was not entirely removed, and, since it has zero radiocarbon itself, acted to make the Oxford sample appear older than it really was, and the Zurich/Arizona estimates were more correct. Whatever the reason, there is clearly no evidence which stands up to the overall correctness of the radiocarbon date.
Here is the radiocarbon corner, fully reconstructed as well as possible:
