Re-examining the Somerton Man Identification: A Forensic Perspective
In July 2022, a high-profile announcement claimed to have solved one of Australia’s most enduring forensic mysteries — the identity of the Somerton Man.
According to the report, DNA extracted from a hair sample embedded in a 1949 plaster 'death mask' led to the identification of the deceased as Charles “Carl” Webb, a Victorian man born in 1905. The story made international headlines and was widely celebrated as a triumph of forensic genealogy.
But as any serious forensic researcher knows, the strength of an identification lies not in the press announcements or the narrative, but in the rigor and detail of the methodology behind it.
For the past two years, I’ve undertaken a detailed review of the hard evidence, the protocols employed, and the statistical reasoning that underpinned the Webb identification. The result is a 40 + page forensic critique titled:
“Methodological Concerns in the Carl Webb Identification of the Somerton Man: A Critical Analysis of Forensic DNA Evidence and Research Protocols.”
This paper is now available for peer review and public scrutiny...
Why I Wrote This Paper
Taking the task on was a daunting prospect. While I have been researching the Somerton Man case for around 15 years and have written more tha.n 1000 posts in that time, an Academic article of this depth required a whole new approach as far as presenting the information was concerned. I don't profess to be an academic but thankfully, when it comes to research, I have a fair amount of very relevant experience. Nonetheless, there was still a learning curve when it came to DNA and forensics.
The Somerton Man case has fascinated hundreds, if not thousands, of researchers for decades, and any claim made as to its resolution deserves not just recognition but rigorous scrutiny.
From the outset, my objective was not to dismiss the identification outright, but to assess whether the methods used met the standards expected in forensic science, particularly when the evidence that existed was historical, degraded, and irreplaceable. What matters is the truth of the information, and that task has always meant that a great deal of focus and attention to detail was essential.
On that note, amongst many aspects, my focus was on the following:
- The presence of Ante Mortem Root Banding (AMRB) in the published hair sample — a feature that can only occur in hair from a living donor
- The absence of a documented chain of custody for the DNA sample used in the 2022 analysis
- The lack of direct testing of Webb family members and the absence of cross-references to earlier credible DNA results
- A demographic statistical model showing nearly 10,000 potential candidates from the relevant cohort
- Historical evidence that human hair was commonly added to plaster busts for realism, often months after death
A Key Forensic Distinction
One of the most significant findings centers on the type of banding observed in the hair sample. While some have interpreted the darkened region as Post Mortem Root Banding (PMRB), the location and morphology are more consistent with AMRB — a feature that cannot form in deceased individuals.
Significantly, forensic literature confirms that AMRB is structurally stable over time, whereas PMRB is highly susceptible to environmental degradation, especially over 74 years in an alkaline plaster environment.
The distinction is not academic; it goes to the heart of whether the analyzed hair could plausibly have come from the deceased at all.
Transparency and Replicability
The paper includes detailed appendices outlining:
- Statistical modeling of the 1905 Victorian male cohort
- Forensic hair analysis protocols and quality control standards
- Chain of custody documentation requirements
- A full APA-formatted reference list and replication instructions
The article is definitely not a matter of speculation. It is a structured, evidence-based critique intended to elevate the standards of forensic discussions and discourse, ensuring that high-profile identifications are held to the same rigor as you would expect in a courtroom.
An Invitation
I welcome feedback from those experienced in forensics, historians, researchers, genealogists, and, realistically, anyone with a genuine interest in the subject. If the Somerton Man case is ever to be resolved, the resolution must be grounded in verifiable science, not just compelling storytelling.
You can access the full paper here. I look forward to the dialogue it may inspire.
Gordon Cramer
Independent Researcher
The Somerton Archive
That’s an interesting read, I became a little unsure when you commented on post mortem/ante mortem banding. I understand the difference between the two but just a shade unclear as to the effects of the environment where the bust was stored. The photos I’ve seen show the bust a a museum type case for display, wouldn’t that have protected it from the outside environment?
ReplyDeleteYes, there’s some truth in what you say, the display case would tend to protect the bust. However that case was only introduced about ten years ago, I don’t have the exact date of that. Before the case, some 60 plus years, the bust was kept either in an open office or in a cupboard. It would have been wiped and cleaned numerous times and all of which exposed the bust to contamination. The real issue raised in the paper is the fact that Post mortem banding, which only occurs after death, deteiorates over time for the reasons spelled out in the document. Not only that but it affects an area of the hair which is close, within .5mm, to the root bulb. The image displayed in the IEEE article does not show the proximity of the banding and instead of a root bulb a tapered end of the hair is seen. That is one very concerning aspect of that image but there is another. As mentioned earlier, Post Mortem banding fades over time. The image looks relatively sharp, you can see the contrast in the colouration. Put those two things together and it becomes plausible to say that what we see in that image is Ante Mortem banding, the kind that is observed in hair samples taken from a living subject. The ‘clincher’ as it were, is the fact that human hair from living subjects was used in the creation of plaster busts for realistic sculpting effects. I hope this clarifies it for you, it was a good question and please keep asking !
ReplyDeleteGratulerer Mr. C. That’s a huge and very detailed job, must have taken months and months to finalise it. I can’t question it yet because I haven’t finished reading it. I liked all the resources and references you included.
ReplyDelete(This comment received from a Norwegian friend, translated as above)
You’ve gone to a lot of trouble here, that’s a comprehensive article, the citations leave very little room for doubt. My question is why didn’t the Professor provide an article like yours?
ReplyDeleteYou’d have to ask the Professor that question.
DeleteThose stats say a lot, the Prof only looked at 4000 men who were in the right cohort when there were more than 9000 in reality. Something not right about that.
ReplyDeleteI just love wading through statistics!
ReplyDeleteThe key find has to be the human hair in the sculpted wig. That says a lot about the sample that was taken but there’s still a question about why he never crossed checked that hair against the ACAD 2018 sample that they analysed.
ReplyDeleteIt’s certainly amongst the key finds. There had been earlier mentions of hair bought from hair dressers salons some years back. The question then became was that hair used by mixing it in with plaster for plaster busts? There was anecdotal evidence that it was sometimes used but that wasn’t really solid or reliable evidence for example hair from salons was commonly used to make wigs and hair pieces.. The breakthrough there came when I looked more critically at the images of the bust and noticed how the hair appeared to be sculpted. That’s when I started to search for information about hair being used in sculpting ‘wigs’ for plaster busts and there it was. But I must stress that by itself that was one key finding amongst others. It’s all in the Academia paper, the link’s in the post above.
DeleteI wondered a lot about the hair. In one of the images the Prof said it was grey in the pic but all along from Dr Bennet’s words it was ginger, Gerry Feltus said the same and the reporter Pat Burgess all said it was a gingery colour. If they were using salon hair that might be the cause?
DeleteThanks Netty, that’s a good point. There are a few close up images of hair in the IEEE article. They were referred to as being the Somerton Man’s hair but what wasn’t said was whether it was of the actual hair that was sent to Astrea for analysis. The outstanding question relates to, this, given that it was hair used for realism in sculpting the ‘wig’ and that it showed Ante Mortem Root Banding, meaning it was from a live subject, was that hair bought in and Mr Lawson had it in his tool kit or was it a last minute thing and was it cut from the head of one of the detectives who were assisting him? Maybe a hair from Mr. Lawson’s head? When it comes to the colour of the hair, if it was bought in, then that hair would have been processed as in disinfectant, soaps, chemicals as well as probably being boiled. The process would have affected the colour. It is also likely that hair embedded in plaster over the years of exposure to contamination, would have discoloured. Going back to Pat Burgess the journalist, there’s a photograph of him examining the bust, I think it was an eyebrow. The comments made by the Professor, I think, were about hair from the man’s head? A good question!
DeleteYou haven’t mentioned the height issue. By my estimation Webb is 3 inches shorter than the Somerton Man’s.
ReplyDeleteThe reason the height of the man is not mentioned in this post is because the height is not mentioned by Professor Abbot in his IEEE Spectrum claim. But so you are aware, the height of Carl Webb was first shown and proven to be 5’8” in a post on this blog in 2024. I then followed that up in a separate and more detailed post in August of this year, here’s the link:
ReplyDeletehttps://tamamshud.blogspot.com/2025/08/how-tall-was-carl-webb-verifiable-answer.html
I added a detailed account in the comments of that Post as follows:
Gordon332August 04, 2025
No problem. We start with the photograph of the Webb family group of 4. It has 3 men standing around Eliza Webb. Let’s call the men A,B and C, being Richard, Roy ab]nd Carl in that order.
We know from records the measured heights of A and of B. There is no recorded height for C or Carl.
If each of the images of those men were full height from head to toe images, it would be a relatively straight forward image matching exercise given that all three men were standing in the same plane.
The challenge is two fold. First, they’re not all standing on the same plane and second, we do not have full head to toe images.
I put it to my mathematically superior friend that perhaps we could carry out an exercise based on comparing the upper body or proportional comparisons.
In this way we overcame the issues of standing on the same plane or one or more of the men were standing on a slight rise or in a dip in the ground. We didn’t need the head to toe measurement. We would still have a relatively minor problem of camera lens distortion but we could account for that by applying an anticipated error factor.
4 coordinate points were placed on the three men’s upper bodies in the photograph. I then measured the distance in pixels between the coordinates for A,B and C. For everyone’s reference you can use Photoshop or Gimp both of which have pixel measurement tools.
The next task was to properly frame the problem as in ‘ If height A = Height B then what is the height of C?
The said mathematically inclined friend created an algorithm into which he entered the precise pixel counts in the appropriate locations and pressed the GO button so to speak.
The algorithm then generated the answer of 5’8” as Carl’s height within an error range of a shade under + or - 2 inches. Physically that meant that Carl’s height lay between 5’6” at the lowest estimate and 5’10” at the highest extreme of the range. The closer you go the the highest extreme, the less likely it is.. the right answer is far more likely to be in the centre of the range.
Separately I applied a possible error range of + or -1 inch for the measured known height of the Somerton Man. At the low end of the range he would have been 5’ 10” and at the highest end, his height would have been 6’. Once again the closer you get to the extremities of the range, the less likely that Haight would be.
You can quickly see that the high end estimate and the least likely estimate of Carl’s height is 5’ 10”. And that compares to the low end extremity and least likely estimate for the Somerton Man also equals 5’ 10”.
By default, the most likely heights for Carl Webb and the Somerton Man are 5’ 8 inches and 5’ 11” with an extremely high probability factor.
That’s about it. Hope this helps.
Hello Gordon, we can only hope that SAPOL and Forensics utilised the bone pulverisation DNA extraction methodology on the West Terrace exhumed remains, just as Toronto Police did on a female who was buried before 1949 and discovered in 1995. Here is the case study:- https://www.dropbox.com/scl/fi/x5s9vkw8uj179b5w1i20w/Anthropology-Report-A.pdf?rlkey=tj1iru3jfqipnjf9zoo6ehbq7&st=lxb2o4ri&dl=0
ReplyDeleteFingers crossed.
Many thanks, Alan. I read through the document, and a fascinating case it is.
ReplyDeleteWhat caught my attention, as no doubt it did yours, was the first, and wrong, DNA identification. I followed through on the YouTube documentary, and it gave an entirely different explanation. It shows the problems that arise when people rush in with half an answer and insufficient hard evidence.
In this case, the first DNA results were inconclusive, shades of the Somerton Man case. It is simply not good enough to make a grand announcement based on incomplete DNA data, even though it may make the researcher look good in the first instance, with lots of media attention. I can certainly recommend downloading the full report and then watch the YouTube documentary on the Danforth Jane Doe here:
https://youtu.be/WhxM3BaqnBY?si=p6fnTqmElzyXidjL
The video sounds so convincing, but it was so wrong.
The bottom line is that an amateur researcher with the best of intentions can have as many opinions as they want, but the truth of the matter is unlikely to be found amongst them.
Thanks again, Alan, a really good, really relevant find.