Jack the Ripper identified?

UPDATE: Go here to see a post from later on this day, detailing Adam Rutherford’s objection to this paper, which he calls a “joke” because of the possibility of contamination, the lack of provenance of the shawl, the sloppy calculations, and the weak genetic evidence.



The paper below is getting a lot of publicity, and its conclusion, that Jack the Ripper—the long unidentified British serial killer—was really a 23-year-old Polish barber named Aaron Kosminski, is being trumpeted in all the media. After all, the Ripper, who killed five women in 1888 in a particularly gory manner, has long fascinated the public as an unknown figure of evil. Hundreds of books and papers have been written speculating about his identity, but nothing conclusive has resulted.

Now, a paper in the Journal of Forensic Sciences (probably free with the legal app Unpaywall; click on link below, or get pdf here) purports to identify the killer using DNA evidence on a shawl linked to the crime. But I’m not really impressed with the paper, as the authors don’t present the results in full, and a piece in Science about the paper identifies further problems.

Here’s the link; if it doesn’t work, I’ll be glad to send you the pdf:

The identification is based on DNA extracted from a silk shawl found at the murder scene of Catherine Eddowes, a shawl recovered by a police officer and passed down to his relatives. (Science, however, says that the shawl was bought in 2007 by a writer, Russell Edwards, and given to Louhelainen. It also notes that “Other critics of the Kosminsky theory have pointed out that there’s no evidence the shawl was ever at the crime scene.”—neither of these facts is mentioned in the paper).

Eddowes, one of the five victims, and one of two killed on September 30, had several wounds, and her left kidney and her uterus were cut out and removed (there is plenty of gory detail, and a photo of the autopsied victim at the Wikipedia page).  A card purporting to be from the Ripper, and referring to the “double event”, is known as the “Saucy Jacky postcard”, and was received at police headquarters on October 1, but its authenticity is uncertain. This is a facsimile of the front (the card has disappeared), and the message is below:

Text on the obverse:

I was not codding dear old Boss when I gave you the tip, you’ll hear about Saucy Jacky’s work tomorrow double event this time number one squealed a bit couldn’t finish straight off. Had not time to get ears off for police thanks for keeping last letter back till I got to work again.

Jack the Ripper

There’s also a “letter from hell” that accompanied a kidney that was sent to a London vigilance committee, but the handwriting on that letter differs from that on the Saucy Jacky postcard.

Aaron Kosminski was a suspect in the murder, apparently because he “hated women and had strong homicidal tendencies”, but even here there is ambiguity, as Wikipedia reports that his name might have been confused with that of a Jewish bootmaker, Nathan Kaminsky (aka “David Cohen”), and even “Aaron” as a first name was recovered from records of Londoners long after the murders were committed. Both men died in insane asylums.

Back to the paper. The shawl, as I said, was made of silk, and had dyes that ran in the rain. The authors suggest that a fancy garment like this probably belonged to the Ripper, not to Eddowes, as Eddowes was a poor woman who worked as a casual prostitute.

The authors located stains on the shawn by visual inspection of the shawl and UV and infrared photography. Bloodstains were distinguishable from semen stains (presumably the killer masturbated on the shawl). Here’s the shawl:

(From Figure 2 of paper): Images of the shawl parts. Upper left: largest piece of the shawl with the blue and brown sections. Lower left: the floral detail on the shawl. Right: smaller piece of the shawl from the blue side.

The stained areas—both putative bloodstains and sperm stains—were washed and DNA from the extracts sequenced; in most cases they isolated single cells from the material and extracted and amplified both mitochondrial and nuclear DNA from those cells, also doing profiles of others who could have contaminated the shawl in the interim. Not much DNA could be obtained, but enough for the authors’ analysis, which shows how refined the technique has become. Single cells can help incriminate you!

They then compared the mitochondrial DNA (mtDNA) from the putative blood and semen samples to DNA from relatives of Eddowes and the suspect Aaron Kosminski.  Although mtDNA is passed on only through females, it is present in the midpiece of sperm, and is destroyed after fertilization. mtDNA is useful for this analysis because it can be passed on intact (save for rare mutations) in the female relatives of the victim and perp; all one has to do is find females in female lines of descent from Eddowes’s and Kosminski’s mothers, and compare their DNA with that on the shawl. Who they sampled and their relationship to victim and suspect are not reported in the paper for legal reasons, though I think they could have given at least those nameless individuals in a line of descent.

The way the data are reported is weird: they appear as simply blocks colored in, denoting a match in the mitocondrial DNA regions, and the matching sequences are not given. Here’s how the matches are shown:

(From paper’s Figure 7): Sequencing results of mtDNA presented as graphical blocks when deviations from the Human revised Cambridge reference sequence (rCRS) have been recorded. Color coding is used to highlight the results (victim = blue, suspect = red, owner = green, and laboratory operator = gray).

As you see 6 regions of mtDNA on the shawl that were variable (apparently single nucleotides, though they’re not clear on this nor give the sequences) matched Eddowes’s sequence, and seven regions of the shawl that were variable matched mtDNA from relatives of Kosminski (there was a non-match for two nucleotides).

The frequency of a match for Kosminski (using the present-day European population) is .019, or 1.9%, while the frequency of the type for Eddowes is 0.0013, or 0.13%.  These are interesting, but also mean that thousands of people in London today could be a match for either sample. The results, then, are intriguing but not dispositive. And they have no database for London, neither now nor of course from 1888, so these figures are at best ballpark estimates. Neither the shaw’s “owner” (not specified) nor the lab boss (or technician) matched the stains.

The 11 nuclear DNA sites were used only to determine what the killer’s hair and eye color were, as these sites are like those used in 23&Me to tell you what you probably look like. They show that the killer had brown hair and brown eyes, which match one eyewitness report (not given in the paper) of what the Ripper looked like. As the authors note:

One of the strengths of this paper is the demonstration of the use of aged single cells as a source of genomic DNA. This was taken further by using in‐house assays for amelogenin sex determination and phenotypic SNP markers. The results were in full accordance with one of the very few witness statements considered reliable: a male with brown eyes and brown hair. Although these characteristics are surely not unique, they fully support our hypothesis. We have no reliable information on how common these phenotypic features were with males in London in 1888, but at the moment, blue eyes are more common than brown in England.

Well, this isn’t very strong evidence, and to be sure the authors don’t make a strong claim that they’ve identified the killer. However, they fail to present all the data they have so that other scientists can judge how strong their conclusion really is. And they seem to omit critical data, like Science‘s claim that there’s not strong evidence that the shawl was in fact found at the murder scene. (There is a chain-of-custody letter.)

So, has Jack the Ripper’s identify finally been revealed? The answer is “probably not,” but the data, at least in the weak form presented here, increase the likelihood that Aaron Kosminski, who was a suspect in the murders, was the killer. But we’re a long way from knowing who butchered those five women. Caveat lector. 


Louhelainen, J. and Miller, D. (2019), Forensic Investigation of a Shawl Linked to the “Jack the Ripper” Murders. J Forensic Sci. doi:10.1111/1556-4029.14038


  1. Dominic
    Posted March 19, 2019 at 10:07 am | Permalink

    Readers may be interested in the book that ‘rescues’ the victims & points out that most of them were not actually prostitutes but have been painted as such –
    The Five by Rubbenhold -https://www.goodreads.com/book/show/37545347-the-five

    • Posted March 19, 2019 at 11:13 am | Permalink


  2. BJ
    Posted March 19, 2019 at 10:36 am | Permalink

    At this point, I see no reason to believe any theory regarding Jack the Ripper’s identity. Unless something truly remarkable and unimpeachable is discovered, we’ll never solve a crime that happened nearly 150 years ago.

  3. Posted March 19, 2019 at 12:14 pm | Permalink

    I’m going to apply Betteridge’s Law to the headline of this article and answer no.

    In fact, this line:

    “The identification is based on DNA extracted from a silk shawl found at the murder scene of Catherine Eddowes, a shawl recovered by a police officer and passed down to his relatives”

    confirms my judgement. The provenance of the scarf must be in dispute and, even if it were genuine, the opportunities for contamination must have been numerous.

  4. Doug
    Posted March 19, 2019 at 12:27 pm | Permalink

    It’s interesting how Jack the Ripper killed five people over the course of a year 131 years ago and we’re still talking about it. Today, school shooters kill several times that number in a few minutes.

    In the 1979 movie “Time After Time,” Jack the Ripper [David Warner] steals a time machine and travels to 20th-century America. When he sees how violent it is, he remarks “Ninety years ago I was a freak. Today, I’m an amateur.”

    • BJ
      Posted March 19, 2019 at 12:43 pm | Permalink

      I freaking love that movie.

      And yes, it is strange that we’re still fascinated by this case after all this time. I’ve been consistently perplexed as to why. Would it be nothing more than a footnote in history if he had been caught?

  5. Ken Kukec
    Posted March 19, 2019 at 12:38 pm | Permalink

    I dunno. Maybe. Or maybe shades of Amelia Earhart.

  6. Jonathan Dore
    Posted March 20, 2019 at 3:43 am | Permalink

    I’ve always thought Francis Tumblety was by far the most convincing-sounding suspect, i.e. someone with a plausible psychological profile for a serial killer of women, in London at the time etc.

%d bloggers like this: