More recent story:

Lucia de B., angel or witch? [HAR 2009]


I just turned away from the lock-picking talk, as the tent was absolutely packed (me being 5 minutes late). I don’t know how many people fit in these convention tents, hundreds, perhaps thousands, but that is the amount of people that after tonight may know how to break every lock you own.

Earlier today I was at the talk with possibly the smallest amount of listeners of this 4-day exercise, you might even say the attendants resembled Cantor Dust. OK, lousy statistical jokes aside, this talk was by statistician Richard Gill of the University of Leiden and dealt with the Lucia de Berk case.

I had heard of the case before. In 2001, a nurse from The Hague was accused of having murdered dozens of patients, and the strange thing was that most of her guilt was determined by statistics: she had been near the victims at the time of their deaths, and although a direct link with the accused in the form of a confession or evidence could not be established, the court found that the statistical likelihood of her being near all these victims at the time of death was so minute, she must have done it.

At the time I thought this reasoning seemed silly, but I have learned early on in life never to argue with statisticians. So imagine my surprise: here was a statician who argued that the court’s reason had indeed been extremely silly, and that an innocent woman had gone to jail.

I won’t bore you with repeating the entire lecture: author Maarten ‘t Hart summarized Gill’s position excellently in this article from NRC (Dutch). Gill’s paper on how likely the chance is that a nurse was on active duty during all deaths concludes that one in nine nurses would have gone to jail (PDF).

Although the case has gone all the way to the Hoge Raad (High Council), with every court on the way finding De Berk guilty of multiple murders and eventually condemning her to life in prison, the Hoge Raad has now decided that a lower court should look at the case again after an important expert reneged on his earlier, vital testimony.

One thing from the talk sticks out in my mind. In the Netherlands, you are rarely allowed to take photos of the accused, the reasoning behind this presumably being that somebody is innocent until proven guilty, and therefore it should be made more difficult to have an accused be condemned in the eye of the public.

Courts will, however, allow artists to make drawings of the persons in a case. For the Lucia de Berk trial, the artist had chosen to draw her as a hagard, mean looking woman somewhere in her fifties. Gills also showed a photograph of De Berk, a happy forty-something who according to the speaker most resembled Uma Thurman.

If you want to know what any of the other HAR talks were about, check the HAR program: every talk has its own homepage with an abstract, and often links to transcriptions or the slides of the talk. If all else fails, googling the name of the speaker could help you locate similar talks or articles.

More recent story:


  1. Richard Gill says:

    I really like the Cantor Dust joke! Definitely not lousy.

    BTW more recently I have numbers like 1 in 25 according to perhaps my favourite model; the “worst” I could come up with is 1 in 1000, which is what you get when you correct Henk Elffers’ 1 in 342 million for the obvious errors in methodology and well documented errors in the data. The randomness being tested is not the right kind for this situation. Hospital environment is not constant, the events in question are highly correlated (eg three events of just one child among the nine or so!), nurses shifts are not purely random either.

    BTW 2: *personally* I am pretty sure that Lucia is innocent and moreover that most likely, no one was ever murdered by anyone (except, technically speaking, for the euthanasia which was removed from the data-set because … well because? … I guess it would have been embarassing). *Scientifically* my well-founded scientific opinion is that there is no evidence on the basis of which to convict anyone for murder; indeed no serious evidence for any murders at all (except the afore-mentioned). Of course as someone said to me the other day “still, she could have done it”. I was too dumbfounded to give one of the good answers; eg “done what?” or “yes and so, in principle, Queen Beatrix could have done it too, doesn’t she also spend a lot of time in the Hague”

  2. Nico says:

    There was no evidence of any murder in the Lucia the case. So what could Lucia have done? Not quite. There were fantasy murders in the sick mind of a group of people. With a group of people worse thing can happen. Be prepared when you meet a group of people.
    Wim Klein calculated the 73rd root of a 500-digit number in 2 minutes and 43 seconds. With his arithmatique he might have killed more insects than Lucia with her magic. In fact he was on a murder place, being the victim.
    Thus quite a nonsense apart fromm the murder of Wim Klein. They never found the murderer.
    See also his farewell show
    A man on his own.

  3. C Edrich says:

    As a long-time friend of Lucia I am really gratified that you’ve reported the way the court ‘artist’ slanted other peoples’ view of her, although I suspect that was a reflection of the entire courts procedures rather than an individual’s decision (I wasn’t there). Richard Gill’s image is indisputably the woman I know and I’d not have known the ‘cartoon’ was of her had it not been for its context.

  4. Beryl Ostting says:

    I’d just like to buck the trend here a bit. I think people may be too focused on the statistics questions here.

    To put it in common-law terms, prosecutors showed De Berk had the means, motive and opportunity to commit the crimes she was convicted of.

    The statistics fall into the category of showing whether the circumstantial evidence presented during trial was enough to convict her beyond a reasonable doubt.

    Obviously the judges felt it was. Her demeanor in court, and total lack of explanation for many incriminating pieces of evidence no doubt contributed to the overall picture.

    Just to point out two, her diary entries shortly after the murders were committed, and other nurses walking in on her while children were dying and she was doing nothing.

    Personally I think she was correctly convicted but improperly sentenced; Judges obviously thought De Berk believed she was an “angel of mercy” for the severely ill children that died; if that’s true, she was badly confused and morally wrong, but certainly not deserving of life in prison without possibility of parole.


  5. Richard Gill says:

    Your comments Beryl Osting are revealing.

    Lucia’s diary entries shortly after the murders were committed were diary entries of 5 years previously when she was working on a ward with terminally ill old people. Of the seven times she wrote in her diary of her secret compulsion, just once it happened to be on the day an old lady had died. This was a ward where people were dying all the time. In fact that particular old lady was the one of the ten cases she was supposed to have had a hand in, which is most indisputibly of all, a completely natural death. The *only* evidence for unnatural is Lucia’s diary entry. This is admitted by the medical experts too! The death would normally have been considered natural, but because Lucia had been there, it is a bit strange. This is what the judges called indisputible medical-scientific proof of unnatural death!! And there are six more such entries on days on which nothing happened at all.

    Two, the weird story about other nurses walking in on her while children were dying and “she was doing nothing”. This is a fantasy story similar to many such stories which were carefully fostered by the prosecution and seized on by the media but without any basis in fact, whatsoever.

    About the courts proof that Lucia’s was lying, there are also two versions. Some journalists in court were impressed by how cleverly the judges tricked her out and got her to say contradictory things, they saw this, as the judges did, as proof of lying. In fact, nobody has ever been able to catch Lucia out saying a lie. Sometimes she recalled events, then heard doctors giving evidence giving a different story, and later she was forced to admit that the doctors must be right (we are talking about events five years previously). Actually further evidence shows that the doctors were mistaken, not Lucia. In fact the judges many times excused doctors from changing their minds about what they saw five or more years ago, while every time Lucia changed her mind they took it as proof that she lied.

    Other people in the court were amazed and disgusted at the way the judges behaved, it was a parody, a sham, a Stalinist show trial; they were stunned and amazed.

    I would recommend, mrs Ostting, that you study Derksen’s book or Knigge’s or Meulenbelt’s report. Do that with an open mind and be careful to distinguish between what we know is based on fact and what is based on rumour, innuendo, gossip.

  6. Richard Gill says:

    PS by the way, at my talk I did briefly show a real picture of Lucia; on the slides on internet you will see a real picture of Uma Thurman.

  7. Beryl Ostting says:

    Hi Gil, I’m not married.

    I would recommend that you re-read Derksen’s book and Knigge’s study: it will increase the volume in your echo chamber.

    For my part, I’ll not tangle with condescending, ego-invested statistics professors. I have my reasons for thinking what I think, as outlined originally, and I’ll just wait and see what the courts decide.


  8. Guys, please keep it civil, all of you.

  9. Nico says:

    Hi Beryl, many scientists of also other disciplines than mathematical statistics, such as physics, are convinced of the innocence of Lucia. Their opinion is based on solid facts, not on feelings that were needed in the beginning of mankind to survive and is characterized by the witchlike picture.
    I saw a posting with your name on Internet, a technology enthusiast living in Amsterdam. Without doubt there is the difference, research and engineering (technology) represent the opposite state of mind.
    Maybe that is the reason you can not appreciate the facts as they are and characterize them in a way which is not scientifically sound.

  10. Marty Hirst says:

    I just hope that those who were so quick to accept the court’s decision that Lucia was guilty as charged will also be able to accept the court’s verdict that she is innocent when they finally get round to spitting those words out. If you have faith in the one you must also believe the other.

    I am, time & time again, totally amazed by the blind faith people still have in this legal system, in spite of all the evidence that blows their reputation right out of the water. This travesty of justice is (finally) about to be put right, not because “they” think it’s the “right thing to do”, but because a few determined people said “NO!! this ain’t right!!”, backed “them” into a corner and forced “them” to act.

    There are many other cases like this out there.
    Wrong place, wrong time, face fits? It could be YOU next time ms.Ostting, I hope there will be people out there who care enough to DO something about it and not just sit back and accept what’s fed to them by the state.

  11. […] Lucia de Berk, the serial killer seemingly convicted on the basis of flawed statistics, received some good news today. Now that her case has been re-opened, the public prosecutor has […]

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

RSS feed for comments on this post. TrackBack URL