Gill’s first talk: Lucia: various documents and a wikipedia link

Tomorrow (Tuesday Nov. 25) evening I plan to talk about the Lucia case, arranging my contribution around three topics, which I hope will all lead to spirited discussion.  There is also a fourth topic which we can look at “if time permits”.

[Remark added a week later: I have added a fifth topic, see Appendix 3 below]

Topic (1). A very recent paper in criminologyIn Search of the ‘Angels of Death’: Conceptualising the Contemporary Nurse Healthcare Serial Killer” by Elisabeth Yardley and David Wilson, Journal of Investigative Psychology and Offender Profiling J. Investig. Psych. Offender Profil. (2014) Published online in Wiley Online Library ( DOI: 10.1002/jip.1434

Topic (2). Some basic concepts from probability and statistics, in particular the couple “Sensitivity and specificity“. I will also mention p-values and likelihood ratios. The wikipedia article “Sensitivity and specificity” has a nice numerical example on the fun topic of bowel cancer, highly relevant now that a huge screening programme has been started in the Netherlands (link and some numbers, below, Appendix 2). I will explain why (act of civil disobedience!) I did *not* submit a sample of my own shit …

Topic (3). My TEDx Antwerp 2014 talkmurder by numbers“.

I am not sure yet what the sequence will be.

In earlier posts I already uploaded the TEDx slides and the Yardley and Wilson paper.

Below (Appendix 1) also a very recent (confidential, draft) paper on the Lucia case by Utrecht philosopher Herman Philipse Herman asked me for comments, so *your* comments will find his way to him …. please read and comment! And we can also discuss!


Appendix 1

Topic (4). Legal Reasoning:   Some Problems of Probability, Illustrated by the Coincidence Argument in the Case of Lucia de Berk

Herman Philipse

Philipse 2014 (draft, confidential)


Which logic of probability should be applied with regard to factual hypotheses in criminal cases? In this article, I discuss two possible logical reconstructions of the so-called Coincidence Argument, which played a crucial role in the conviction of Lucia de Berk by the Court of Appeal of The Hague in 2004. If the argument is construed as an instance of the Law of Likelihood, nothing follows with regard to the probability that Lucia was a serial killer. If, however, the Argument from Coincidence may be interpreted charitably as an instance of Bayesian updating, the Court of The Hague did not fathom the diversity of data needed in order to make it valid. Since there is no better logical reconstruction of the Coincidence Argument, I conclude that law faculties should teach probability logic (inter alia!) during their courses on legal reasoning.


Appendix 2

Some material on Topic (2). Some basic concepts from probability and statistics: Sensitivity and specificity

Sensitivity = true positive rate = rate that true positives are correctly classified positive
Specificity = true negative rate = rate that true negatives are correctly classified negative

Prevalence = background rate of true positives

eg, bowel cancer example from wikipedia; a short R script

sens <- 67/100          # 67%
spec <- 91/100          # 91%
prev <- 1.48/100        # 1.48%

1000 * prev * sens              # 9.916

1000 * prev * (1 – sens)         # 4.884

1000 * (1 – prev) * spec         # 896.532

1000 * (1 – prev) * (1 – spec)        # 88.668


Appendix 3

Topic (5) Lessen van Lucia

An attempt (by me, in Dutch) to understand what actually happened at JKZ in the months preceding Amber’s death.

This document is controversial and confidential.

Print Friendly, PDF & Email

One thought on “Gill’s first talk: Lucia: various documents and a wikipedia link

  1. Here is a link to an attempt by me (in Dutch) to understand what went on at the Juliana Children’s Hospital in the run-up to the Lucia case.

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title="" rel=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>