Monday, September 10, 2012

Hat trick - finale

Years went by, and I passed through a series of jobs in information technology, with less and less enthusiasm. The applications of computer science had never excited me.

The story of the Wise Men had already been forgotten, and Knowledge-based systems were no longer hot and sexy. Jan Treur's Artficial Intelligence group had quietly moving into the next biggest thing: Agent Systems, without too many people realising that the contribution to the field of knowledge-based systems had been minimal, from a theoretical, practical, scientific or any other point of view. In fact, the old papers on knowledge-based systems were reused, and sold again as new results on agent technology, which was not too surprising, given their overall vagueness. To be honest, I did not really care, until a former colleague brought the following to my attention:
Brazier, F.M.T and Treur, J., Compositional Modelling of Reflective Agents; Intl. Journal of Human-Computer Studies 50:5 (1999)
The paper is predictable agent systems versions of our earlier paper on logical methods in protocol analysis, and more than 6 pages are literally copied from the original, without proper reference. For Treur, this fits in a long tradition of selling his own results more than once, but for Brazier, this is a pathetic case of stealing research. In both cases, it is plagiarism, or scientific fraud.

Somewhat naieve, I thought the issue would be quickly sorted out through one or two letters to the responsible committee within the Faculty of Sciences of the VU University. However, I had underestimated the clamshell nature of the department.

To cut a long story short, over a period of more than one and a half year, a grand total of 10 professors, including prof. dr. J.W. Klop, director of the Division Mathematics and Computer Science, prof. dr. W Hogervorst, dean of the Faculty of Sciences, and prof. dr. H.E. Bal, chair of the Scientific Council of the department, were involved in this case. Most of them did not seem to really care, and none of them had the guts to openly express their opinion on the problem. I consulted prof. dr. J.A. Bergstra, and he advised me to consult a wise man for a confidential advice that would be binding for all involved. This was a mistake; till the present day I doubt Bergstra's role on this, as he appeared to be more closely involved with Treur than he had admitted.

I feel it is mainly the requirement of confidentiality that does not work in delicate cases like this. Any authority will think twice before publishing an wishy-washy opinion on scientific fraud: there is a reason why court cases are public. Anyways, as soon as I realized the expert chosen, prof. dr. J.-J. Ch. Meyer, had closely collaborated with Treur in the recent past, it was too late. The binding advice consisted of some loosely formulated opinions on the matter, without any conclusion. The wise man wore a black hat.

The case of the Wise Men was my main motivation for turning my back on science. In my opinion, modern scientists fall into four categories, which are best summarized in Dutch: "onderzoekers, onderwijzers, ondernemers en onderkruipers". That is: researchers, teachers, enterpreneurs, and rats. Over the past decades, teachers have been quietly replaced by their competitors, and it has become impossible to make a career out of teaching; a sad development pushed forward by a government who believes that science is a short-term investment. Those thriving on this new wave are competitors who know how to play the games of money and media, and while I have no problem at all with scientists who popularize or monetize their research, I do have big issues with the rats who lost the connection with science. 

Issues with modern science are leveraged by the clamshell nature of traditional universities, who are fighting tooth and nail to keep up appearances, covering up cases of scientific misconduct until bubbles burst. Well-known Dutch examples of Anthonie Stolk, René DiekstraDiederik Stapel, and Dirk Smeesters,  show an introvert culture where scientific fraud pays and is not always easy to expose. I believe it is all too easy to burn a man like Stapel at the stake, with his colleagues, co-authors, employers, PhDs and publishers watching. For more than a decade, the community had loved to hang around with the man with golden hands, ignoring signals that his performance was too good to be true. It takes a community to create a fraud.

The Department of Mathematics and Computer Science proved to be an excellent environment to make scientific career, at the cost of others. The 'Tweede Fase', once intended as a plan to educate young researchers, eventually degenerated into a system that abuses ambitious young researchers as cheap labour forces, kept quiet with empty promises, all too willing to pay with money and energy for the ambitions of those who are a few steps higher on the ladder towards scientific mastership. The Ponzi scheme never worked for me, while it paid the Braziers and Treur of the community called 'science' really well.

Till the present day, this case has never shown up in any of the official reports of the Faculty of Science of the Vrije Universiteit Amsterdam. Taking into account that the issue was reported to the 'Wetenschapscommissie' of the Faculty, it once more confirms my suspicion that effort was taken to cover the whole issue up. The current Faculty of Science seems to be as corrupt as the old Department of Mathematics and Computer Science.

I do not really regret my leaving. After all, I am still wearing my white hat.

3 comments:

@notpicnic said...

http://pearltrees.com/p/cRoiL

Anonymous said...

Do you know of any other cases of scientific misconduct in the Computer Science Department? A co-worker told me of one case of data-fraud in a PhD thesis, and a plagiarized Master's thesis, both from the time frame you were working there?

Izak van Langevelde said...

I remember a PhD defence which was forced through, although one of the members of the reading committee had serious objections for reasons of sloppy statistics. However, the date had been set, relatives of the candidate had bought tickets, and the department had already embraced her as a staff member, so the opponent was overruled. The plagiarized Master's thesis is from a later date, and I have only indirect information about a plagiarized PhD thesis.