Chairman's impression of the "Evolution of Complexity" Symposium

I believe the whole conference and the symposium in particular were a great success (probably the best one, together with the 1991 Workshop, that was organized by the Principia Cybernetica Project until now).

Although the conference was started much too late for such an ambitious program, in the end it managed to appear well-organized. Of course, many contributing speakers did not turn up (only one of the invited speakers, J. Jaynes was absent, but got replaced by R. Petrella), mostly because they did not manage to arrange the funding or the journey in the short time that was left between the acceptance of their abstract and/or scholarship request and the start of the conference. But this was not felt as a great problem because of the large number of speakers (some 200 in total). A greater annoyance, for me at least, was the lack of time: 20 minutes per lecturer is really too little for a good presentation and discussion.

The "Evolution of Complexity" symposium was kept tighter in hand by me than the other "streams" (except the "Quantum Structures"), which lacked a chairman responsible for the whole. Thus, it seems I managed to have less delays and to better fill up the holes from the people who were unexpectedly absent (Robert Glueck, A. W. Raifu, S. Norvaisas, P. Stokes) by extending discussions or coffee pauses, by adding new speakers (R. Cottam appeared a little bit out of the blue at the last possible moment), and by letting one speaker lecture twice (thanks for the additional effort, Richard Coren!). The final programme, with all abstracts (and full papers if submitted) is available at

The most frustrating part of the chairman role was that I had to continuously look at my watch and remind speakers when they had only 5 minutes left (usually after they had just introduced the subject domain), in order to ensure that the next speaker could start in time. Thus, I was not really in the best position to concentrate on the content of the lectures. Yet, I got the impression that the general quality of the research and the presentations was quite high. I heard from others who had also been in different streams that our Symposium had generally a higher quality.

The attendance was quite satisfactory too. I estimate that on average some 50-60 people were present, with a high point of 80-90 during the final panel discussion. The discussions concluding each of the 3 sessions were clearly a good idea, and engendered many enthusiastic debates. Again, it was a pity that we had to stop in order to leave enough time for people to get a lunch, because it seemed as though we might have continued for at least an hour each time.

Especially the final panel discussion was quite lively. The panel members were F. Heylighen, C. Joslyn, F. Geyer, B. Edmonds and John D. Collier. Chris Langton, from Alife fame, had originally agreed to participate, but he afterwards apologized for not getting out of his bed in time after a long night discussion ;-) After each panel member had given his own impression of the general "Evolution of Complexity" domain in 5 minutes, and replied to the others' comments, someone in the room got so eager to widen the discussion that he nervously interrupted us with the plea to allow the public to participate (which was supposed to happen anyway at that time). Perhaps 20 or 30 people intervened in the subsequent hour, and it was again lunch time which forced us to cut short the debate.

As any insider would have predicted, there were no clear conclusions, except that the topic is very interesting but needs a lot of additional research. Most obviously lacking is a general definition of complexity. Yet, there seemed to be at least an implicit consensus on what are the most important issues to be resolved: definitions, mechanisms underlying metasystem transitions and growth of complexity... That is something which is rarely achieved in wide-ranging domains such as cybernetics and systems theory. As such, the symposium seems a good starting point for an on-going collaborative effort, which would involve most of the partipating lecturers.

After the conference, I had some long discussions with my colleague Principia Cybernetica editor, Cliff Joslyn (the third editor, Val Turchin, unfortunately was unable to join us). We agreed about several practical issues concerning the further organization of the project. One of these was to create a new medium for people wishing to keep in touch with the project's activities: an announcements mailing list. This list would complement our existing email discussion list PRNCYB-L, by only broadcasting the most important news items, without on-going discussions. Thus, it would carry much less messages than PRNCYB-L (which fluctuates around 1-2 messages per day on average), which would make life easier for people with limited time or email capabilities, who don't wish to get lots of messages in their mailbox. On the other hand, all messages sent to the new announcements list would be automatically broadcasted to PRNCYB-L, so that there would be no need for double subscription, and so that people who wish to discuss announcements can still do so.

I checked with the VUB computer center, and they told me there is no problem in creating such a list. I expect to have it running in a few weeks, and will make all practical details about subscription available by then. In the meantime, symposium participants are invited to subscribe to PRNCYB-L (see if they wish to participate in on-going discussions, and/or read the Principia Cybernetica Web ( if they wish to get information in their own tempo.

Another way to build on the work done at the symposium is to publish a Proceedings with papers by the symposium contributors. The conference organizers are presently discussing possibilities with some international publishers (Kluwer in the first place). I cannot say anything more concrete as yet, but I hope there will be one or more proceedings volumes for the conference. Ideally, I would like to have a separate volume for the "Evolution of Complexity" symposium, though we might also include some contributions from other conference streams. In the meantime, I am collecting the texts of the presented papers ( I have already 5 or 6), and making them available electronically over the web ( I would welcome any further submissions.

Note however, that these texts are likely to be quite different from the eventual published papers, in the sense that the final versions will need to follow as yet unspecified guidelines about length, format, etc. and may undergo further refereeing. For the published volume, quality of the papers and coherence of the whole are for me the most important criteria. This means I would not mind including papers from people who were not present at the symposium, but whose abstract was selected and fits into the theme. I would also appreciate it if different contributors would read each other's papers, and use the ideas they get from other to try to develop a more coherent whole. That is why I would like to get draft versions available over the Web now, as a first stage for the final publication.

The "Einstein meets Magritte" conference was meant to introduce the transdisciplinary Center "Leo Apostel" (CLEA), which was responsible for the overall organization, on the international scene and to a wider audience within the Free University of Brussels (VUB). It seems to have succeeded very well on both counts. Several people (including K. Diller and B. Martens) have asked me about possibilities to spend a sabbatical or a visit at CLEA, and people from other research centers (e.g. the Center for Theoretical Studies in Prague led by I. Havel) have expressed their interest in starting a formal collaboration. I will discuss these possibilities at the next board meeting of CLEA (of which I have now officially become an "associate director").

The very positive response within the VUB seems to indicate that CLEA will be able to rely on a lot of good-will created by the conference, making it likely we will get sufficient funding and administrative support for such collaborative projects. The most pressing practical problem at the moment is lack of office space, but there are promises to arrange a complete building for CLEA. I will also propose a formal association between CLEA and the Principia Cybernetica Project, and perhaps suggest individual associations between CLEA and the PCP editors Cliff Joslyn and Val Turchin.

Let me conclude by reiterating my wish that the symposium contributors who felt really involved with the subject would keep in touch with us at the Principia Cybernetica Project, either through one of the mentioned mailing lists or at least through the Principia Cybernetica Web.