STRUCTURE AND CHANGE IN COMPLEX SYSTEMS by Wallace H. Provost Jr.

INTRODUCTION As thought ascends to ever higher levels, previously valid premises must yield to more comprehensive points of view, . William N. Johnston In 1958 I moved my family to a small village in the Berkshires, convinced that the danger of a nuclear holocaust was too great to chance remaining in the city. In the 1960s everyone was convinced that early in the twenty-first century the population of the world would reach the point where people would be standing on each others shoulders. Now, as I am writing this in the 1980s it appears that our crystal balls have turned into pits of despair. Even the most optimistic projections fail to present a reasonable course that will carry the human race into the twenty- first century without major cultural upheaval, In fact, aside from a few political dreamers, I don't even find anyone predicting 1983 except as an economic disaster area. Being an optimist, it is my strong feeling that the reasons for all of this negative reaction to the course of human history is our lack of understanding of human culture and human cultural change. In my search for this understanding, I find that it is likely that our quandary is deeper than just social and cultural systems, that what we seem to lack is a good understanding of complex systems in general. . Norbert Wiener said in 1950 that within ten to twenty-five years chess machines would turn chess into a simple parlor exercise. Twenty-five years of development have simply shown that computers are at best mediocre chess players, barely the level of a good novice. There are other things computers have not proven themselves the masters of. In the 1950s TRW began experimenting with a computer controlled cement kiln. At the same time the Japanese began an even more ambitious project. Now, some thirty-five years later, while computer operated control loops are common, and both robots and computerized assembly lines are becoming more popular every year, no one has successfully operated a complex industrial process totally with a computer, Repetitive simple processes are the forte of the digital computer. We just do not understand complexity enough to develop an algorithm for controlling such processes with the kind of simple logic we have taught our computers to respond to.If the human race survives the twentieth century, and I am sure it will, the understanding we develop of complex systems will play an important part in helping it come about.