Solving the frame problem

Murray Shanahan has been actively involved in research on the frame problem since the late eighties. When a scientist looks back at the past of his eld, including his own early work, he can see how poorly the subject was understood years ago, and then two attitudes are possible. One is to think: It's amazing how naive and confused we used to be; much of this work should have never appeared in print, and the rest should have been expressed very diierently! Or he can adopt a wiser view of the history of his eld as a reasonable|maybe necessary|path towards the better understanding that is available today. This respect for the past, with all its embarassing mistakes, is an attractive quality of Shanahan's book. He believes that the history of attempts to solve the frame problem isn't an arbitrary record of failures which culminates in the discovery of a truth which, in hindsight, seems obvious. Rather, it seems to me to reeect a natural order of enquiry. The dead ends that people have investigated are much more than mistakes. They're obvious possibilities, which any thinker engaged in studying the frame problem is wont to investigate. A deep understanding of the knowledge representation issues surrounding the frame problem can only be acquired by getting a feel for the space of mathematical possibilities and where they lead. p. xv] The book is about the history of research on the frame problem as much as about our current state of knowledge. This may be the only way a good book can be written today on a subject as volatile as the theory of commonsense knowledge and reasoning.