Complexity: A Guided Tour
The best single volume introduction to complexity science, and the orientation this trail otherwise lacks: seventeen primary sources with no synthesis between them. Mitchell writes from inside the mainline rather than adjacent to it. Her doctorate was supervised by Douglas Hofstadter and John Holland, whose 1992 work is a waypoint here, and she has been at the Santa Fe Institute throughout, so the book is a practitioner's account of her own field rather than a journalist's tour of it. It covers the formal core in sequence: dynamics and chaos, information and entropy, computation, evolution and genetic algorithms, then networks and scaling, before turning to the harder question of whether complexity science constitutes a unified field at all. That last move is why it earns a place here rather than merely being useful. Mitchell devotes a chapter to the problem of defining complexity, surveys the competing measures on offer, and concludes that no single definition commands agreement, which she treats as an honest description of a young field rather than a defect to be argued away. For a corpus that keeps meeting confident claims about what complexity really is, that is the useful posture, and it comes from someone with every credential to claim otherwise. Read it before or alongside the ascent. Where Weaver poses the problem and Anderson states the principle, Mitchell supplies the connective tissue: what a fitness landscape is, why scaling laws matter, how the pieces relate, at a level that assumes curiosity rather than mathematics. The honest limitation is that it is a survey, not a contribution, and it stops in 2009, before the network science and machine learning developments of the following decade. Tom recommends it in the corpus's own reading list, and it remains the book to hand someone who asks where to start.
Save this paper to a bibliography, keep notes on it, and export citations. Create a free account →