I'm a big fan of this whole "Very Short Introduction" series. I have several, mostly in Kindle but this one snuck it's way into my pile at a bookstore. Most are 4 and really do give a quick survey of a whole field, so I usually read if I'm starting to study or my knowledge is old. This one is quite similar but it misses several aspects of Complex Adaptive Systems (CAS) that I'm already aware of, while going into depth on others.

I re-read it and dropped my rating considerably. A more critical reading found a considerable number of important error. I had skimmed over the very non-standard definition of an ecological niche (he thinks the human body is a single niche for organisms that inhabit it), unsupported generalisations (e.g. all complex adaptive systems become more complex, while in fact the opposite happens in many cases, e.g. parasites becoming simpler than the free-living organisms they evolved from), category mistakes (it's not individual systems that necessarily become simpler or more complex but evolutionary lines (whether that is an animal vs a species or a town market versus a computerized commodity market).

Holland explains the big division in Complexity studies between CAS complexity and complex physical systems such as the atmosphere but then spends very little time on the latter. This was fine by me as today I'm interested in CAS. But then, maybe it would have been 5 if he'd at least listed some of the other approaches to CAS at the same level of detail as physical systems. For example he does not even mention Bayesian Networks . Now I think of it, his whole approach is constructive (building CAS from bottom up) with little mention of analytic (understanding CAS top down via learning models).

There is a small amount of mathematics in the book, but you don't need to know how to work it. For example, it is useful to know what a partial differential equation is, but you don't need to know how to construct one, let alone solve it.