Two interesting books that I picked up in the last week. Obliquity, by John Kay (former Director of the Institute of Fiscal Studies) is about "why our goals are best achieved indirectly", why it is often best to take a step backwards in order to move forwards, why we should not abandon accumulated knowledge too quickly and think we can recreate everything from first principles.
Examples include the folly of the "first principles" thinking of the tower block designers of the 1960s which quickly dehumanised those living within them, to the search for the shortest route across Central America (not looking for short east-west routes but stumbling across the oblique Panama cut), disorganised market systems (a triumph of obliquity) and the success of businesses that prioritise the building of meaning through a business (hard to quantify) versus businesses that prioritise the maximisation of shareholder value (they often fail because they pick on only one part of the multiple objective function).
The book is a warning not to treat models as a substitute for judgement and experience. My favourite reference in the book is to a "Professor of Decision Sciences" who is headhunted by a prestigious University. He asks his colleague for advice--should he take the post?. You, of all people should be able to do without my advice, says his colleague. The Professor of Decision Sciences says "Don't be silly, this is serious".
The book is all about the need for feedback loops, adaptation, evolution (see Owen Barder's column on this) and the proposition that direct (as opposed to indirect) action makes sense only if you are clear about higher level goals and you have sufficient knowledge and control of the systems that achievement is based on (and that does not often characterise social and economic change).
Systemic Action Research by my colleague Danny Burns (2010) covers similar territory (also using a high-rise tower block example as did Kay) but is a more academic book, covering the research methods for making sense of a complex and uncertain world. "Sense making is often about making creating a whole out of fragments. But it is about more than juxtaposing and arbitrarily linking them. It is about finding the patterns that connect them and constructing a meaningful narrative to hold them". The book describes different types of action research ("coming to know") and resonates with Kay's position of "we learn about the structure of a problem by the process of solving it".
These books do not reflect the mainstream of development thinking, although this is changing. Feedback loops, constant adaption, lateral systems thinking -- these things don't fit terribly well into the campaigning, 3-year turnaround, silo-ed world of current development assistance (and research). As one who was schooled in models, linearity, and direct problem solving, but has spent the last 20 years trying to blend the two approaches --direct and indirect--more seamlessly, I can recommend both of these books.