Skip to main content

Posts about book-reviews (old posts, page 16)

Thinking, Fast and Slow

Thinking, Fast and Slow

Daniel Kahneman


I'm slightly conflicted about this book (and I'm writing this review without having quite finished it). Some of the insights are quite astounding, and the basic premise -- that people are poor decision-makers, and that this basic fact needs to be reflected in all our social, economic and scientific systems -- is one that's well worth our taking on board.

As a scientist, I spend time working within and developing systems of enquiry. There's a temptation to think that good people working with goodwill will generate accurate, honest and balanced results. The message from this book is that this isn't at all correct: ability and integrity are insufficient, and we need to design our systems with this in mind. What this means for practical scientific enquiry is unclear. For example, is peer review the best system for assessing research results? How about for research funding proposals? Does bibliometrics a better rear-view mirror of quality than human assessment? All these are valid questions thrown-up by a more behaviourist perspective on human enterprise.

However, I did feel that the book itself is very laboured. In fact I feel the same way about it as I felt about books like The End of History and the Last Man and The Tipping Point: How Little Things Can Make a Big Difference: that they're basically essays that have been extended to book form, and struggle to maintain that length. A more concise presentation would have omitted a lot of supporting evidence, obviously, but whether that would weaken the appeal of a popular science book is questionable, and more exploration of the consequences and potential mitigations of the science would perhaps have made for a better read.

It's also slightly annoying that only American universities seem worthy of being named: research done elsewhere is attributed to "a British university" or "a German university". I know the author is at a US institution, but science is an international endeavour that deserves better treatment.

3/5. Finished 14 November 2012.

(Originally published on Goodreads.)

Gödel, Escher, Bach: An Eternal Golden Braid

Gödel, Escher, Bach: An Eternal Golden Braid

Douglas R. Hofstadter


This is a book that's either loved or hated. It has a cult following amongst mathematicians and computer scientists: Hofstadter himself is a respected computer scientist with an interest in biological modelling.

The "plot", such as there is, revolves around explaining Godel's incompleteness theorem: the idea that any formal mathematical system has inherent limitations and inconsistencies. Godel incompleteness is to mathematics what Turing computability is to computing: a bound on ambition, a constraint that prevents some problems from being explored. The former perhaps has less significance to everyday life than the latter, but still remains one of the cornerstone discoveries of 20th century science.

Hofstadter also zones in on the familiar discussions about art and mathematics, the appeal that recursive, self-reflecting art works (like those of Escher and Bach) have to the mathematically inclined. The weaving of these themes within the book is quite astonishing, and at times illuminating. However, it does mean that this is not a book one can dip into: it requires concerted and prolonged effort, which it only partially repays.

The scope is both impressive wide and restrictively narrow, with the reader emerging with an understanding of a problem whose everyday relevance can be questioned, but also with an exposure to a wide range of aesthetic and scientific problems that he might otherwise never consider.

3/5. Finished 11 November 2012.

(Originally published on Goodreads.)