The Information: A History, a Theory, a Flood by James Gleick
My rating: 5 of 5 stars
Loved this book. This contained everything I could ask for in popular science: a very engaging prose style and a bunch of interesting new material about topics that I love, all woven into a compelling narrative across a very broad canvas illustrated in depth with characterful anecdotes and curious asides. Even topics where I thought I was already well-informed were brought into a new light.
Gleick starts with the earliest forms of non-verbal communication – how writing evolved, and how that may in itself have catalyzed changes in the way humans actually think. He discusses communication over distances by a great discussion of talking drums, and that alone was practically worth the price of admission. I had no idea how they actually worked, and the setup of how they mimic the rising and falling tones of the tonal tribal languages, but need much more florid phrases to disambiguate what they are saying because they have tone alone to work with, was brilliant.
Next, we are introduced to the very first dictionaries, and what an odd thing they are if you’ve never seen one. Alphabetical order, for example, is an alien concept that we just take for granted – just think how pictographic languages might (not) work with that. And which new words deserve to be included? Again, not obvious concepts.
From here we move to the core concepts of numbers, via Charles Babbage and his engines of computation. This was familiar territory to me, but still I found new insights and interest, not insignificantly down to just the sheer depth of the material being presented. Once we have numbers, we get to codes, and from there to how you might code a language – Samuel Morse makes a great appearance here.
Now comes the more mathematical chapters, as Claude Shannon tries to work out what information actually is, driven by the desire of his employers (Bell corporation) to cram more information, in the form of morse, or other, coded language down a noisy channel. This expands into a much broader section that’s effectively about mathematics as a language, which touches on Godel’s incompleteness theorem and builds up towards Maxwell’s demon and how information and energy are in some intriguing mathematical ways directly interchangeable.
I’ve seen all of this before, and what worries me still is that Shannon’s definition of information doesn’t correspond very well to our natural, every-day understanding of information. This means that the mapping back from conclusions drawn from information theory back to our every-day world may give us some warped conclusions. The concept of logical depth, from Charles Bennett, fits this more closely, and gets a brief mention at the end of these chapters, but could have done with more coverage for me.
No modern book about information can avoid touching on Dawkins’ concept of the meme, and Gleick covers this in the concluding chapters. He builds from DNA as a code up to this, and here I might criticise his lack of any discussion of epigenetic effects: we now know that DNA isn’t simply the code for life, but that there are a whole number of other things that affect for example what parts of the genome are expressed, and these often come from outside of the chromosomes themselves. Examples abound, but include the way in which many reptiles will grow out of an egg as one gender or another depending on the temperature of incubation, or how a regular bee will become a queen bee if fed the right chemicals (in the form of royal jelly). None of this was discussed, but it could have been incorporated – the person that reads a text gets different meanings depending on what it understands about the world already, after all.
Overall my criticisms are minor in the face of a great book – five stars for sure, and one to keep for a re-read another day!
