Darwin among the Machines

Darwin Among the MachinesWow, I finished it. I feel like a great weight has been lifted from my shoulders.

I don’t normally do this, but before writing this review I checked the rating for George B Dyson’s Darwin among the Machines at Amazon. I’d heard great things about the book and wanted to see if I was just out on my own with my opinion of it. Amazon rating: 4 stars. So yeah, I pretty much am.

But I’m calling the Emperor clause. I believe he has no clothes. The book does have an interesting theme, but that theme is more “a history of computing” than anything to do with “the evolution of global intelligence” (the subtitle of the book). But the basic problem is that while Dyson might be a Very Smart Guy, he doesn’t know how to write and communicate clearly. Seriously, this book was a slog… I read it in 3-4 page chunks (started it back in June) because the style was so awkward it made my head hurt. I’d often have to read a passage several times to figure out what point he was trying to make. Also, Dyson uses a *lot* of quotes. There’re 30 pages of footnotes for the 230 pages text and the quotes tend towards lengthy passages. I’m going to estimate that 70% of the book is quotations. Why is that a problem? Because it means the there’s no unified ‘voice’ to the book. A theory voiced by an individual from the 16th century is going to read very differently from one voiced by a modern individual (not to mention the changes in language over those years). So you’ve constantly got to ‘switch gears’ in your mind as you read.

Here’s a passage, more or less at random:

When the Spanish armada entered the English Channel in July 1588, a network of fire beacons raised the alarm, cradling the newborn Thomas Hobbes with fear. The invention of the telescope in the early seventeenth century extended the distance between relay stations and allowed more complex symbols to be distinguished. The feasibility of a “method of discoursing at a Distance, not by Sound, but by Sight” was addressed by Robert Hooke in a lecture “Shewing a Way how to communicate one’s Mind at great Distances,” delivered to the Royal Society on 21 May 1684. Having advanced the optical instruments of his day, Hooke showed that “’tis possible to convey Intelligence from any one high and eminent Place, to any other that lies in Sight of it, tho’ 30 or 40 Miles distant, in as short a Time almost, as a Man can write what he would have sent, and as suddenly to receive an Answer as he that receives it hath a Mind to return it… Nay, by the Help of three, four or more such eminent Places, visible to each other… ’tis possible to convey Intelligence, almost in a Moment, to twice, thrice, or more Times that Distance, with as great a Certainty as by Writing.”

Robert Hooke (1635-1703) was a brilliant but difficult character whose “temper was Melancholy, Mistrustful and Jealous, which more increas’d upon him with his Years.” Possessed of “indefatigable Genius,” his creative output was astounding, despite ill humor and ill health. “He is of prodigious inventive head,” reported contemporary John Aubrey, adding that “now when I have sayd his Inventive faculty is so great, you cannot imagine his Memory to be excellent, for they are like two Bucketts, as one goes up, the other goes downe. He is certainly the greatest Mechanick this day in the world.”

– Darwin Among the Machines by George B. Dyson, pp 133-134

Again, I just pretty much randomly opened the book and grabbed a passage. You can see the proportion between Dyson’s own words and historic quotations, and you perhaps will wonder what Hooke’s character really has to do with a global intelligence developing like an Orson Scott Card character in the spaces between networked computers. I know I did.

That said…it *is* an interesting book from a historical perspective. I rather wish Dyson had just written a “history of computers and technology” and forgotten about the intelligence aspect. As it stands, I found the book difficult to read and rather unfocused. I never really got the point he was apparently trying to make, in any but the vaguest of ways. He certainly didn’t provide any evidence that would convince me there’s some kind of ‘machine intelligence’ percolating along at the speed of light in our networks. And I *think* that was the point he was trying to make.

2.5/5 stars from me.