I’m interested in the nature of consciousness. It’s a pretty weird phenomenon, right? And if you’re going to be intellectually curious about something, you might as well start with the capacity to reflect on intellectual curiosity. Etc. And after studying the topic of consciousness a bit I have come to believe in the idea that human minds are composed of multifarious cognitive actors. I don’t get proscriptive about how many or what their natures are, just that there is more to a human mind than can be adequately modeled as a singular entity.
When I heard about Marvin Minsky’s book The Society Of Mind it seemed like something I would find interesting. What is interesting to me is that I did not! Minsky is a well-known computer scientist famous for half a century of work with the MIT AI lab. But this book… I just didn’t like it.
Seriously it’s like some video game designer from the 1980s attended an introductory philosophy course, went home every day after class, smoked a ton of weed, and dictated stream of consciousness about how it relates to his day job; then a year later when he’d stopped smoking weed, he started an amphetamine habit which enabled him to scrupulously organize all of the previous material. For example, it was full of weird prose.
…we learn to gain our ends by sending signals to those great machines, much as the sorcerers of older times used rituals to cast their spells.
How’s that for a bizarrely bad analogy?
After reading enough bits like, "Our thoughts about our mind-experiments are mind-experiments themselves," I started to get the feeling that Minsky was desperately trying to play whatever game Douglas Hofstadter was playing in GEB (a book that I’ve been slowly reading for several years now). This seems confirmed by the prominent Hofstadter blurb on the back cover.
But this is no GEB. Not even close. It’s not interesting or entertaining or well crafted. There is a lot of fatuous nonsense disguised as deep wisdom. "The less there is in a treasure chest, the more you’ll be able to put in it." You’ll be able to put a lot in this one!
Minsky’s theory of humor is terrible and deserves no further comment. His definition of intelligence is weird and not in an interesting way. He defines intelligence: the mental problem solving processes we don’t understand. Seriously. I’m concisely paraphrasing from page 71 and that just leaves a lot to be desired. It does give me confidence to continue assuming that everyone who ever mentions AI (including me) doesn’t know what they’re talking about because they’re unlikely to be able to provide a philosophically sensible definition of just the "I".
I have seen praise for the copious epigraphs which can be found on almost every other page. Buddha, Proust, Webster’s dictionary. I found them mostly bewildering and uninteresting. There were some exceptions.
What we call a mind is nothing but a heap or collection of different perceptions, united together by certain relations and suppos’d, tho' falsely, to be endow’d with a perfect simplicity and identity.
That was David Hume reminding us that simply reading Hume would be more interesting and instructive. What’s new?
The tragedy of this is that I don’t think Minsky is wrong to emphasize collections of cognitive agents. It’s just that this book doesn’t help say anything more than its title.