I am in this rather pensive mood, and also feeling a bit melancholy, about time passing, and passing so quickly. 2013! How unbelievable; this was a date in the far future, how can it be here now? How amazing, too. Twelve years after 2001, and only six years away from 2019, the year Blade Runner is set in.

The Society of Mind

So, yes, this post is random. I am sitting here in my mum's living room, my daughter lying on one of the sofas, and the dog Marvin on the other, both asleep and snoring. And I will just write about the things that have been on my mind in these holidays, things I thought I could put in a blog post or posts.

I wanted to write about Marvin. Not Marvin the dog, but the person the dog was named after. Marvin Minsky. More precisely, I wanted to write about a book he wrote, called Society of Mind. I read this book in just a few days, whenever I found the time, and till late at night; I somehow felt compelled to read it, after I had rediscovered it, lying on a shelf. My late father who was a big admirer of Minsky's had bought it once. The book was published in 1986, so you could argue whether it is not a bit dated. I don't know anything about the current state of Artificial Intelligence so I couldn't say. Yet I think the ideas in this book can stand on their own. Also, the way the book is written and layed out is interesting in itself. It is written in the form of essays that each span up to one page, in the edition I was reading. And the layout and typography look quite modern too. But these are just factors that support the content.

Easy is the hardest thing

So what is it about? Luckily there is a good wikipedia entry about the Society of Mind theory, so I don't have to explain it all. I like this description of the book: "It is a collection of ideas about how the mind and thinking work on the conceptual level". Yes, there are lots of theories and models, there are no proofs about how things work in reality. And yet these ideas don't seem far-fetched, actually many seem very plausible (although I don't believe at all in his explanation of foreign accents! - he suggests you lose the ability to learn the precise phonetics of a language at puberty, so there's no risk of you picking up the phonetics of your child's baby language. Hmm) The reason why they are quite plausible is that they stem from attempts to build machines that have some abilities of the human mind, starting with just very "basic" ones, things that children learn. One fascinating conclusion in the book is that it is actually much more difficult to make a machine do something we regard as basic, than things that require higher mathematics and logic. It is easier to build a chess computer than a robot that carries out actions that we learn as little children. We think of things as basic because we were not conscious of learning them, and always had them available by the time we had learned to think. Minsky is disecting the processes that it takes to, for example, build a tower from building blocks. It really is quite complicated if you think about it.

What brains do

The book touches on theories of child development, psychology, and of course programming in its aim to explain how thinking and perceiving might work, and how consciousnes might arise. Minsky at one point sums up his findings with "Minds are simply what brains do". There is no "hidden ingredient", not what we think of as a soul. This also means there isn't really a free will. Everything that happens in our mind, every decision we take, is a consequence of what we have learnt plus random events. This is not what we like to hear, and in fact we need the illusion that there is a "me", an individuum that is "in control", constant and immutable. We need it for our mind to function properly. For my part, I find this all makes sense, and it echoes things I have read about, or thought before. Not that I'd ask myself these things all that often.

I wouldn't even say I am generally that interested in theories of consciousness, I used to be a lot more. But I have returned to this interest for a few days and it was a fascinating excursion. And I think a lot of the ideas from that book will stick in my mind, just mostly unconscious, but some of them might pop up here and there.

As one last thought, of course another consequence of these ideas, if they were right, would be that we could in fact build thinking machines, given we find out enough about how the subprocesses work that together make up the work of our minds..

I would have liked to write even more about some implications of the book's ideas, it is all coming back to me now, but I have to leave it here. I did not think I'd write so much anyway.

It is all good, as probably my most important short/medium-term goal for now is to write more. And it doesn't matter for the moment what it is about. It doesn't matter that the start of this post has no apparent connection to the rest of it (indirectly it has because it's somewhat about the mind too, I guess). I have a feeling there might be a number of random posts on here, about things that just pass through my mind - actually that's what a lot of blogs do, isn't it. It's a good exercise for getting things from your unconscious before your very eyes, and doing it "publicly" provides some perhaps necessary cohesion and format. And I don't need to feel guilty about writing it, as I am not urging anybody to read it. In fact not many will read it, but some might, and that will be just the right people.