Review of Marvin Minsky's Society of Mind

<h3> Introduction </h3> This is a small review of <a href="http://www.amazon.com/Society-Mind-Marvin-Minsky/dp/0671657135">Society of Mind(SoM)</a>. I urge anyone who is looking for a more in depth analysis to check Push Singh (R.I.P) <a href="http://web.media.mit.edu/~push/ExaminingSOM.html">review</a> or just to read the book itself. It is the magnum opus of Marvin Minsky, and it is more a philosophical and psychological venture into the nature of the mind than a technical AI work. There are very few works around like this one. It basically sets a very abstract framework of how he thinks the mind works. The ground that is covered in the book is staggering, and I think it is one of the reasons why this book is not as famous and cited as it should. It is the culmination of the ideas of two genius, Minsky and Papert, on how the mind structure itself, the world, time and beyond. The structure of the book is also something very uncommon. It is organized in several mini ideas, one per page, and those ideas are themselves organized in larger modules or chapters. So Minsky used his society idea for the structure of the book itself. What that results is there are technical aspects that are not immediately clear, but run through the book, much like a though. So, it is a difficult book to understand, but one that is easy to read I guess. I'm going to try and give some insights from several quotes I have gathered in my path in the book. It is very disconnected and floating around, but the book is in this format, and I for sure lack the competence to formulate a coherent string of though were Minsky himself failed to. Like he says: <blockquote> My explanations rarely go in neat, straight lines from start to end. I wish I could have lined them up so that you could climb straight to the top, by mental stair-steps, one by one. Instead they're tied in tangled webs. </blockquote> Well, I'll use his justification also: <blockquote> I'm inclined to lay the blame upon the nature of the mind: much of its power seems to stem from just the messy ways its agents cross-connect. If so, that complication can't be helped; it's only what we must expect from evolution's countless tricks. </blockquote> :) <h3> The book </h3> The book provides several kind of falsifiable (perhaps some are too abstract) theories and hypothesis floating around the book, but don't worry, Minsky is well aware of how science marches <blockquote> The only course left for us is to study the mind the way scientists do when something is too large or too small to see - by building theories based on evidence. </blockquote> That is not what the book is about of course, like I said before, it provides a framework, or several frameworks and hypothesis, to begin building experiments and improving the hypothesis and frameworks(or destroying them). There is a lot of talk about what things mean, and how meaning is merely the connections that one does between scarcely existing things. <blockquote> The secret of what anything means to us depends on how we've connected it to all the other things we know. That's why it's almost always wrong to seek the ''real meaning'' of anything. A thing with just one meaning has scarcely any meaning at all. </blockquote> This connectionism is keenly related to Papert's principle and to how people learn (the ideas that grew into this book originated from Papert and Minsky dialogues, but they went into separated directions, Papert went into pedagogy and the applications of these ideas into constructionism, Minsky remained interested in the construction of an intelligent machine). I disagree with what Minsky's concept of intelligence. <blockquote> Our minds contain processes that enable us to solve problems that we consider difficult. ''Intelligence'' is the name for whichever of those processes we don't yet understand. </blockquote> Then he goes and adds <blockquote> Like the concept of ''the unexplored regions of Africa'', it disappears as soon as we discover it. </blockquote> I think that our concept of intelligence is flexible and mutable, but it is not bound to disappear, it will just change with our perception of the world. There is also several Buddhist quotes, mostly about the nature of suffering (attachment). Since I identify a lot with those though, I'll cite one of them: <blockquote> Do not become attached to the things you like, do not maintain aversion to the things you dislike. Sorrow, fear and bondage come from one's likes and dislikes. </blockquote> The most interesting part for in was the description of what is called Papert's principle: <blockquote> Some of the most crucial steps in mental growth are based not simply on acquiring new skills, but on acquiring new administrative was to use what one already knows. </blockquote> I think the development of this principle is one of the aspects of the book who is most valuable in a profound personal way. It is, as Papert would say, a powerful idea This idea, in my opinion, is the embryo that gave birth to constructionism. There is several ideas that resonate with those found in other works by Papert, specially <a href="http://www.amazon.com/Mindstorms-Children-Computers-Powerful-Ideas/dp/0465046746/ref=wl_it_dp?ie=UTF8&coliid=I3S1J23V2VTONF&colid=25BYFGJI3WTE5">Mindstorms</a>. Like for instance <blockquote> Our best ideas are often those that bridge between two different worlds! </blockquote> The book goes on through a lot of ground. Sincerely there is quite a lot to learn, even from casual reading, like this little pearl from Buddha: <blockquote> If the mind were an ego-personality, it could do this and that as it would determine, but the mind often flies from what it knows is right and chases after evil reluctantly. Still, nothing seems to happen exactly as its ego desires. It is simply the mind clouded over by impure desires, and impervious to wisdom, which stubbornly persists in thinking of ''me'' and ''mine''. </blockquote> Well, I'll say it one more time, Minsky is a genius, one of the greatest minds to work on AI and psychology. This is a great work, so go ahead and fetch it! My bet is that this book and its successor will still be studied for decades, and will be looked back as reference for a lot of subjects. <h3> The present and the future </h3> There is a <a href="http://www.amazon.com/exec/obidos/ISBN=0743276639/marvinminskyA/">successor</a> for the book, and I have it in hands, so there will be a review of it in the future. Not that I have digested the SoM, but perhaps a little peek on the development of the ideas will make me understand them better. This book is a point of view from an AI father about intelligence, and it is a must read. But there is a lot of finer tools for analysing the brain these days. My hope on cracking the intelligence and consciousness enigma is to look at the brain with those tools and frameworks like those shown in this work. One last quote: <blockquote> One can acquire certainty only by amputating inquiry. </blockquote> Go ahead, read the book with a lot of inquiry, and enjoy! 2009-05-02T02:40:12Z 45 review-of-marvin-minsky-s-society-of-mind Review of Marvin Minsky's Society of Mind

2009-05-13T21:23:08Z