Stewart sets out to prove that evolution is progressive. He points out technology, where it's easier to show that there has been progress (if one only looks at the last seven centuries or so and mainly in the West). He then points out that there has been no comparable progress in social life. The reason is that change in technology is driven by a mechanism (the market) that selects and reproduces technological improvements, whereas change in society is not driven by a mechanism that selects and reproduces social improvements.

The second step consists in proving that cooperation is a powerful mechanism for evolution. Stewart argues that whenever organisms team together new adaptive opportunities open up. The larger the cooperating group the more explosive the growth in adaptive opportunities. He then prevents an objective by showing that cooperation is counterbalanced by selfishness, and thus it never evolved fully.

Stewart believes the mechanism that evolved cooperation is neither Hamilton's kin selection nor reciprocal altruism: they are both too weak. To understand what drives cooperation, Stewart investigates what could restrain selfishness and comes up with a surprising answer: management. The degree of cooperation that one finds in the most cooperative organizations is due to some kind of hierarchical organization in which some kind of management minimizes the "damage" that individuals can cause to the whole. Stewart believes that organizations focused on cooperation become more competitive than the others. Stewart shows that evolution tends to create managers: they benefit if the cooperation works, and therefore have a vested interest in making it work. The organization becomes highly cooperative even though its members and its management are thoroughly selfish.

External management can do the job, but Stewart believes that there is also internal management at work. Multicellular organisms are an example of cooperation whose management is distributed internally. Distributed internal management basically consists of hard-wired predispositions towards cooperation (and towards limiting the damaging effects of competition) that are reproduced in each of the members of the organization. Stewart notices that such distributed internal management works towards stability, and this could escalate to a form of "dictatorship" that prevents any kind of innovation. The ideal form of distributed internal management is one that restrains competition without stifling innovation.

However, the fact that organizations are possible and make sense does not necessarily imply that they will evolve: that also depends on how good they are at tapping into the potential of cooperation. If they simply proceed by blind trail and error, they might not go too far. If they are capable of self-discovering the potentials of their internal cooperation, they will yield organization at ever larger scales at an accelerating pace. They will, in other words, enjoy and foster greater evolvability. Stewart argues that evolvability has already evolved in the past: he does not believe that the evolution due to natural selection is totally blind. In his opinion genetic system have improved their ability to evolve because the pattern of mutations is inherently biased towards those mutations that are more likely to yield favorable adaptations. He thinks that some genes are more likely to mutate than others because that mutation is more likely to yield an evolutionary advantage. Basically, the genes that determine gene mutations have evolved themselves, and they have evolved to improve evolvability. The pattern of genetic mutations can actually be quite smart, not random at all.

That said, the mechanism of genetic mutations is still severely limited by the fact that it works over generations: it cannot adapt an organism during its lifetime. Stewart believes that the mental life of humans constitute a notably remedy to that limitation: humans can adapt during their lifetime and pass that knowledge on to future generations. Humans are the product of a dumb genetic mechanism that was actually smart enough to evolve beings capable of adapting during their lifetime. Therein lies the potential to further accelerate the evolution of life.

Stewart then proceeds to the describe the ideal self-evolving organism, who has to break free of the psychology created by the old narrow-minded short-term evolution in order to achieve longer-term evolutionary success. The thinking has to be different. Here is where the book turns into a manual of management, and it becomes quite vague. Humans need to learn self-management and so on, but it all amounts to simply learn how to model evolution of complex systems.

As far as the Earth goes, so far evolution has organized molecules into prokaryote cells, and these cells into eukaryote cells, and these cells into multicellular organisms, and them into societies, and Stewart can see a near future in which the whole of the planet will be organized under the management of humans.

Stewart believes that the lessons of cooperation should be applied to governments, creating governments that, for pure self-interest, will end up benefiting all the individuals of society. Democracies satisfy this requirement but are still limited in their appeal to cooperation. Stewart wrote the book during the gay 1990s, so he is convinced that free markets work better than governments: "When governments have competed directly with markets in the organisation of economic activity, governments have proved inadequate". One wonders how Stewart feels today, after the Great Recession caused by free markets at a time when the biggest success story is the economy created by the central planning of China's communist party. He writes that the free market "makes it profitable to develop a more fuel-efficient motor car engine or a tastier ice cream" forgetting that the market developed highly inefficient cars that might have led to more wars and economic crises until governments stepped in and passed laws forcing the market to make fuel-efficient cars, and forgetting that food has been getting less and less healthy, causing millions of deaths. We would not have the Internet, nor the freeways of the USA, if it weren't for government bureaucracies. Some of the premises of Stewart's historical arguments will make professional historians cringe, but that's not the core of his argument, so let it be. Stewart proposes that a "vertical market" is a better system than today's democracies: a vertical market is a market not of goods but of rules and regulations. This is the system that, in his opinion, would perfectly align the self-interest of the managers and the self-interest of the managed.

Throughout the book there is an intimate connection between self-evolution and cooperation, the latter being considered a prerequisite for the former. This vision of cooperation on a planetary scale escalates into a formidable crescendo towards the end.

I have three main comments. One is purely technical, one is ontological, and the third one is more existential in nature.

Firstly, Stewart assumes that conscious mental processes are not a form of trial and error, and he neglects neurological literature that seems to point to the opposite conclusion: that even our most cherished private life may be just the outcome of a very complex set of trial and error processes. The very structure of the brain (neural networks) and the electrochemical processes that take place over it are simply variations on the concept of trial and error. So at a very physical level what goes on in human mental life is just a frantic multitude of trial and error processes. Stewart may have been fooled by his own brain into thinking that his own brain is smart, when in fact his brain is just a very complicated (by human standards) set of trial and error processes.

Secondly, Stewart keeps talking about evolvability but it is not clear what would be the goal of that evolution: a mere increase in the chances of surviving in the face of environmental changes? It is clear that he thinks the next step in evolution is cooperation, but cooperation for what? Grandiose statements like "Successful participation in the future of evolution of life means we must continue to form cooperative organisations of larger and larger scale" are actually cryptic: evolution towards what? What exactly is the "success" that comes with this? And what happens if we don't build those ever larger organizations? He preaches "longer-term evolutionary success for humanity" but does not explain what that success would entail: peace, money, eternal life, wild sex orgies, ...? He reiterates that the ultimate objective should be "evolutionary success": is this merely an insurance against extinction? or does it mean that we will rule over the universe? or what? And would the benefit be for the race as a whole (do i really care what happens to the human race one million years from now?) or for the single individual alive today and her/his family? Stewart hints at the fact that some of us (I'm one of them), having been programmed by old-fashioned evolution for short-term goals, cannot even envision what that "success" consists of, but then why should we care? I personally don't want to cooperate with the opposite team when we play soccer, and i don't even want company when i go hiking on the mountains. It might just the the old evolutionary process at work, but i value my solitude in the rare occasions when i can get it. It is not clear what Stewart's utopian cooperative society is meant for: what is the "success" that it will yield to me? Cooperating for the sake of cooperating was the dogma of communism and many other failed ideologies (most of which ended up killing millions of people in what was probably a very failed evolutionary experiment). I literally don't know what he is talking about, and i just sense that he is talking about something that i might not desire.

Thirdly, this is a book obsessed with contributing to the future evolution of life in the universe. The goal is evolvability for the sake of evolvability. The goal is the evolution of organizations that are best at evolving. The point that Stewart ignores is that "evolution" means "destruction": when a species evolves into another species, the old species has typically disappeared. The individuals have physically died, and so have their descendants, and so has their "world". Or the new species has taken over the world from the old species, and the old species is now what monkees are to humans. Why would the old species encourage and even accelerate the evolution of a new species that will either destroy it or make it obsolete?

The simplest objection to Stewart's vision of self-evolving beings is that nobody truly wants to evolve. Change for the sake of change is not what we want. Most people want to remain human beings (and not something else) living in the society that they know and surrounded by the environment that they know. Most people fear change, that alters their lives, their families, their values, their jobs, their status. The ultimate change is death. In fact, rich developed societies increasingly try to minimize change, ford example by preserving the environment, for example by creating institutions that will last at least one's lifetime. Why would anyone want to accelerate change?

If i mutate into something else, that something else might be superior to me in every possible way but it is not longer me. What exactly is an improvement over me as I am now? over my world as it is now? Is it still me? Is it still my world?

Creating smarter humans for the sake of creating smarter humans is a paradox. There are no doubt in my mind that more intelligent beings can exist and will exist. In fact, we don't need to think of a new species: there are much more intelligent individuals in my own species. Einstein was certainly more intelligent than me. That does not mean that i would want someone to remove my brain and replace it with Einstein's brain in my body: that would not be "me" anymore, it would be him.

TM, ®, Copyright © 2011 Piero Scaruffi