Nine laws

Try it today. Look up at a flock of birds that flies overhead. In the city, you will find pigeons; on the seashore there are tiny stints; near rice-fields in the evening you'll see egrets; if you venture out to the Rann of Kutch or Orissa's Chilika Lake, you may even find flamingos. Don't know about you, but I find something deeply satisfying about watching flocks of birds. How do they make those intricate, complex and yet elegant moves?

One theory is that each bird follows some simple rules. Computer scientists have accurately simulated the movements of flocks by making individuals follow simple rules. (Of that, another time). But intriguing as such simulations are, we don't really know if it's how real birds work.

Nevertheless, the point of the theory is that out of simplicity, you get complex behaviour.

Out of next to nothing, nature produces something. Often, something extraordinarily beautiful.

Kevin Kelly, co-founder and still "Senior Maverick" (I want that title) of Wired magazine, has put some thought into these ideas. Looking at esoteric biological and computer science research, with a lot else in between, Kelly compiled what he calls the Nine Laws of God. They are not the only laws that can be applied to creation, but he once wrote that "these principles are the broadest, crispest and most representative generalities."

Here are Kelly's laws, with a few lines about what each means to me. They got me thinking on all kinds of lines. Perhaps they will do the same for you. (This is from an early listing; compare to how Kelly now looks at them).

1) Distribute. Many things in life -- flocks of birds, your intelligence, the performance of the economy -- has an essentially distributed character. Each of them is made up of a number of smaller components that connect in some way. The whole is frequently more than the sum of the parts -- case in point, the moves that flocks of birds can produce. The phenomena around us that we find most interesting, stimulating and true to life are invariably rooted in distribution.

2) Control from the bottom up. When many things come together to form one system, when they are all connected, things begin to happen independent of any central authority. (Think of how mob behaviour can quickly spiral out of control of any one leader). This means that overall control of a large distributed system must grow out of smaller, simpler, local authority: village panchayats, for example. Control must rest at the lowest possible level. At the bottom.

3) Positive feedback. If something works for you -- an idea, a tool -- use it again. And again. As you do this, you refine and strengthen your skills. Nothing succeeds quite like success, so never have qualms about repeating it.

4) Grow in little bits. No complex system can work without having evolved from simpler systems that work. For example, a country cannot change overnight from a centrally planned economy to a market-driven economy -- such an attempt is bound to fail, or at least stutter, as several countries have found. Therefore, start with small, simple parts. Put them in place. Let them operate on their own; let them communicate with each other; give them time to work out their particular problems. If all that works, complexity will grow out of these simpler blocks. That is, complexity can only be built, bit by simple bit, from simplicity that works.

5) Keep diversity alive and healthy. That's how you change with a changing world. A large, homogenous system can adapt only via large, convulsive efforts. But if a heterogenous system encourages diversity, it can keep changing in small ways, adapting where and when necessary. That increases robustness and spurs innovation.

Uniformity kills. Diversity breathes life.

6) Honour your mistakes. Making mistakes is an integral, even critical, part of life. If a company wants to stay ahead of the competition, it must keep finding new avenues, methods, products. It must break from the conventional. Getting off the beaten path is a crucial idea; it is also indistinguishable from making errors. Creating something, by definition, means making mistakes. So recognize that. Celebrate mistakes: making them means you are making progress.

7) Don't pursue perfection. In any complex system, there are compromises and tradeoffs. In building a car, do you use lightweight materials so that it is zippy, but a deathtrap in an accident? Or do you use such heavy sheetmetal that it is safe even in a nuclear attack, but moves like a sloth? Naturally, neither. Nor can you hope to build a car that is both built like a tank and as nimble as a Maruti 800. Instead, you build a little bit of both -- safety and speed -- into your car.

You can't have perfection, so don't waste time chasing it. Get it working, that's all.

8) Walk the line between change and constancy. If you've reached equilibrium in a project, treat that as a sign of doom. If your creation is to live and grow, you must have frequent change. At the same time, relentless change alone also leads to disintegration. So there must also be some amount of stability. A successful creation is in what Kelly calls "persistent disequilibrium" -- forever balancing on the crest between change and constancy.

9) Change changes itself. Large, complicated systems all evolve: they change according to certain rules. Evolution is all about how things -- animals, plants, institutions -- change. But because such systems interact, because they affect each other, the rules for change themselves change. Think of a carnivore animal evolving powerful jaws to catch and kill its prey. Responding to that, the prey evolve horns, or sharp hooves, or simply speed, to protect themselves. In turn, the carnivores must evolve other killing techniques: their own speed, for example.

In a deep sense, evolution is about how the rules for change change. To turn nothing into something, you need not just rules, but self-changing rules.

There you are: Kevin Kelly's Nine Laws of God.

"I believe," Kelly wrote, "that one can go pretty far as a God while sticking to these nine rules."

What do you think?