After the machine

The emerging arch-trend of our era is more biological rather than algebraic. The new prevailing metaphor is no longer the machine, with its precisely defined input and output functions, but rather the self-evolving system.

Ludwig Wittgenstein (1889-1951), an Anglo-Austrian philosopher, once remarked that he was writing for people "who had not yet been born, who would breathe a different air and think in a different way".

Wittgenstein's own career as a philosopher embraced a profound change in perspective that presaged a change now taking place in many fields.

The early Wittgenstein took part in a centuries-old effort by Western intellectuals to algebraicize the world by analyzing it in terms of symbolically expressible principles that work like machines with precisely defined pieces interacting with each other in precisely defined ways. The later Wittgenstein pointed to a way that leads beyond abstract machines and symbols.

Descartes mechanized the concept of space, and from there Newton mechanized the concepts of force and motion. Now, after centuries of preoccupation with mechanization, we seek a deeper understanding of exactly what "livingness" means.

In one way or another, all living things are creative. In particular, they participate in creating themselves. But what is creativity? Not all mere novelty qualifies as creative. Creativity involves bringing something rare and valuable into existence, not simply causing arbitrary change. What does that mean from a physical perspective?

The quest for new insight into precisely how living things differ from machines will turn out to be a leitmotiv of much of this century.

Symbols are a convenient way to cope with complexity. The algebraic, symbol-manipulating approach to knowledge has been critical to the development of science, but now the complexity at the heart of many scientific and cultural endeavors is often irreducible.

The quest for new insight into precisely how living things differ from machines will turn out to be a leitmotiv of much of this century.

As Wittgenstein remarked, this is a "different air," a different approach to knowledge and meaning. The emerging arch-trend of our era is more biological rather than algebraic. The new prevailing metaphor is no longer the machine, with its precisely defined input and output functions, but rather the self-evolving system.

The following are some of the characteristics of self-evolving systems. They

  • create and modify their own priorities.
  • respond to changes within themselves.
  • remain maximally open to interaction with their environment while still maintaining their structural integrity.
  • have behavior that is not expressible in terms of a set of explicit procedures.
  • lack a distinctly identifiable internal pilot because control emerges dynamically in the self-evolving system as a whole.
  • can serve as pilots for larger systems, providing motivation and intention.

Of all technological shortcuts based on the symbol-manipulating approach to controlling nature, digitization has been by far the most impressive. Aspects of the world that were once impossible to mechanize have now been reconfigured as tamer, virtual realities.

Despite the success of digital technology, self-evolving systems will probably never be based on software. Livingness cannot be virtualized. Nature's unabridged messiness is the basic stuff of life, not an externality that can be abstracted away.

Self-evolving systems will most likely be implemented through biomolecular engineering, a discipline still relatively in its infancy.

Biomolecular engineering will use complex systems as tools for understanding and harnessing other complex systems, without needing to construct explicit, intervening representations. Technologies supporting self-evolving, emergent systems will entail ineradicable uncertainty, which means that self-evolving systems will not be machines.

Challenges raised by playing one layer of complexity against another will provide opportunity for new forms of quantitative analysis to flourish.

Cultivating surprise

An essential element of evolution is surprise. The evolutionary future always lies beyond the power of any current process to predict or create. In that case, how can a system evolve itself, and what exactly is self-evolution?

The creativity that drives evolution means more than inventing new moves in a game; it means rebuilding the framework within which the game itself is executed. Such creativity usually unfolds orthogonally to any current reference or deliberate action.

A creative system has to continuously bootstrap itself in ways that are categorically unmechanical. Creative depth springs from self-evolution.

For that reason, evolution depends on open protocols that allow otherwise independent structures to interoperate. Open protocols form a stage upon which further development can unfold, making development possible without necessarily causing or directing it.

As an example, although DNA is often thought of as a basis for encoding and transmitting information, perhaps a more fruitful concept is that DNA is a protocol around which complex developmental activities spontaneously organize.

Although machines can be built that learn from their mistakes or that use random input to generate novelty, a creative system has to continuously bootstrap itself in ways that are categorically unmechanical. Creative depth springs from self-evolution.

Discovering a scientific basis for creativity is arguably one of the most important challenges we face. Creativity is crucial for us as individuals, organizations, nations, and as a species that threatens to multiply itself to death.

Michael Webb, 2001

home ]