Scientists pursue system to explain facts that slip through the cracks

By H.T. Goranson  / 

Tue, Sep 06, 2005 - Page 9

An unusual meeting of scientists took place in Paris this summer, when scientists gathered to brainstorm about the need for a new science, one that could be as revolutionary as Einstein's insights were a century ago.

Most scientists assume that the basics of science are known. In terms of big challenges, the conventional wisdom is that few things are left to discover. The remaining options are said to fall into three groups: "grand scientific quandaries" (such as uniting gravity and electricity into one theory) which require a huge investment and first-world infrastructure; "data collection," which is the field work associated with archeological digs and biological/genetic surveys; and "science-informed problems," such as combating AIDS or addressing global warming.

Beyond that, many believe the only hard work will be to use existing laws to benefit humankind in new technological ways. Who can argue? After all, today's models work.

But an emerging group of scientists points to phenomena that current theories do not address well. These problems are exceedingly common and artfully avoided because the science that would account for them just doesn't exist.

This missing science would describe processes and how entire systems evolve.

Individual scientific disciplines are understood fairly well. Physics, at least the physics we encounter as ordinary humans, is well mapped. Chemistry and biochemistry are similarly solid -- there are some things we don't understand about the body, but it is believed that the basic machinery of how cells and molecules interact is known. Slightly apart from these are the new social sciences, which deal with humans and societies.

In each of these areas -- physics, biochemistry and social science -- the theories are mature and largely uncontroversial. Each discipline has its own language and its own separate machinery. Rarely is a scientist an expert in more than one area, because the worlds and languages are so different.

This means that we can't answer complex questions that depend on more than one field. Consider the brain, for example. This complex organ is composed of molecules that interact using the principles of physics. That information moves according to the laws of electricity. There is also a system of specialized cells and these interact, exchanging chemicals that also convey information. The interaction between two brains adds another level: Here, information is exchanged by means of languages, signs and ideas.

Information is at work on each level and comfortable theories explain how the separate ones operate. But information is being exchanged between the levels as well. There is no science that explains this, even at the most rudimentary level.

To cope with this deficiency, some scientists have tried to reinvent the tools of one level in order to apply them to another. This leads to such things as a "language" at the level of cells and "energy" behind organized societies. Sometimes the transplant works well enough, but it does not address the central problem: What is the nature of the information conveyed between each level, and how is it conveyed?

As it happens, nearly every system in the world is composed of such layers. A similarly embarrassing lack of understanding about how the whole system works exists in every case.

The group of interdisciplinary scientists that met in Paris is loosely affiliated under the banner "Foundations of Information Science [FIS]." They have been working in a distributed collaboration for eight years and come from a range of countries and specialties. As well as scientists from the relevant disciplines, art theorists, psychiatrists, language experts and philosophers are beginning to participate in the discussion.

Most of the group is banking on the "transplant" strategy, in which the basic science of one layer is placed in another, tweaked and supplemented in order to be functional. Others think that a whole new approach is required. They assume that many scientific conventions are accidents of history and new abstractions can be devised. If the FIS group is lucky, there will also be some radical input from thinkers that do not presently have access to first-world infrastructure.

Independent thinking is an underestimated factor. Nearly all the activities collected under the banner of "science" have developed institutional tendencies that are similar to their economic counterparts. It is usually assumed that developing economies need to build resources that emulate those in the developed world, but this could actually stifle the most creative thinking. Many scientific disciples are going through a revolution, and a lot of those ideas -- sometimes revolutionary ideas -- are bubbling up from labs and research centers that are unaffiliated with large institutions.

Twenty years of the most advanced thinking for mathematical algorithms came from a Soviet empire starved of computing power. The cleverest, most literally earth-shattering notions of physics for two generations came not from the expected powerhouses of the day, but from Budapest. Iran has a tradition of architectural design that has revealed key insights to cognitive scientists. Today, some of the most radical new ideas in second-generation artificial intelligence (so-called "autonomous agents") are incubating in Prague.

The most creative breakthroughs became famous events. When Einstein added new abstractions to the language of physics, the identity of space and time changed. The FIS meeting in July was the unheralded beginning of an attempt to remodel the universe in such a way. The group identified the gaps that need explanation. They will expand their pool of thinkers to include scientists from necessarily innovative regions. Following this, they will identify which problems might be solved if this new science is developed. Or if you wish, discovered.

H.T. Goranson is the lead scientist of Sirius-Beta Corp and was a senior scientist with the US Defense Advanced Research Projects Agency. Copyright: Project Syndicate